Wikivoyage talk:Search Expedition Voyage Tips and guide

Swept in from the pub

Contents

meta:Talk:Interwiki map#Wikitravel needs to be nominated for removal again, but before that can happen the wikitravel: interwiki link (and associated templates) need to be replaced with their WV equivalents and the old templates nominated for deletion. This process has already been completed on en: fr: simple: but there is a long list of small Wikipedias in other languages where links to WT still need to be removed or replaced.

What would be the best way to get the word out to editors which actually speak these languages? K7L (talk) 12:40, 14 January 2013 (UTC)Reply

Templates: bjn bn id jv jv map-bms mk mr ms ms or or pam pt ru ru sa sa sh si sk sl sr su ta te th tr uk vi yo xmf

Interwiki links: commons de en-source en fa fr ja ko en-mw meta pam pl pt ro ru-source ru tr zh-books xmf

Interwiki links done: af ar ast as az bar bat be be-x-old bg bjn bn bo br bs ca cbk-zam ceb cs cy-books da diq dv el en-books eo-books eo es fa-source fi fy gd gl gu haw hif hi hr ht hu ia id-books id ig ilo io is it-quote it ja-dict jv ka km kn ko-source kw lad lo lt map-bms mr mdf mi mk ml ml-dict mn ms mt nah nds-nl ne new nl nn no or pl-news pt-news pt-dict qu sa sco sd se sh simple si sk sl sm species sq sr su sv ta te test th tl to tpi tr-books tt uk ur vi vi-dict war wa wuu xh yi-dict yo zh zh-yue

One question I have is this: some of the above (especially the English non-Wikipedia ones) are in userspace. Do we want to go around changing people's userpages? On many projects that's highly discouraged. --Rschen7754 02:50, 15 January 2013 (UTC)Reply

Only in the article space. One can limit their search to that. Travel Doc James (talk · contribs · email) 04:48, 15 January 2013 (UTC)Reply

I'll take a look at the above. I spent last week creating accounts on every WMF wiki, and I'm autoconfirmed on a lot of them. --Rschen7754 09:17, 15 January 2013 (UTC)Reply

By the way, I've struck the Test Wikipedia from the second list as that's all non-mainspace links. If I find others like that I'll strike them too - hope you don't mind. --Rschen7754 09:19, 15 January 2013 (UTC)Reply

However, removing the wikitravel: prefix entirely will break these users' pages - do we need to use a hardlink instead? --Rschen7754 09:57, 15 January 2013 (UTC)Reply

An external link to WT is preferable to an interwiki link for a couple of reasons: 1) a proper external link gets rel="nofollow" which tells search engines to ignore it and 2) the interwiki prefix needs to go away before a request can be made on meta: to finally remove it from the interwiki table. I'm not sure if meta: looks at userspace when deciding whether a prefix is in use or can be deleted. K7L (talk) 00:13, 16 January 2013 (UTC)Reply

Swept in from the pub

Hi! I was just wondering if there's any way we could improve Wikivoyage's standing in Google page rankings. At present, if I search for 'travel guide', 'free travel guide', 'wiki travel guide', 'london travel guide' and 'new york travel guide'. The site didn't appear on the first page for any of these queries nor on the second, despite the fact that another Wiki-based travel site does appear in these places. Is there any way we could improve our exposure? --Nicholasjf21 (talk) 16:42, 6 February 2013 (UTC)Reply

I guess the best way to increase a site’s exposure is by implementing SEO methods. Even Google has a number of tools that can help. Adwords will also increase traffic to this site. --Saqib (talk) 18:04, 6 February 2013 (UTC)Reply

Most likely we're incurring a duplicate content penalty as a long list of articles written by users now here are still sitting on another Wiki-based travel site. That issue will go away once inbound links which were pointing at the other site are updated to point here. Links from Wikipedia are still being fixed in many obscure languages, see #Links from Wikipedia in other languages, #Putting a Wikitravel: namespace in the search box actually takes you to WT and #Sister project link templates above. One particularly annoying link to the other site was in the disclaimer at the bottom of imported pages on this wiki, bugzilla:41983. The new MediaWiki version (1.21wmf9) which fixed this bug and one other (bugzilla:43220, the ability to change the {{listing}} format) was deployed today - which should slowly but surely make a big difference to search engines mistaking an abandoned, ad-infested copy of this content on another Wiki-based travel site as somehow being the "main version".

We should be doing more to verify that (usable, guide, star) articles here which link to Wikipedia have a {{wikivoyage}} sibling project box linking back to us from the /* External links */ section. WP is the sixth-largest site on the Internet and not to be underestimated; if our content is relevant and useful to Wikipedians for a particular destination, we should say so. Certainly any templates on en.Wikipedia which used to point to another Wiki-based travel site now point to us (about three thousand were cleaned up using a robot script on WP) but other languages still have many outdated links which need to be changed to "Wikivoyage". Let's get our own house in order across WMF projects before sinking money (from where?) into AdWords-style programmes. K7L (talk) 20:53, 6 February 2013 (UTC)Reply

If we make this site substantially better this will attract readers, press and Google. But yes currently our Google rating is bad. Travel Doc James (talk · contribs · email) 21:26, 6 February 2013 (UTC)Reply

Could we get a bot running to sync interwiki links between WV and WP articles. Our articles already have the WP links, so it would be easy to find the appropriate WP articles to stick links into. I was pretty shocked the other day [1] to realize there was no interwiki link to our Chicago guide! --Peter ^Talk 21:57, 6 February 2013 (UTC)Reply

For that matter, could WP get {{sister project links}} fixed so that it actually links directly to the intended destination page if given a parameter (like every individual template for siblings) instead of launching a pointless Special:Search enquiry? It's not a huge issue as most links are using {{wikivoyage-inline}}, but if wp.fr can create a viable {{autres projets}} which isn't kludgeware, why can't en:? K7L (talk) 22:47, 6 February 2013 (UTC)Reply

I understand it will take time to build up the links direct to us to increase our google rank. However, in the interim, we could try to play and build up the word voyage. If I search Japan Travel Guide, we aren't in the running. However, Japan Voyage Guide, has us as a first hit, and Japan Voyage has us on the first page. We could update Travel guides at Wikivoyage to Voyage guides at Wikivoyage. In our PR, we could then mention Just google voyage. --Inas (talk) 22:56, 6 February 2013 (UTC)Reply

I think the way Wikipedia has handled us in w:Template:Sister project links is unfair. Only 3 sister sites are automatically omitted unless manually added: Wikidata, Wikispecies and us. Wikidata is omitted because an automatic link would never work; the item number must be entered manually (I also believe that Wikipedia is not sure whether it wants to link directly to Wikidata yet). Wikispecies isn't auto-linked for a similar reason: because they're articles have species under their scientific names while Wikipedia articles take vernacular names. Scientific names must be entered manually. But us? There's really no valid reason.

It could be said that they don't want to automatically add links to a travel guide from an article on say, a person. But then there's an inconsistent approach. You don't "define" people in a dictionary. You can't have "quotations" by a building. You can't have up-to-date "news" on an event that took place 100 years ago. There shouldn't be a double standard that is biased against us. I think we should start a discussion on the talk page of that template, arguing for Wikivoyage's automatic inclusion. That would bring thousands more links and also save us having to go through every location article adding voy=location to it. Thoughts? JamesA ^>talk 07:58, 7 February 2013 (UTC)Reply

As a Wikipedia editor/admin, I would support it. Wikidata is in its own category entirely as it's a database and has all sorts of different software because of that. --Rschen7754 08:33, 7 February 2013 (UTC)Reply

Can Google be contacted and the situation explained? Seriously...would it be reasonable to contact Google & let them know that this site is the legitimate heir to WT content. I tried finding an appropriate contact email, but they probably receive so many emails that it would be highly unlikely a real person would read an email and take any action. However, anyone tech-savvy enough might be able to find the right topic on Google's website to get this site higher in the results. It also might be worth posting a question on the Webmaster forums, where an employee might pick up the discussion and get our site bumped up. Links: Contact, Webmaster forums, Webmaster tools & help, Site not doing well in search.

It might also be worth bringing up on a relevant inter-wiki forum, like some place on Meta. AHeneen (talk) 09:40, 7 February 2013 (UTC)Reply

Knowing how large and evil Google is, they're probably already well aware of the situation and how it came about. While it is possible someone higher up in the WMF could contact their Google contacts (which I'm sure they have considering Wikipedia is on nearly every Google results page), I don't think much will come of it. There's no harm in trying, as Google apparently supports the open, free community. Another thing: Google purchased Frommers recently, so may be working on a competitor, in which case, we're in serious trouble. JamesA ^>talk 10:26, 7 February 2013 (UTC)Reply

I strongly suggest that everyone puts more effort into developing and implementing new features related to maps, listings, mobile version, offline version, etc: anything that makes WV different from (and better than) WT. The content has been copied, the links have been redirected — that's all fine, but it only makes us a good copy (fork, sibling, whatever...) of WT. I think it's time to go further and intensively work on new features and ideas instead of discussing how to persuade Google. --Alexander (talk) 14:19, 7 February 2013 (UTC)Reply

• Question. Is anyone working on doing something about "links" to Wikitravel like the one on the bottom on Poipet#Go_next? If there are many like this a bot should be able to do the task. In fact a template would be better. --MGA73 (talk) 16:41, 7 February 2013 (UTC)Reply

These links are using rel="nofollow". I don't know whether it matters or not=) --Alexander (talk) 16:59, 7 February 2013 (UTC)Reply

The rel="nofollow" was only added yesterday; it could take months for the full effect of such a change to be reflected by Google as it must index every page on the site, one by one. K7L (talk) 20:21, 7 February 2013 (UTC)Reply

I mean is it not better to change the link to Wikivoyage instead? --MGA73 (talk) 10:02, 8 February 2013 (UTC)Reply

• Notice - I've started a discussion here on Wikipedia, regarding auto-inclusion. It's successful passing could mean thousands of extra links for us. I'd appreciate some supportive comments. Thanks, JamesA ^>talk 07:35, 8 February 2013 (UTC)Reply

Swept in from the pub

There's a lot of ideas floating around lately about increasing our Google ranking and ensuring we are the most popular travel wiki on the web. I think the best way to go forward would be a new expedition focused on improving these aspects of our site. It cannot be emphasized enough how important it is to move up the Google rankings, as the site can't be successful if we don't have many viewers. Of course, more viewers will naturally come with more content and better quality, but we already do that quite well. The following is some suggestions and ideas I had that the Expedition could focus on:

• Useful addition of the Wikivoyage template to Wikipedia articles
• The continuous project of removing that site which cannot be named links from WMF wikis
• Working on improving cross-linking between other WMF wikis such as Wikinews, Wikibooks, Wikiversity, Commons, etc
• Verifying and monitoring our Alexa profile
• Promoting the inclusion of a "As featured on Wikivoyage" badge on tourist bureau websites about high-standard articles
• Working on the widespread inclusion of Wikivoyage links on dmoz
• Other common SEO techniques and alternative methods of garnering attention

I think in the long-run, a focused project with set goals will be very beneficial. Thoughts? JamesA ^>talk 08:43, 10 February 2013 (UTC)Reply

That sounds like a great idea to me. Attracting people through Google is absolutely key to our success. Is there already an "As featured on Wikivoyage" badge, or does one need to be made? --Nicholasjf21 (talk) 12:01, 10 February 2013 (UTC)Reply

Perhaps we should be contacting the owners of non-WMF sites which still link to the old wiki, explain the situation (perhaps with a link to the NYT coverage) and ask that they update their links? DMOZ is limited as they only link any given site once, the cleanup of WT templates from obscure foreign languages in Wikipedia is ongoing, not sure how much Alexa can do for us as they're just a traffic counter. The main concern should be to update inbound links to point to WV. K7L (talk) 19:55, 10 February 2013 (UTC)Reply

Let's start it, at least it will be a gathering point for this discussion and a few others going on. I don't suppose we can think of a better name than SEO Expedition? --Inas (talk) 23:03, 10 February 2013 (UTC)Reply

Oh yes, I forgot to mention that: it'll definitely need a much better name. I couldn't think of anything good at the time! JamesA ^>talk 06:02, 11 February 2013 (UTC)Reply

Search expedition? K7L (talk) 18:50, 11 February 2013 (UTC)Reply

• Notice - I've created the Expedition as a trial at Wikivoyage:Search Expedition. The name can always be changed later :) Please join and help me out, everyone! JamesA ^>talk 00:37, 14 February 2013 (UTC)Reply

Article that have completely copied text and no WP links, seem to not rate on Google at all. Like, way off the 15th page or something stupid. I'm currently experimenting with Picton. I've created the new article Picton (New South Wales) that doesn't exist on WT or WV until I created it. The links to Picton (New Zealand) weren't on WP. I'm slowly adding these links, rearranging text, and monitoring the Google Rank, to see how this affects it. As a basis for this, I'm noting that our recently updated policy articles and the pub seem to rank okay - as the content is unique. --Inas (talk) 03:01, 14 February 2013 (UTC)Reply

I like the experiment you're undertaking to check Google's rank decisions. It is true that articles that are near identical to WT will rank very, very far down the list. It may be worth making a list of our largest articles, and having users go through and modify/rewrite a lot of the content to make us unique. JamesA ^>talk 06:44, 14 February 2013 (UTC)Reply

I'm trying to do something similar with North Vancouver. Right now, if I type "North Vancouver travel guide", it doesn't show up in the Google search at all. If I make the search "North Vancouver travel guide voyage", the WV page shows up #2 or #3. I just added a link from WP tonight, plus I'll keep updating the sections and see if that helps at all. -Shaundd (talk) 06:48, 14 February 2013 (UTC)Reply

It's so annoying, especially when added the word "voyage" still doesn't bring it to the top. It'd be interesting to see what happens now. Please add your name here if you're interested in helping! :) JamesA ^>talk 07:00, 14 February 2013 (UTC)Reply

Doesn't anyone know if anybody on other WMF sites is given access to the Google Webmaster Tools? It could be useful to see how effective expedition is? --Inas (talk) 03:21, 14 February 2013 (UTC)Reply

Yes, that would certainly be a good idea. Is there a way to set up multiple accounts to monitor the one website, so any interested administrator could take part? I believe it is possible for us to set up the Webmaster Tools ourselves, along with verification on WOT and Alexa. However, you need admin rights, and I don't have them (just yet, I was meant to get them today actually :P ). The process would be to "upload" a HTML file to the website, which equates to creating a page here which is restricted to admins due to the .html tag at the end. After I'm given the rights, I'd be more than happy to do all 3. JamesA ^>talk 06:51, 14 February 2013 (UTC)Reply

I think the "verification" requires actual access to the servers, not just the 'admin' status of being able to (un)block users and (un) delete pages on the wiki. K7L (talk) 13:19, 14 February 2013 (UTC)Reply

Apparently verification could be done through adding temporary keywords, but it seems admins are actually not able to edit keywords through those MediaWiki links on the main expedition page. I will attempt to contact WMF tech support and request the code additions so that I (and other admins) will have access to WOT, Alexa and Google Webmaster. JamesA ^>talk 03:44, 18 February 2013 (UTC)Reply

Did you do this yet ? --Inas (talk) 22:27, 4 March 2013 (UTC)Reply

No, sorry. I've been reaaallly busy. I got verification through WOT, however, by personally applying. The other two probs wouldn't do that. Please feel free to do it yourself if you have time. JamesA ^>talk 12:10, 11 March 2013 (UTC)Reply

• Notice - I've been verified as a site administrator on WOT, so have responded to the IB comments. I have also created a Quantacast account for us, but have not 'Quanitified' us as that requires further community discussion and WMF assistance. Alexa will also need some WMF assistance that I am seeking. JamesA ^>talk 04:53, 27 February 2013 (UTC)Reply

Post the link please, James. -- Alice^✉ 06:39, 27 February 2013 (UTC)

What would you like linked? Our Quantacast profile is here, as mentioned on the project page, although there is no visible change there. Our WOT profile is here, but there is no visible change either apart from my response. I also forgot to mention that I created a new Wikivoyage email account for cases where we need to register somewhere. So far, I've only used it on Quantacast. I will gladly hand over the log-in details to any administrator who wishes to use/access the account. JamesA ^>talk 06:46, 27 February 2013 (UTC)Reply

Also, current a google search for some of our policies "wikivoyage be fair", "wikivoyage copyright", returns the interwiki link from WP above the native WV link. Curious. Google certainly likes WP and WT. So many extlinks can't be wrong. --Inas (talk) 05:26, 14 February 2013 (UTC)Reply

One of the criteria for higher ranking on search engines is how often site page is referenced and talked about from elsewhere. Not a small task but we need to start answering questions on blogger, discussion groups, Facebook, Twitter, etc. with links back to Wikivoyage pages. --Traveler100 (talk) 07:15, 14 February 2013 (UTC)Reply

That is another excellent idea. All those sites will be useful, but I see our key resource as the Thorn Tree Forum [2] over on Lonely Planet. Already we've had 6 people reference us! [3] Just answering requests with some basic info, then a link back to the source here will be a great improvement. JamesA ^>talk 07:22, 14 February 2013 (UTC)Reply

As per this, it's clear that the IBadmins are monitoring our every move. They managed to make an account on WOT and post a negative review on our profile there within hours of me posting the link. I guess they have nothing better to do with their time, now that no one edits WT. Anyway, I think it brings up some concerns with them copying our ideas. I'd suggest possibly moving this Expedition to a private forum/mailing list, where members must opt in, so that we can discuss the many more ideas to come without them being immediately plagiarised. Thoughts? JamesA ^>talk 07:26, 14 February 2013 (UTC)Reply

As usual, you're not wrong, James. Click on the little red envelope in my signature to e-mail me. -- Alice^✉ 08:24, 14 February 2013 (UTC)

I'm not sure it matters that much. I'm for totally ignoring our friends over at IB. Lets just proceed to make our site the most popular we can, by creating the best and most innovative content, and letting people know about it. If IB choose to create controversy, it isn't a bad thing. Let's hold the high ground. --Inas (talk) 09:55, 14 February 2013 (UTC)Reply

I agree, but it's not just stirring up trouble. IBadmins could easily copy any keywords we decide to add, so that they are always above us in the rankings, and various other things that improve the search ranking of their site, and decreases ours. That's not just about who plays dirty games, but who has the more successful site. JamesA ^>talk 10:02, 14 February 2013 (UTC)Reply

It is all about the community of contributors. A handfull IBadmins can not compete with the volume of good information and updates that a strong community can contribute. --NJR ZA (talk) 14:00, 14 February 2013 (UTC)Reply

True if you're considering our main goal of creating a good travel guide; however, here we are rather narrowly concentrating on publicity and SEO and we should not underestimate the damage that can be done by a concerted campaign of mis-information - you only need to look at the soaring cases of whooping cough and diptheria because of vaccination objectors to see that, sadly. -- Alice^✉ 01:09, 15 February 2013 (UTC)

Our priority is new, accurate and up-to-date information. It's always been our priority. But IB is not a group to be underestimated, with their numerous financial resources. I've seen numerous situations like ours, especially over at Wikia where I used to edit. A few years ago, Wikia's corporate board forced a new skin on all wikis, which meant about three times as much ads, a very confusing layout and constant popups. Vector, Monobook and other skins were outlawed. For many communities, it was the final straw and they jumped ship. My particular community decided against that. Notable communities that forked from Wikia and consequently failed due to SEO and traffic issues include WoWWiki, WikiSimpsons, Halopedia, SmashWiki, The Vault and others. (see w:Wikia#Domain_and_skin_assimilation) As far as I know, despite nearly all of the community forking and more content, most of these failed and Wikia wikis still lead the Google rankings. JamesA ^>talk 04:01, 15 February 2013 (UTC)Reply

One key difference between Wikivoyage and this collection of wikis; we have Wikimedia behind us. Google weights Wikipedia results very strongly. That leaves us a huge task of cleaning up this mess, of trying to wake anyone with ability to update anything at DMOZ, of contacting the owners of non-WMF sites which had linked to WT, of ensuring that any usable/guide articles here which link to wikipedia: have the corresponding sibling links back and that those links point directly to the article instead of to a Special:Search page. The sooner this can be done, the sooner search engines will start penalising WT for being duplicate content and an outdated copy of WV. We also need to remain visible on WP to attract new editors from that project. The task is about as far from glamorous as sorting through the two thousand pages with images still missing, but completing it will make it easier for new editors to find us and join us in creating a good travel guide. K7L (talk) 17:11, 15 February 2013 (UTC)Reply

I think it is evident from the Alexa data that links from Wikipedia have a drastic effect on the number of visitors to Wikivoyage (the peak around mid-January coincides with banner announcing the launch of Wikivoyage that was displayed for a week at the top of every Wikipedia article.) Unfortunately the current links from Wikipedia to Wikivoyage are buried at the bottom of often quite long articles on various countries and cities. Perhaps we should place those links at a slightly more prominent place? I've made a mock-up at Wikipedia:User:Ruud Koot/Sandbox/Rome. —Ruud 22:10, 15 February 2013 (UTC)Reply

That would be grrrrrrrrrrrrrrrrrrrrrreat for both the human and the arachnid traffic.

Perhaps we should have a separate section on the project page where Wikipedians can indicate their Wikipedia user name and then co-ordinate their discussions at Wikipedia - or is that still frowned upon? -- Alice^✉ 22:19, 15 February 2013 (UTC)

The mock-up looks interesting. Another spot where it could go is at the bottom of the first, large template on the right (what we would call on WV the quickbar). Wikipedia doesn't like "tourist attractions" on their pages. However, some pages have a "tourism" section. On pages which do have a "tourism" section, maybe we can put the Wikivoyage template in that section. All these proposed changes, however, will be tricky to implement...it will be very hard to get the blessing of the Wikipedia community to change their policy and allow the Wikivoyage link in a location other than in the last section. AHeneen (talk) 23:18, 16 February 2013 (UTC)Reply

Also, let's see if this has any effect on search results. According to Alexa "trekking in nepal" is one of the highest impact search queries for WT, but it will never make for an encyclopedic article. —Ruud 00:29, 16 February 2013 (UTC)Reply

[4] Ha. —Ruud 01:30, 16 February 2013 (UTC)Reply

A very good example why some topics are best not discussed so transparently...

You see who did the templating (their only contribution so far to the English Wikipedia)? -- Alice^✉ 04:39, 16 February 2013 (UTC)

Thanks, if I have seen that earlier I would have blocked him on-sight instead of opening an SPI. Perhaps this will reveal some additional sleeper accounts, though. —Ruud 09:57, 16 February 2013 (UTC)Reply

Time to fetch a laundry basket? Socks are coming out in the wash, this and this are the same user. K7L (talk) 17:37, 16 February 2013 (UTC)Reply

A heads up that all the socks I pulled in that SPI case have been locked by stewards. Well done guys, stay vigilant. WilliamH (talk) 14:18, 17 February 2013 (UTC)Reply

I've gone through Travel topics and added templates to dozens of WP pages. This is something that could be done by a bot, but is much better handled by a person for a couple reasons: 1) it's placed in the right spot (more visible), 2) placed on the right page (many travel topics are missing a WP link), & 3) links to the same WV page are placed in multiple, related WP pages. I've gone through the travel topics through "health". However, feel free to go back through the WP linked to and look for related pages where the WV template can be added. Once travel topics are done, phrasebooks (on respective WP language pages), and itineraries (when appropriate) need to be added. I'm curious as to what WP policy allowed the w:Trekking in Nepal page to be created as a redirect to WV. AHeneen (talk) 17:10, 16 February 2013 (UTC)Reply

This would appear to fall into the same category as w:Attention whore and w:Cocksucker. Both are protected soft redirects to Wiktionary (in an identical format) for topics which don't have an article with that name in Wikipedia. K7L (talk) 17:52, 16 February 2013 (UTC)Reply

I think there is specific WP policy guidance on this here. -- Alice^✉ 01:48, 18 February 2013 (UTC)

It's actually the next section down on that page: Soft redirects from Wikipedia to a sister project. There appear to be three criteria for such pages to exist: 1) the page should cover a topic that does not deserve an article on WP, 2) there is not an appropriate article to redirect to, and 3) the page should be commonly wikified (and likely to be recreated if deleted). To quote: "Please keep in mind that only commonly wikified words should become soft redirects. We don't need a soft redirect for every possible word or phrase to be included in Wikipedia.". This gives us a limited scope to work with.

We should probably reciprocate. For popular tourist attractions, a disambiguation template would say something like: "[Name] is found in [location] and travel information for [name] can be found at [location]. For more about [name], see [Wikipedia page]." For example, Sydney Opera House does not deserve its own page. It's disambiguation page would read: "Sydney Oprah House is found in Sydney and travel information for Sydney Oprah House can be found at Sydney/City Center. For more about Sydney Oprah House, see Sydney Oprah House on Wikipedia." Travel topics would need different wording. Using a template would be tricky, so that the words sound right, but would keep wording consistent. For topics which could be developed into a proper WV travel topic, a line about plunging forward and starting a page on the topic would be necessary. Travel topics like w:2014 FIFA World Cup, w:Skiing in Australia, w:Hillwalking, w:Ski resort, w:Long distance motorcycle riding, & w:Travel magazine would be topics worth the inter-wiki disambiguation between WP page specifically covering the topic and related Wikivoyage page(s), hopefully helping us when people type such topics in search engines (combination of topic and association of Wikivoyage with travel). Having lots of disambiguation pages like this would be really great for search engine rankings, correct? It would also help users who search the site, instead of having to look through search results. The downside is that this is best done manually. AHeneen (talk) 03:11, 18 February 2013 (UTC)Reply

How are we differentiating making a disambiguation page with a Wikipedia link from including that link in each listing in an article, and what would be the reasoning we'd use to explain the difference to other users? Ikan Kekek (talk) 03:24, 18 February 2013 (UTC)Reply

I've started a proposal at Wikipedia's Village pump. As to whether or not this proposal is accepted or not should be up to the Wikipedians, so I would suggest editors who are primarily active on Wikivoyage to either abstrain from "voting" or at least clear state their affiliation. —Ruud 18:42, 17 February 2013 (UTC)Reply

Is there a Wikiproject on Wikipedia to help coordinate activities across the two sites? If not, could one be set up? 86.44.163.139 22:55, 16 February 2013 (UTC)Reply

I think coordination between our two sites would be within the scope of Wikipedia:WikiProject Travel and Tourism. That WikiProject is not terribly active, though, but still a better option than starting a wikiproject solely dedicated to WV-WP coordination. AHeneen (talk) 18:36, 17 February 2013 (UTC)Reply

Would one be necessary? We can just coordinate things here, and if Wikipedians want to comment on our actions, they can do so here: in one place. JamesA ^>talk 00:43, 18 February 2013 (UTC)Reply

When a WP page contains "{{Wikivoyage}}", what do search engines see? Do they see literally just "{{Wikivoyage}}", or something more, like http://en.wikivoyage.org/wiki/Special:Search/Place_name? Just wondering how effective the template really is in improving rankings. Nurg (talk) 10:10, 18 February 2013 (UTC)Reply

They see the exact same rendered HTML that gets send to your browser; not the plain wikitext. —Ruud 10:31, 18 February 2013 (UTC)Reply

And they don't get Special:Search, they get a direct link (thankfully). JamesA ^>talk 11:25, 18 February 2013 (UTC)Reply

Spiders are told (in robots.txt) not to index links to special: pages. We therefore don't want to encourage creation of templates pointing to Special:Search when they should point directly to a destination - launching a search for something whose location is already known is a waste of time in any case, likely slower (for the server) and not encouraged by Wikipedia's external links policy. Not sure what's up with w:pt:template:correlatos, it only shows projects with explicitly specified targets, but is still a mess of dead or pointless "if no target specified, blindly invoke Special:Search/{PAGENAME}" coding instead of linking directly like w:fr:autres projets. It would seem that some other languages copied code from "sister project links", which in its original form was an ugly kludge based on just blindly going to every Wikimedia project with a Special:Search based on the current page name. As w:pt: has both "correlatos" and "correlatos simples" (the latter being the blind cross-wiki search) the correlatos really shouldn't be invoking this unnecessary special:search. Another thing to watch with blindly linking cross-wiki using {PAGENAME}: these links get tracked into other language Wikipedias when articles are translated, and a blind link to {PAGENAME} from a Georgian language Wikipedia to a Texan-language article on Wikivoyage will fail because the Georgian character set is not the same. (Russian is less of an issue as we have a Russian-language Wikivoyage). K7L (talk) 20:22, 20 February 2013 (UTC)Reply

Melbourne at WP has had a link to WV since 1 Dec 2012 and a 2nd link since 7 Feb 2013. But I don't see that article in the 1st 100 results when I Google:

melbourne wikivoyage site:en.wikipedia.org

Any idea why? Nurg (talk) 01:41, 16 March 2013 (UTC)Reply

It irritates me, because I'm trying to improve the Melbourne article to a much better standard and I'd be sad if no one even looks at it. I think it's because Google thinks that our WV Melbourne article is a duplicate of the WT one, so they punish us. Over the next few days or so, I'll try and change up the content and improve the main article. We'll revisit after a few weeks and see if things improve. JamesA ^>talk 01:50, 16 March 2013 (UTC)Reply

For some searches on travel topics (e.g., "smoking travel" and "laundry travel wiki") we are now being listed above WT in Google. —Ruud 12:12, 18 February 2013 (UTC)Reply

After creating Picton (New South Wales) here (it doesn't exist at WT), I've been slowly adding some content, and linking to WP. The article appears first for on a search for "Picton (New South Wales) wikivoyage". I haven't manage to devise any other search where it will appear in the first 10 pages of a good search. There are some very obscure pages, including a couple of WT pages for completely different destinations, that appear higher. --Inas (talk) 22:26, 25 February 2013 (UTC)Reply

What do you get if you search for "wellsford travel guide" (without the quote marks). WP does not link to the article as yet, btw. Nurg (talk) 10:24, 26 February 2013 (UTC)Reply

When I search for "wellsford travel guide" it's the sixth result for me on Google - looking good! --Nick (talk) 10:28, 26 February 2013 (UTC)Reply

Wikivoyage seems to appear high up the list when searching for new articles that weren't created on Wikitravel. This is true of some Bangladesh articles I've worked on, where "____ travel guide" usually gets first result! That's even without adding Wikipedia links. Google seems to punish us harshly for having identical content to Wikitravel, so it may be worth setting up a project (another expedition?) whose job it will be to rewrite/modify all our guide-standard articles so that we will not be counted as a duplicate. We wouldn't have to add new info or completely start from scratch, but just shift info around and express ideas differently. I think that will be important in at least getting us on the Google search results, then the Wikipedia links will eventually push us above Wikitravel. JamesA ^>talk 11:38, 26 February 2013 (UTC)Reply

I would argue that we don't want to go down the path of having people change articles solely "to be different" - that will happen over time, and our goal should be first and foremost to create a great travel site, not to have the best Google rankings (the latter will follow the former). Creating more incoming links is a good idea, and some of the other suggestions in this expedition have been great, but part of rising in rankings will also be a matter of waiting - Wikivoyage has only been officially launched for 42 days, and Google penalizes new sites [5], so as we prove we're not going anywhere that will also help significantly in raising the rankings. -- Ryan • (talk) • 15:52, 26 February 2013 (UTC)Reply

For some reason, search engines seem to think that Wikivoyage:Don't tout is a good page. We have the top position for various bits of text from that page, including the included counter-example "Well-appointed rooms invite you to a wonderland". Go figure. We must be linking that page often lately? :) (The other possibility is that MSM coverage of our launch is linking there when raising the "this new travel wikipedia is going to attract all the spammers and pitchmen" question.)

As for the "new site" bit? Not sure on that, Wikivoyage has existed since 2006. K7L (talk) 19:17, 26 February 2013 (UTC)Reply

Perhaps someone who speaks German could run a few tests on German search terms to test the "new site" theory with some URLs that haven't changed since 2008. In my day-to-day work there is a lot of effort expended making sure URLs don't change or are updated with redirects, etc, to ensure both that bookmarks still work and that search engines don't think they've found a new page when an old page is moved to a new URL, thus sacrificing the SEO weighting of the old URL. -- Ryan • (talk) • 20:38, 26 February 2013 (UTC)Reply

Picton (New South Wales) is obviously not considered a new page. I can still only get it to appear in results at all by using the word wikivoyage in the search. However, Tarcoola, which I created yesterday, with very little content already appears on the front page google results.

As far as the "haven't been here long enough" is concerned, I think freshness is valued by Google. I know from a couple of blog articles I write, that they very quickly go to the top of the results, and fade away over time.

As for the "make deliberate changes to rankings" argument. No, I agree the changes will come with time. But I don't think we should shy away from making otherwise beneficial changes (improve section headings, improve listing format, etc) that will give us some divergence in the text. --Inas (talk) 22:13, 26 February 2013 (UTC)Reply

Well, Lake Alfred, which was created 18 Nov 2012 and thus not at WT, doesn't appear within the first 5 pages of results for "Lake Alfred" travel guide. Polk City, created 19 Nov 2012 and not at WT, appears at the bottom of the second page of Google search results. In both cases, the Wikipedia article appears at the top of the second page and the text excerpt displayed has "Travel guide from Wikivoyage". Both are small towns (<5000 people), so it's not like there's a lot of travel info on the web!AHeneen (talk) 04:25, 27 February 2013 (UTC)Reply

For me, the Lake Alfred result comes in at number 9 on the first page. --Inas (talk) 04:48, 27 February 2013 (UTC)Reply

I just double-checked and the Wikivoyage page is not found in the first 10 pages of results for either Lake Alfred travel guide or "Lake Alfred" travel guide. Now, the Wikipedia page is the last result on the first page when using quotation marks and the second result on the second page without using quotation marks. AHeneen (talk) 05:29, 27 February 2013 (UTC)Reply

There's some major search localisation going on. The google.com.au results differ from the google.com results. Using Picton (New South Wales) as an example, 'picton new south wales travel guide' is second (out of 400k) in the .com domain, but nowhere in sight for .com.au (1.1m searches). I tried the same for Marina Bay, an already existing Singapore city district with no wp link, and it jumps up and down on 'marina bay travel guide' depending on domain (second page for .com.sg and .com.au, first page for .com). So they might not be seen at all by people already in the country because of Google's redirected local search. - Torty3 (talk) 06:12, 27 February 2013 (UTC)Reply

Google does both localization and personalization. If Google notices (through your account or even your underlying IP address) that you visit Wikivoyage a lot, it will likely list it higher in subsequent queries. You should be aware of this when comparing rankings. —Ruud 13:14, 27 February 2013 (UTC)Reply

I forbid Google from keeping my search history. Therefore I would have thought this wouldn't have an effect. I'm seeing Picton (New South Wales) number 11 result now, and Tarcoola as number 10. It looks to me like if you have a new article here, link from Wikipedia, and don't have other specialist guides already in existence, the article will make the front page of google results. --Inas (talk) 23:44, 2 March 2013 (UTC)Reply

It seems that both Mapcarta and Tripline have started pointing their links to us instead of WT. Both have a decent Alexa ranking a large number of backlinks. —Ruud 01:02, 7 March 2013 (UTC)Reply

Mmmm, intriguing new site, but can you provide an URL where I can see this? I only see links to Wikitravel here: http://tripline.net/waypoint/New_Zealand-6170477577521002A203B68C33D0FBB9

I do hope you're right, Ruud and keep developing that great parsing engine! -- Alice^✉ 01:18, 7 March 2013 (UTC)

Google for "site:tripline.net wikivoyage". And even if you replace WV with WT the pages, or at least the few I checked, obviously not the one you just found, have already been updated to refer to WV.

There still seem to be a few other websites that link to WT on a large scale, some of which should be easy targets for "conversion", although I'm not quite sure if I should name names at this point. —Ruud 01:33, 7 March 2013 (UTC)Reply

Targeting mirror sites or sites that link to WT should be on the "Tasks" list. Maybe create a subpage with a list of known mirrors/link-sites and we will gradually contact them all? It may also be worth writing up a standard email letter. JamesA ^>talk 05:52, 7 March 2013 (UTC)Reply

After pointing out to them that Wikitravel now includes a hotel booking tool, a large hotel booking site just replied to me that they will update their links before the end of next week. More mails are being written. —Ruud 22:38, 7 March 2013 (UTC)Reply

Nice work! Did you already know that the hotel booking site was using WT links, or did you obtain the backlinks from a list? There are some free tools out there that allow you to check a site's backlinks. I may be worth going through the list of WT backlinkers and encouraging them to make a fair decision regarding the current circumstances. JamesA ^>talk 00:49, 8 March 2013 (UTC)Reply

I found the hotel booking site through Google a few days ago. I've since come across quite a useful tool to find backlinks. —Ruud 17:31, 8 March 2013 (UTC)Reply

The hotel booking site appears to have changed their links at the bottom of articles to Wikivoyage, but I think the ones within the articles still point to WT - hopefully they'll get round to changing those too. --Nick (talk) 13:49, 9 March 2013 (UTC)Reply

Picton (New South Wales) that was my first experiment, is now on the first page of google results. So, it seems clear that new content with a single link from WP is sufficient to make the front page of Google results. Insufficient for bing, however. --Inas (talk) 04:57, 11 March 2013 (UTC)Reply

Does anyone know what we would need to do to have Wikivoyage included as a sister project at the bottom of http://www.wikipedia.org/ ? We're already linked from http://en.wikipedia.org/, but a link from the root page would be helpful as well. -- Ryan • (talk) • 20:52, 2 March 2013 (UTC)Reply

These screens are stored as templates, such as meta:www.wikipedia.org template. Ask on the talk page there. K7L (talk) 23:16, 2 March 2013 (UTC)Reply

Thanks, I've commented at meta:Talk:Www.wikipedia.org template#Re: Multiple updates needed. -- Ryan • (talk) • 23:32, 2 March 2013 (UTC)Reply

This is now done, and http://www.wikipedia.org now includes a link to Wikivoyage (meta:Talk:Www.wikipedia.org template#Sandbox). -- Ryan • (talk) • 02:34, 13 March 2013 (UTC)Reply

http://www.wikinews.org/ and http://wikisource.org/ also need to be updated. The process appears to be similar as for Wikipedia (above) if anyone else wants to do it, otherwise I'll follow up. I think all of the other projects include links to Wikivoyage, but if someone could confirm it would be appreciated. -- Ryan • (talk) • 02:56, 13 March 2013 (UTC)Reply

Heads up - our pagerank dropped from 6 to 0 last week. --Peter ^Talk 17:15, 3 March 2013 (UTC)Reply

My understanding is this is just for http://en.wikivoyage.org page. The wikivoyage.org page is better, as is a page like en.wikivoyage.org/wiki/Chicago. Still a bit of a mystery why it would drop so sharply with little change. How are our webmaster tools coming along? Out current Chicago is 3, WT is 5. Both less than 6. --Inas (talk) 03:11, 4 March 2013 (UTC)Reply

The European Union article is 8. --Inas (talk) 00:06, 5 March 2013 (UTC)Reply

That's 'cos it's been hacked about considerably from when it was imported as a "REDIRECT". Page rank is hard to calculate any way now Google don't officially use it and not much of a guide to organic SERPS anyway (do a search on EUROPEAN UNION TRAVEL to see what I mean)... -- Alice^✉ 00:18, 5 March 2013 (UTC)

I have. I agree that it isn't such a big deal. Other pages that are new seem to rate much lower though, so it is curious. My primary concern is why the main page would drop from 6 to 0. That implies to me that Googlebot thinks there is something surreptitious going on. --Inas (talk) 00:30, 5 March 2013 (UTC)Reply

It's called a "duplicate penalty", and in the short term (a year or two) the only way round it is to kick up enough of a stink that Google manually manipulates their results. They always pretend that they never manually override their (continually evolving) search algorithm, but if you kick up enough stink (as in lots of newspaper and magazine articles) they eventually will. Meanwhile, we would be well advised to re-visit some ancient (and patently stupid) policy decisions in a conscious effort to put as much blue water as possible between us and Wikitravel... I do my best, but I'm continually slapped down by some of the "traditionalists" that have no real conception of how important some of this stuff is to our potential readership. -- Alice^✉ 08:01, 5 March 2013 (UTC)

I don't think you're right, in this case. A main page that is totally undup has google rank 0. Other pages that are identical have no such problem. I think there is something else going on here, focussed on the entry page. --Inas (talk) 08:40, 5 March 2013 (UTC)Reply

The "duplicate penalty" is applied to the entire domain - not individual pages and, believe me, I wish I were wrong. Come back in a year or two and we'll have an answer if nothing else is done in the interim. It certainly does help mitigate the penalty if highly internally linked pages (such as the main page) don't have duplicated content. However, most of our article pages right now are linked to Wikitravel via the standard template on the bottom of each which says something like: "This article is derived from the article EU on wikitravel.org/en in its revision as of 22:05, 25 July 2008 (UTC)." even though, in this example, that statement is clearly mendacious since the Wikitravel article was simply a re-direct on that particular date! -- Alice^✉ 10:00, 5 March 2013 (UTC)

I clearly understand that. However, as I just said it doesn't explain the significant variation of page rank between the main page 0 and many other site pages that rate as high as 8. And exact dup pages having significantly higher page rank the main page. That is what I am addrsssing here. I understand that the site overall will have a lower rank. --Inas (talk) 10:31, 5 March 2013 (UTC)Reply

As part of further experimentation, I'm going to delete and recreate Yowah. I'm the only author of that article, so there should be no attribution issues. --Inas (talk) 02:06, 6 March 2013 (UTC)Reply

I wouldn't worry about this too much just yet. It seems that the Main Page, and only the Main Page, has been removed from Google's cache, but it still appears in search results. Furthermore we only seem to be crawled infrequently during Google's monthly or so deep crawls and not on more regular fresh crawls. So don't expect any changes here to have any instantaneous effect on Google's results as it often does on Wikipedia. —Ruud 10:10, 6 March 2013 (UTC)Reply

I resubmitted the Main Page to Google and it's now back in the cache. The page rank still seems to be zero, however. —Ruud 22:35, 7 March 2013 (UTC)Reply

I know Google is a focus, but I've noticed that many pages don't even seem to be listed in bing. Even searching for the place name with wikivoyage doesn't bring them up. WT seems (anecdotally) to rate higher than in google. --Inas (talk) 00:55, 8 March 2013 (UTC)Reply

Yes, we should consider other search engines. I'm also an occasional Bing user and become frustrated when trying to search for Wikivoyage and can't find it at all. Sometimes searching with "wikivoyage" does work, like saying "Qatar wikivoyage" (strangely enough, the Doha article ranks #1 instead of Qatar itself). I'm not sure what we can do about it. I presume the methods we use to improve Google ranking like fixing backlinks, changing interwiki links and removing duplicate content will work on Bing too, but is there certain Bing-specific methods? JamesA ^>talk 01:19, 8 March 2013 (UTC)Reply

Yes. And there are arcane SEO tricks (white hat, not black hat) with Google, too. But I'm certainly not going to spill the beans in a public forum like this one. I'll just say that I wish the people here playing silly buggers (they know who they are) with some tiny, otherwise trivial edits I make are shooting ourselves in both feet! -- Alice^✉ 05:54, 8 March 2013 (UTC)

I've submitted en.wikivoyage.org. The Bing webmaster tools are here. AHeneen (talk) 05:01, 8 March 2013 (UTC)Reply

This probably isn't this expedition's top priority, but... :) One of this expedition's declared tasks is to find a better name for itself that's a bit more all-encompassing. As this expedition is now involved in a lot more things than just search engine optimisation, I think something more like 'Popularity Contest' would probably reflect the nature of this project better. --Nick (talk) 07:49, 8 March 2013 (UTC)Reply

Sounds good to me, Nick. -- Alice^✉ 09:36, 8 March 2013 (UTC)

I sort of like the double meaning of the current name. —Ruud 17:32, 8 March 2013 (UTC)Reply

I agree - it would be nice to have some sort of pun in there. Project Knife & Fork? :) --Nick (talk) 01:20, 9 March 2013 (UTC)Reply

I propose the Who's ya daddy Expedition. Texugo (talk) 02:23, 9 March 2013 (UTC)Reply

Popularity Contest describes its purpose very well, but let's say it is a little too "in-your-face" and that we must be a little more diplomatic in our language. JamesA ^>talk 09:13, 11 March 2013 (UTC)Reply

"Site promotion Expedition". Nurg (talk) 09:34, 11 March 2013 (UTC)Reply

That could work, but is straddling the line of being touty. I'd support it, but would like to hear other opinions. JamesA ^>talk 09:57, 11 March 2013 (UTC)Reply

'Voyage Around the World (Wide Web)'? --Nick (talk) 23:25, 13 March 2013 (UTC)Reply

"Outreach Expedition"? Texugo (talk) 01:20, 16 March 2013 (UTC)Reply

The suggestions have dried up. Let's decide on one. Nurg (talk) 06:28, 28 March 2013 (UTC)Reply

I'm a big fan of Popularity Contest. Its self-deprecating humor is disarming and rather Wikivoyagey ;) --Peter ^Talk 06:46, 28 March 2013 (UTC)Reply

It could be seen as bad practice. Popularity contests have a very negative connotation where I come from. On one hand, we're dismissing ideas like a side-by-side comparison of the two sites as it looks like we're somehow resorting to low blows, yet we name one of our expeditions as "Popularity Contest" which is clearly against Wikitravel. JamesA ^>talk 07:10, 28 March 2013 (UTC)Reply

My point was that the negative connotation is self-deprecating humor about SEO in general, which shows that we're not cutthroat and not taking ourselves too seriously. And I don't see why you think it means we're focused on Wikitravel—it's a popularity contest with every other site on the web! --Peter ^Talk 17:00, 28 March 2013 (UTC)Reply

When I suggested the name I did mean it in the vein that Peter suggests - as a sort of tongue in cheek look at what we're doing here, rather than a serious attack on any other website. Maybe World Domination Expedition would be a better name? :) --Nick (talk) 17:58, 28 March 2013 (UTC)Reply

I won't oppose that name change, but I still feel it is a little too tongue-in-cheek and looks like we're playing the game of stealing audiences rather than improving our guides (which is obviously false). JamesA ^>talk 01:18, 29 March 2013 (UTC)Reply

Oh, but it is what we do: the same thing we do every day, James—try to take over the world! --Peter ^Talk 03:17, 29 March 2013 (UTC)Reply

Well that is Wikipedia's goal, so I guess it should be our goal too! >:D JamesA ^>talk 03:20, 29 March 2013 (UTC)Reply

Bump! - One final name suggestion - 'Profile Booster'! --Nick (talk) 16:21, 10 April 2013 (UTC)Reply

There is a lot of overlap between this expedition and Wikivoyage:Ways to promote Wikivoyage. The introductory paragraphs for both are essentially the same in meaning. Effectively this expedition could be a subsidiary of the other page. Do we need to sharpen the focus of this page before settling on a new name for it? Or even merge the two? Nurg (talk) 03:00, 7 April 2013 (UTC)Reply

That page is definitely a useful tool that we should utilise, but I think that and the Expedition are slightly different. While the page is there to give users tips on how to promote Wikivoyage, the Expedition is about bringing elements of the community together to discuss other ways of promoting our site. The Expedition is for users to work together to come up with ideas and work on specific projects. Already, on this talk page, we've come up with numerous ideas. Though I do feel like we need to choose a major idea, sort of like a 'Collaboration of the Month', and all work on it together to promote the project. Whether that be rewriting key articles and seeing if it affects search results, approaching mirror/link sites and asking them to swap to WV, creating banners for Tourist Bureau sites, etc.

By the way, Promotion Expedition is another name this expedition could take. JamesA ^>talk 09:55, 7 April 2013 (UTC)Reply

That sounds like a workable name to me. --Nick (talk) 00:53, 29 April 2013 (UTC)Reply

Someone should probably keep a close eye on this Twitter account. —Ruud 17:33, 8 March 2013 (UTC)Reply

How to Report Spam on Twitter? K7L (talk) 18:11, 8 March 2013 (UTC)Reply

Well he seems to be doing more harm than good to himself, so far. You sort of have to wonder why he hasn't been fired for gross incompetence yet. —Ruud 23:55, 9 March 2013 (UTC)Reply

http://wikitravel.org/wiki/en/index.php?title=User:IBobi&diff=next&oldid=1946416

That is interesting! --Nick (talk) 21:40, 13 March 2013 (UTC)Reply

http://wikitravel.org/wiki/en/index.php?title=User%3AIBAlex&action=historysubmit&diff=2005168&oldid=1991029

Ooooh - how odd...! --Nick (talk) 23:12, 13 March 2013 (UTC)Reply

Call me a pessimist, but they're probably just trying to pretend they're part of the community instead of being on IB's payroll. —Ruud 14:17, 14 March 2013 (UTC)Reply

That's what I was thinking. Funny though, that it is basically an admission on their part that being associated with IB might produce unwanted reactions/attention. Texugo (talk) 14:20, 14 March 2013 (UTC)Reply

I think you're probably right, but it's still an interesting development nonetheless. --Nick (talk) 14:44, 14 March 2013 (UTC)Reply

Related to the above it looks like almost all reference to IB on the other travel wiki have been removed, whilst the Orbitz ads aren't appearing for me any more either - is something happening over there? (PS I'm not trying to turn this into a spying thread - I'm just curious!) --Nick (talk) 19:07, 10 April 2013 (UTC)Reply

Slightly off topic, but are we any closer to sorting out the WV presences on Twitter and Facebook? I realise there's a separate page for this, but it sort of come under the remit of improving our popularity. --Nick (talk) 00:06, 10 March 2013 (UTC)Reply

As I stated elsewhere, I contacted the page admin of the most popular Wikivoyage page on Facebook, but never received a response. As I am getting a little impatient, I will be trying to make contact with the WMF and Facebook over the next few days so that we can "claim" the page under our ownership, due to WMF trademarks. JamesA ^>talk 09:20, 11 March 2013 (UTC)Reply

I think both the Facebook and Twitter account are still operated by the Germans, to it would be best to contact them. I don't think you want to try and "steal" it from them. —Ruud 11:07, 11 March 2013 (UTC)Reply

The Germans operate the 'Wikivoyage - deutsch' page and another "Wikivoyage" page meant to be interlingual that only has a handful of likes. The largest, main English page is run by an unknown user who we've been unable to determine. See [6]. JamesA ^>talk 12:07, 11 March 2013 (UTC)Reply

The comparative edit stats, article and comments from Peter Daams of Travellerspoint are interesting. Ruud posted it in the pub but it's worth noting here too. Nurg (talk) 23:27, 8 March 2013 (UTC)Reply

I'm confident that 'Soren' at the bottom of those comments is IBobi - the account's profile picture matches the only image uploaded by him here. --Nick (talk) 16:09, 9 March 2013 (UTC)Reply

Yes, but 'Søren' with the same first/last name but with slashes through the Øs is a legitimate user. Doppelganger? Impersonation? K7L (talk) 23:10, 9 March 2013 (UTC)Reply

He really is everywhere isn't he. (Hi Paul!).. Good find with the dup images though, Nick, you should post that on google plus. --Inas (talk) 22:59, 10 March 2013 (UTC)Reply

Almost tempting to print off a pile of the shill posts and send them to WT's advertisers with an explanation that the site looks to be spamming heavily and might be more of an embarrassment than it's worth? K7L (talk) 03:55, 11 March 2013 (UTC)Reply

I'd strongly, strongly suggest drawing a clear line between actions that promote Wikivoyage and those that might be seen as attacking Wikitravel, and that we stay firmly on the "promoting Wikivoyage" side of that line. While the apparent w:Astroturfing against Wikivoyage is hugely dismaying, let's maintain the moral high ground. That said, it might make sense to put together a page that captures some of the suspected astroturfing as a reference for when people ask about those comments (for example: [7] or [8]), but beyond that, I think we're best ignoring it to the greatest extent possible. -- Ryan • (talk) • 04:29, 11 March 2013 (UTC)Reply

Seconded. Several of Paul's comments backfired on him and we really don't ever want to get into that position. I think it would be a good idea to expand the Wikivoyage:Wikivoyage and Wikitravel pages, so we can point people somewhere if they get, understandably, confused about the whole situation. —Ruud 11:48, 11 March 2013 (UTC)Reply

I've just discovered that www.en.wikivoyage.org, does not bring users here, but redirects them to Meta. Could that be hurting our search rankings? --Nick (talk) 01:02, 9 March 2013 (UTC)Reply

I just tested that & Firefox complains that the security certificate is invalid for that name. Pashley (talk) 12:39, 25 July 2014 (UTC)Reply

"The certificate is only valid for the following names: *.wikipedia.org, wikipedia.org, m.wikipedia.org, zero.wikipedia.org, *.m.wikipedia.org, wikibooks.org, m.wikibooks.org, *.wikibooks.org, *.m.wikibooks.org, wikidata.org, m.wikidata.org, *.wikidata.org, *.m.wikidata.org, wikimedia.org, m.wikimedia.org, *.wikimedia.org, *.m.wikimedia.org, wikimediafoundation.org, m.wikimediafoundation.org, *.wikimediafoundation.org, *.m.wikimediafoundation.org, wikinews.org, m.wikinews.org, *.wikinews.org, *.m.wikinews.org, wikiquote.org, m.wikiquote.org, *.wikiquote.org, *.m.wikiquote.org, wikisource.org, m.wikisource.org, *.wikisource.org, *.m.wikisource.org, wikiversity.org, m.wikiversity.org, *.wikiversity.org, *.m.wikiversity.org, wikivoyage.org, m.wikivoyage.org, *.wikivoyage.org, *.m.wikivoyage.org, wiktionary.org, m.wiktionary.org, *.wiktionary.org, *.m.wiktionary.org, mediawiki.org, *.mediawiki.org, m.mediawiki.org, *.m.mediawiki.org, *.zero.wikipedia.org

Hi Ruud,

Nice to hear from you and its cool of you to support Wikivoyage. As you may guess I'm a big fan of Wikivoyage and am glad that you support re-using content. We do try to use the content ethically given the constraints that we earn money from the site. Thanks as well for your kind support.

The nofollow attributes were a mistake. They are a carry-over from the days of Wikitravel. When we switched to Wikivoyage we forgot to remove them. But no worries. I've fixed that and they've all now all gone.

Thank you also for the reference to Mapcarta. Its helpful as we do virtually no marketing of our own.

I believe a great way in which you could help us and also open the door to other mashups and maps on Wikivoyage would be to set up the goal of having accurate coordinates for every Wikivoyage destination. At the moment it seems that only about 60% of Wikivoyage destinations have coordinates. Unfortunately for Mapcarta that means we're providing a poor user experience because we can't associate those places on the map or get accurate photos or hotels without coordinates.

Also please let me know if you have any other suggestions, especially in regards to improving our relationship with Wikivoyage or in ways we can help support the project.

Kind regards,

<name redacted> Mapcarta.com

That's very nice of them. It sounds like a good relationship for us to have. It sounds like now we could do with accelerating our attempts to add co-ordinates to articles. Could we use a bot? --Nick (talk) 00:37, 10 March 2013 (UTC)Reply

I thought project:script nominations had mentioned a 'bot already being run here to link WV articles to WP city-level articles and grab the city-level co-ordinates. K7L (talk) 03:13, 10 March 2013 (UTC)Reply

The word "travel" does not appear in our sidebar, header or footer ("travellers' pub" being the closest). Would it make sense to change the "Destinations" link text to "Travel destinations"? Is there some other way to get the word "travel" to appear at least once on most pages? Without the ability to add a meta description or other keywords it seems we're possibly doing ourselves a disservice by leaving out the most relevant term we expect people to use when searching for our site. -- Ryan • (talk) • 03:38, 13 March 2013 (UTC)Reply

Is there an extension we can install to have keywords? I'm pretty sure they form an integral part of SEO, so we should take advantage of them. If not, 'Travel destinations' sounds good. It might provide a bit more consistency, as all the other links in that top navigation have two words. JamesA ^>talk 04:07, 13 March 2013 (UTC)Reply

I've made the change to MediaWiki:Sidebar, and think it's an improvement in clarity, SEO questions aside. --Peter ^Talk 04:37, 13 March 2013 (UTC)Reply

Thanks Peter. James - re: keywords, having just looked into this it now appears that meta keywords and meta descriptions are no longer relevant for search ranking purposes ([9]) - apparently I need a refresher in the latest SEO tricks. The latest Google guidelines are at [10], and the fact that we now have the word "travel" in the page text (and the page title) helps us fulfill the recommendation to "think about the words users would type to find your pages, and make sure that your site actually includes those words within it." -- Ryan • (talk) • 04:56, 13 March 2013 (UTC)Reply

I've greatly expanded the keywords in MediaWiki:Keywords & MediaWiki:Keywords-0. I honestly don't understand exactly what the difference between these are and how they function, but assume they are invisible keywords that are associated with our site. The terms on each are identical. I've included a variety of synonyms for travel as well as other travel-related terms. The original version did not have spaces "wikivoyage,travel,travel guide,trips,tourism...", is that necessary for this feature to function? AHeneen (talk) 02:24, 15 March 2013 (UTC)Reply

Does anyone know what MediaWiki:Keywords does? I think we need to be careful about possibly creating spammy meta keywords as that can adversely affect SEO if you're seen to be trying to game the system. Currently, however, I don't see a meta keywords tag in the source. -- Ryan • (talk) • 03:05, 15 March 2013 (UTC)Reply

Absolutely nothing. MediaWiki does not emit them anymore and if it would, Google would simply ignore them. —Ruud 03:09, 15 March 2013 (UTC)Reply

Swept in from the pub

Hey guys! I've started editing Wikivoyage last week or so and it's been fun to get back to wikiediting, as well as being able to share my travelling tips and insights in a, hopefully, lasting way. Hats off, kudos, congratulations and thanks to everybody involved in creating, maintaining and developing this project, including each and every editor.

That said, after a few fun moments the thought started nagging me: does anybody even read that?

So I went to Alexa and saw that:

• [11]
• and in particular that: [12]

What do you guys make of it?

It seems like after a short peak after WV's launch, apparently aided by a campaign at Wikipedia, the site's popularity is waning and not even touching WikiTravel. There seem to be trends affecting most travel websites and WV is out. That, and the humorous bit about WV's visitor group:

Based on internet averages, wikivoyage.org is visited more frequently by males who are over 65 years old, have no children, are graduate school educated and browse this site from school.

...signifies to me that there may be little more than enthusiastic wikieditors that actually come over, and the graphs do not give away any trend for this changing.

Is this site going to be a wikimaniacs' little secret and a fun playground, or is there a plan and effort to make it a more popular point of reference. It's still fun editing even for a single person who'd ever read that, but I just wanted to know where the community sees, plans and wants it headed.

Kindest, PrinceGloria (talk) 06:23, 26 March 2013 (UTC)Reply

I think we need to start thinking more mobile and web based and less making sure the pages are printable. This is were the majority of users are. Making the experience interesting, informative and useful for the person using a smart phone, in-car browser and tablets is important. For example go to a location page and dial straight to a restaurant. Do we have a project for mobile connections? Also more links from other sites such as GoogleEarth markers. However the most important thing is probably more content. Finding useful information on a first visit is the only think that brings people back again.Traveler100 (talk) 07:04, 26 March 2013 (UTC)Reply

Thanks for replying. While I might agree with you on some accounts, I believe this is not the major problem. The experience is indeed extremely important, but for now to me it seems that awarness, traffic, and then engagement are more important at that point. There seems to be very little awareness and/or willingness to visit WV at this point. I do not think the lack of a mobile version is the culprit. I also believe there is enough content to justify visitor levels on part with WikiTravel, after all the content is to a large extent the very same. PrinceGloria (talk) 07:13, 26 March 2013 (UTC)Reply

PS. On a totally separate note that warrant a SEPARATE discussion, I would actually more often prefer to print out things than access from my smartphone, due to roaming charges and simple practicality of wanting to have my maps and notes with me at all times. I believe for now, WV is pretty average on both accounts. It is good for browsing, but not optimized for printing either, lacks maps, customization properties et al. I still do believe those are things that should ever be discussed down the road, if the users ever show up to make those for them.

This is a problem that will most likely solve itself over time. By way of avoiding redundant commentary, you might want to have a look at this thread where a similar issue was discussed (relevant section begins with the comment by me, timestamped 11:51, 11 March 2013). -- AndreCarrotflower (talk) 07:20, 26 March 2013 (UTC)Reply

PrinceGloria - see Wikivoyage:Search Expedition and its discussion page. Nurg (talk) 10:35, 26 March 2013 (UTC)Reply

Our Alexa figures have got worse over the last couple of weeks, not so much in terms of the number of visitors (our reach has generally remained a bit above 0.004%)[13] as the number of pages they view once they get here. This has plummeted from 10-15 (in February and early March) down to around 5, almost as bad as WT.[14] The last big drop we had was during the launch; I don't see why there would be a similar drop now. There is a similar drop in the time spent on site.[15]

So while we should certainly try to get more visitors, this recent change seems to be more about what people do once they get here. I don't think Alexa's clickstream data sheds much light on this, although it's not good to read that people go to Google as they leave us more often than than they use Google to reach us. Anyway, maybe we shouldn't worry too much about short-term changes like this, especially when Alexa's figures seem to have other peculiarities. They say more of our visitors come from New Zealand (7.7%) than the United States (6.8%), for instance. --Avenue (talk) 02:45, 27 March 2013 (UTC)Reply

I have a good explanation for the New Zealand thing - I've read that somebody did their best to take care of an article about a town in New Zealand in terms of external links to it, Google positioning et al. It also seems that this article did not have a WT counterpart. This only shows how very few visitors does WV have, so that a little factor like that can affect visitor count. PrinceGloria (talk) 06:01, 27 March 2013 (UTC)Reply

Where did you read that? At Wikivoyage_talk:Search_Expedition#First_results or somewhere else? Nurg (talk) 10:32, 27 March 2013 (UTC)Reply

I might have thought of that one indeed. To us up North it is easy to confuse NSW with NZ ;) --PrinceGloria (talk) 11:30, 27 March 2013 (UTC)Reply

Oh, c'mon. We're ranked 7,700th out of millions upon millions of websites on the web. I don't think it is accurate to say linking to a small NZ town from Wikipedia changed the whole mechanics of our audience and made us most popular with New Zealanders :P While I agree that we need to improve readership, we are not at all "completely ignored" as far as I can tell, and the fairly high Alexa ranking shows that. JamesA ^>talk 13:13, 27 March 2013 (UTC)Reply

A better explanation for the New Zealand thing is that Alexa results aren't really worth the paper you'd print them out on.[16] SimilarWeb gives a more sensible regional breakdown (keeping in mind that the period covered extends back before our launch),[17] although it seems a bit optimistic about Australia. It also seems odd that they think "Private Massage In Tbilisi" is one of the main searches leading people here. --Avenue (talk) 13:58, 27 March 2013 (UTC)Reply

Thank you so much for that, I got a much-needed kick out of "Private Massage in Tbilisi"! SimilarWeb sure has nicer graphics and clearer way of presenting their info, but I guess what really counts is not whether we rank as 7000th, 70000th or whichever site in the whole world, but how many people actually come over and read articles, especially those not editing them. I know such statistics were accessible sometime ago at Wikipedia, it would be great to know that readership is growing, regardless of whether they came her searching for [private massage in Tbilisi] or architecture in Helsinki. PrinceGloria (talk) 16:11, 27 March 2013 (UTC)Reply

Wow, the stats between those two websites seem to vary very much! I would also like to see a WMF-sanctioned statistics like used to exist on Wikipedia, as those are some strange search terms. JamesA ^>talk 08:15, 28 March 2013 (UTC)Reply

I believe our traffic will go up considerably once Google figures out that we're the better source for this data. LtPowers (talk) 14:14, 3 April 2013 (UTC)Reply

Sorry if I'm doubling an already existing discussion somewhere. I was just wondering.. If I understood correctly we're being left out of search results because the content is a copy of content over at WT. That makes sense. Surely that'll change over time as content here evolves, but would rewriting parts of the most-searched-for destinations speed up that process? Do we know what the most searched for destinations are (or rather were, back at WT)? JuliasTravels (talk) 13:15, 27 March 2013 (UTC)Reply

That would probably help, as pages that are either new or have had their content changed significantly rank much higher on Google, sometimes in the top 10. (e.g., try searching for "travellers' pub" or "picton new south wales travel".)

Google Panda severely penalizes duplicated content, although we seem to be getting hit than some other mirrors, perhaps because we're pretty still pretty "young".

However, I don't think we should be making changes purely for SEO reasons. On the other hand, fixing up poor grammar and outdated listings and other content expansion or improvement can never hurt.

A few of the most visited articles can be found at [18], but they are the obvious ones: the articles featured on the main page, the continents and capital cities (London, Paris, Rome, ...)

—Ruud 16:40, 27 March 2013 (UTC

Well, the idea that we shouldn't make changes purely for SEO reasons seems obvious, but I'm not sure it is. Thinking more about it, I don't really see why not. Of course we would never want to decrease quality of wording, but as long at quality is improved or at least kept the same, it actually seems like a good use of time. I mean: in terms of accelerator effects for the development of Wikivoyage, an hour spent increasing search engine visibility might have a far bigger positive impact than an hour spent on fixing stubs for remote villages in rural India. For me, having unique intros and other text for top destinations would be a good thing in any case, for SEO reasons but also to distinguish ourselves in the eyes of visitors. Also, capitals and other well-developed articles are not the ones that suffer a whole lot from bad grammar, missing sections and such problems. Without an active approach, they are likely to be the articles that will be developing least of all, won't they? JuliasTravels (talk) 17:20, 27 March 2013 (UTC)Reply

There will likely be a lot of pushback on any effort to change articles when the primary goal is SEO-related, since that doesn't directly focus on the goal of writing the best possible travel guides. However, more efforts like Wikivoyage:Country surgeon Expedition that have a clearly-designed set of goals for improving high-visibility articles would be welcome, and a side-effect of those efforts will be that our articles begin to distinguish themselves from other travel wikis. -- Ryan • (talk) • 17:29, 27 March 2013 (UTC)Reply

I'm not sure it necessarily is worth changing articles just for SEO purposes as I think the rewards will be pretty limited! My veryu little experience with SEO has shown that you need an awful lot of effort to cause a fairly small change over a long period of time. As Ruud and Ryan have said, we might be better just editing as normal - improving existing pages and creating new ones. Unfortunately, I don't think there is a 'quick fix' here. --Nick (talk) 19:20, 27 March 2013 (UTC)Reply

It's not purely for SEO and it's not about quick fixes. I do think creating unique content is a goal and it would be silly to overlook all the most visited places in that light. I'm just trying to think how (my) energy is best spent. To me personally, there are two ways of editing. On one side there's the places I've been to, that I know, and have knowledge to share about. On the other side however, there's the work you just do for the general good of the site: fixing stubs of places you've never heard of, cleaning up bad grammar etc. And in that category, I don't particularly mind what article I'm editing. I do however prefer larger impact :-) I think creating visibility and thus attracting new editors is key, as we will never write the "best" travel guide with our current active users alone. And clearly, popular articles will have a bigger influence on that than remote places. I'm not saying anyone should do anything, I was just asking about the effects of changing high profile articles. So, thank you for that :-) (ps: The Country surgeon Expedition was a good one indeed, I should look at it again.) JuliasTravels (talk) 21:49, 27 March 2013 (UTC)Reply

It would be hard to do this with some of our star articles, which are already really well written, and honestly a bit hard to improve on. It kills me that the immense work I put into, say, Washington, D.C. and Chicago comes at our expense simply because it was done pre-migration, but I don't want to rewrite either if they are star-quality. The district articles, though, require more updates, which is good. --Peter ^Talk 23:51, 27 March 2013 (UTC)Reply

Oh, agreed, I wouldn't dare :-) Frankly I wouldn't even know hów to improve Chicago more than in small details perhaps ;-) Rewriting any of those top articles would be a pointless exercise. Despite the good collection we have though, it's a limited number. What I'm rather talking about is something like.. say...South Africa, France, Berlin or India. Key destinations, much searched for and okay articles, but with plenty of room for improvement. Rewriting their intros, for example, wouldn't per se be a priority as they are functional, but I think they would greatly benefit from an inviting intro like Chicago has. If that now has positive side effects for the site visibility, it becomes a more interesting task. That's all. JuliasTravels (talk) 12:42, 28 March 2013 (UTC)Reply

I did suggest this in passing above and do think it is a good idea. There does seem to be quite a few opponents, but I think that we just need to look at the bigger picture and the flow-on effects of such changes. An increase in viewership is significant for our site, considering that we are actually hovering quite low at the moment and concerns are being raised about whether anyone uses us at all. I don't think it can be said that the rewards would be limited. Not only will it improve readership, but also improve our general awareness and ability to provide the best quality travel guides. No use writing great guides if there's nobody here to see them... I say we trial it on one of the heavily-searched articles as mentioned above, and see how it goes. Maybe India, because then we can tie it into the Country Surgeon Expedition. Also, I don't think it's fair to say that our Star Articles are perfect and shouldn't be touched. As we recently discovered, some are actually in a fairly dismal state, such as Singapore until a recent renewal. JamesA ^>talk 13:29, 28 March 2013 (UTC)Reply

I suppose I didn't express my view particularly well in the post above. What I'm trying to say is that I don't think we should change articles for change's sake. If we can improve them, but with the side-effect of potentially higher search engine rankings then it's probably worthwhile. My point about limited rewards is not that doing this wouldn't ultimately pay off, it just would take a long time and a lot of editing to push the whole site rather than individual articles up the rankings. --Nick (talk) 18:01, 28 March 2013 (UTC)Reply

Well, maybe I didn't express my original question clear enough either then. We agree on change for change's sake, it's really rather about improving high-visibility areas that can benefit from attention, but perhaps wouldn't have been a priority if not for other side-effects. For the record though, I'm not sure you're last remark is entirely true. I think the whole rankings question is one that takes a long time, but I imagine highly popular articles should have a bigger impact also for the simple reason that they are way more likely to be linked to from other sites, no? And more links in turn would push up the site as a whole? (I'm asking - I'm not expert at all.) I'm not saying huge direct impact, but surely more than edits on small remote destinations, I guess. JuliasTravels (talk) 10:11, 29 March 2013 (UTC)Reply

I think we're both singing from a rather similar hymn sheet! Ultimately, more links would of course push WV up the rankings - it'll just take a lot of work! Improving high profile articles is always a good idea and it probably would have an impact over time - links are indeed very important! :) --Nick (talk) 23:36, 29 March 2013 (UTC)Reply

For some time jamguides.com was used to hold a regularly-updated backup of the content from the old site. While I would like to get that site running again using the XML dumps for Wikivoyage, until that happens I've simply made the site redirect to Wikivoyage (as of 30 minutes ago), so http://jamguides.com/ redirects to http://www.wikivoyage.org, http://jamguides.com/travel/en/Chicago redirects to http://en.wikivoyage.org/wiki/Chicago, etc. From what I've read I don't think this should be seen by Google as an attempt to game the system and thus hurt Wikivoyage, but if anyone has extensive experience with site redirects and knows otherwise please let me know and I'll remove the redirects. -- Ryan • (talk) • 16:55, 21 April 2013 (UTC)Reply

What happens with languages which have not yet been imported to Wikivoyage (Chinese, Hungarian, Romanian, Finnish)? K7L (talk) 17:23, 21 April 2013 (UTC)Reply

All of the redirects are pattern-based so the redirect simply builds a replacement based on rules (grab the language and article name from the old URL and build a new URL) and then performs a redirect: http://jamguides.com/travel/fi/Helsinki redirects to http://fi.wikivoyage.org/wiki/Helsinki which in turn appears to redirect to http://incubator.wikimedia.org/wiki/Wy/fi/Helsinki. -- Ryan • (talk) • 17:31, 21 April 2013 (UTC)Reply

Very possible that articles which did exist before the site moved aren't in the incubator for these few languages. I recall it being a problem when dealing with templated WT links which are still spammed onto Wikipedia articles in these languages - until the content is imported, we have nowhere to which to redirect. K7L (talk) 17:42, 21 April 2013 (UTC)Reply

I can guarantee that there will be some old jamguides.com links that now redirect to non-existent articles - I didn't try to add any intelligence into the redirects, I just wanted to try and have the majority of incoming links and spiders be redirected somewhere useful. The only concern is whether or not this type of redirecting would have any negative effect on Wikivoyage's SEO; I don't think it should, but posted here just in case I'm misreading currently-available SEO guidelines. -- Ryan • (talk) • 18:43, 21 April 2013 (UTC)Reply

We're a couple of months in since this particular expedition set out, but we've not seen a particularly dramatic (if any) improvement in affairs. Is there anything we can be doing and co-ordinating from here to improve our standing? --Nick (talk) 18:19, 26 April 2013 (UTC)Reply

Ditto! PrinceGloria (talk) 22:10, 24 May 2013 (UTC)Reply

Has anyone initiated any attempts to get Wikivoyage listed in Dmoz? Traditionally a dmoz listing has been an important step in improving Google rankings, although I'm not sure how true that is anymore. Looking at dmoz listings for the old site ([19]) it seems there are potentially a number of categories that Wikivoyage could be included in, although at a minimum we need a single listing as a travel wiki resource.

If someone has already investigated could you provide an update? And if not, is anyone willing to volunteer for the task of navigating the dmoz bureaucracy to get us listed? -- Ryan • (talk) • 19:43, 29 April 2013 (UTC)Reply

I think we definitely need to navigate the bureaucracy and come to some kind of memorandum of understanding with the site owners whereby we are allowed to freely add our guides to the "Travel guides" section of places. Right now, it's a matter of individually writing a listing for EVERY destination, submitting it, and hoping someone keeps track of the category. Some categories haven't been updated or monitored for years. I think we deserve it, considering we give dmoz hundreds of links from our own site. If they refuse or aren't interested, I think we can simply remove their links from our site, as they don't really serve the traveller that well anyway.

As a short-term measure, I've submitted out site to the general Travel Guides category, along with Victoria (state) and Melbourne. I'll keep you posted on whether anything happens. JamesA ^>talk 10:29, 2 May 2013 (UTC)Reply

Look at the history for any article we imported from WT during the move — that is, a large majority of articles on the site (though many have been improved here) including nearly all the articles for major tourist destinations. Each history here currently has the entire WT history, with a link to WT for every edit 2004 to late 2012, plus one link per edit to the editor's WT user page if any. For a popular article, that can be hundreds of links. Do search engines look at those? If so, we are giving WT tens of thousands of links to raise their rank. They, as a matter of policy, give us none at all.

Whether or not it affects SEO ranking, I'd like to see just two WT links per history. The oldest starting-point line in the file should be something like this:

This article was imported from the Wikitravel article (link to specific WT version) on (date). The history to that date, including author attributions, is at (another WT link).

As I see it, this is a useful cleanup step, making a clearer distinction between WV and WT edits and getting rid of a great deal of ancient history that the vast majority of readers will not care about anyway. The few who do care are likely to be exactly the ones most capable of following links and tracking down info on WT when required.

There are a number of problems with this proposal. First, it affects attribution and that is a legal obligation from the license, so we cannot even consider it seriously without comment from WMF lawyers. Second, it is a large task, obviously something that would require a bot. Finally, even assuming the lawyers say just linking to WT discharges our legal obligation to attribute, we cannot rely solely on those links. We'd need some sort of backup plan and backup data to ensure we could still attribute even if the WT site went away.

Comment, anyone? Pashley (talk) 21:06, 29 April 2013 (UTC)Reply

For attribution purposes, and also to be able to trace article changes and development, my opinion is that we should 100% have full article history here. With regards to SEO impact, all article history pages are disallowed by http://en.wikivoyage.org/robots.txt per the "Disallow: /w/" rule, so they should have no impact on search rankings. -- Ryan • (talk) • 21:16, 29 April 2013 (UTC)Reply

Swept in from the pub

Quality backlinks is an important factor for search ranking. According to ahrefs.com, WT is losing about 150,000 backlinks per month:

March 1st: 3,974,000
April 1st: 3,775,000
May 1st: 3,667,000

I tried to access the stats for Wikivoyage, but was asked to register so I give up for now, if anyone has time I would be interested in the stats too, hoping we get tons of quality incoming links :-) Nicolas1981 (talk) 09:43, 30 April 2013 (UTC)Reply

Our backlinks went way up in March, from 494k on 1 March to 683k on 1 April. But we were essentially flat for April due to a decreased number of new backlinks and a slightly increased number of lost backlinks. We're at 695k now. It'd be really nice if Google stopped hiding us. LtPowers (talk) 14:58, 30 April 2013 (UTC)Reply

I found the comment posted here, that WT's demise is not going to be as swift as predicted, but in community at least we're excelling, quite interesting. We just need to up our popularity! --Nick ^talk 15:42, 23 May 2013 (UTC)Reply

I'd like to remind everyone that there are only a handful of wikis left that still have used Wikitravel templates and deprecated [[wikitravel: ]] interwiki link code. I suggest users who want to help to start replacing usage of the remaining Wikitravel templates with a Wikivoyage template, linking to en: if there is none in the local language. The last interwiki links should also be removed, if replaced with a standard http: link if appropriate. There is quite a task ahead of us, as although the remaining wikis are few, their links number in the hundreds. See the Whatlinkshere of the Wikitravel template at the Malay, Hungarian, Finnish, Chinese and Romanian Wikipedias. JamesA ^>talk 10:08, 28 May 2013 (UTC)Reply

Swept in from the pub

First, please take a look at this: [20]

Then, allow me to reflect that, when I joined the project quite some time after it's big launch, the number of active contributors stood at 533. I checked today and got the same number.

I went to [Wikivoyage:Search Expedition] and discovered that the policy seems to be to "wait for more and better content and only then promote". Specifically, I was worried seeing that articles that link to namesakes in Wikipedia are often not linked back. Then, I discovered that such automatic interwikilinking is not taking place because of the policy to only link to really good articles. That struck me, as Wikipedia links quite automatically to other language versions, even if the other language version of the article is still a stub. Perhaps there are policy differences that govern that, but I have already learned that in the wiki world, the policies are to be discussed, not treated as gospel.

I believe that with 533 editors and the daily amount of edits we have, it will take an infinite amount of time to bring Wikivoyage to a really high content standard, and a number of editors that low does not even guarantee basic maintenance of the content we already have (the world is changing rapidly, so information in many articles becomes outdated really quickly). If we are to wait with promoting the manure out of Wikivoyage to attract as many editors and gain critical mass until Wikivoyage becomes a place rife with quality content, we are in a vicious circle.

Edits at Wikivoyage talk:Search Expedition are sparse. Either everybody's working their buttocks off there with no time to talk, or perhaps rather simply there is not enough membership, enthusiasm and commitment to actively purse the cause. I guess the guys who volunteered for the project aren't really getting enough support. Perhaps there is something we could collectively do to help them move faster with the things. Perhaps everybody should consider themselves a member of this Expedition and ask to be assigned tasks.

At any rate, I believe Wikivoyage is a great project and I would so much like for it to succeed and become as natural for people to look up and participate in as Wikipedia. But, for now, I am feeling we are applying makeup to what slowly becomes a dead corpse. This Frankenstein's monster needs a big zap to come alive or else it decomposes on our tables.

What is it that is not happening but could? PrinceGloria (talk) 21:25, 27 May 2013 (UTC)Reply

I am not certain what can be done other than trying to improve articles here, linking to them from elsewhere whenever it seems relevant, and suggesting to others that they contribute. Anything beyond that I'm content to leave to the Search Expedition. Presumably they'll let everyone know if they find something good that needs more people.

Anyway, the chart you show does not indicate the absence of a pulse. It shows fluctuations around .004% of web users visiting here with a fall in the last few days under .002. Worrying, perhaps, but not disastrous. It is too soon to tell if the recent drop is just another fluctuation or indicates some larger trend. Also, even .001% of a billion users is 10,000 users a day which is not entirely shabby. Pashley (talk) 21:55, 27 May 2013 (UTC)Reply

(edit conflict) Progress has been slow, but we have hit a bit of a bottleneck with Google -- we don't show up in search results unless you specifically ask for "Wikivoyage". That's an indication that they see us as duplicate content. There are a number of efforts underway to try to fix that. For one, we're trying to get access to Google's (and Alexa's) webmaster tools, but it's tricky because WMF runs the servers. Barring that, we're hoping that as more people find Wikivoyage, Google will eventually realize that WT is an out-of-date archive copy of our content.

There has also been discussion about SEO techniques, but much of that is occurring "in the shadows" so to speak, so as not to tip off WT as to our plans (or to give them ideas).

I do have to question some of your premises, however. I'm not aware of any rule, either here or on Wikipedia, that suggests only the best Wikivoyage guides should be linked from Wikipedia. We would absolutely like to maximize the number of links from Wikipedia to Wikivoyage (without going overboard), but so far as I'm aware, the only thing holding us back is the effort needed to do so.

-- LtPowers (talk) 21:59, 27 May 2013 (UTC)Reply

Hopefully something like this would eventually be handled in Wikidata, but that's a long ways off. --Rschen7754 02:15, 28 May 2013 (UTC)Reply

Wikivoyage:Links_from_Wikipedia#When_to_link does mention that our guides must be usable or above to be linked to from Wikipedia. However, that has been, and should be taken with a grain of salt as the Wikipedia policy as far as I'm aware is to list any sister project if a respective article exists. Re: Search Expedition, things are moving slowly, but have not stalled; we should be prioritising a number of new initiatives and tasks in the coming weeks. JamesA ^>talk 02:35, 28 May 2013 (UTC)Reply

Am I the only one that enjoys WV lack of search engine popularity? I have no doubt that WV will eventually reach a critical mass and be the go-to guide for travelers, and when that time comes it will be awash with special interest groups and well funded and probably misguided attempts to alter pertinent data and we will be dealing with a whole other level of problems. How does the quote go? " First they ignore you, then they laugh at you, then they fight you, then you win." -Mahatma Gandhi. WV is already the best travel guide for some areas, it is just a matter of time before everyone else catches on and I only hope that it does not lose something in the process. Lumpytrout (talk) 04:03, 28 May 2013 (UTC)Reply

Travel wiki is always a small wiki. WT has seen same or even lower level of activity, but it is recognized as a useful source of travel information, even though this information becomes outdated. I see no reason to worry, at least as long as this site has an active user community and sees new developments implemented (page banners are a very good example). The large number of links from Wikipedia may be good for search engines, but it is likely to annoy and upset new users, who will see Wikivoyage as a collection of stubs. This also relates to the policy/practice of keeping hundreds of empty and hopeless articles on less-known destinations. --Alexander (talk) 04:54, 28 May 2013 (UTC)Reply

For comparison, we're what's considered a "medium-sized" wiki since we're not one of the ~600 WMF wikis that rely on m:global sysops and since we have >10 admins. All the other Wikivoyages are small wikis. --Rschen7754 05:10, 28 May 2013 (UTC)Reply

I'll second guess my instinct to not bother with this thread, just because the concern seems so frankly ridiculous. Having made this project a hobby and even a job at one time, checking in on a near daily basis for over 6 years, I can say with confidence that we have never before had such a vibrant community, more top-notch edits, more creativity and innovation, better content control, and just generally such a collective feeling of hope for a bright future.

Search engines are made by design to help users find the best information, so let's worry about creating and presenting it. We have some online travel titans out there to wrestle with (none of which start with the letter W), and I'm looking forward to the competition. --Peter ^Talk 06:31, 28 May 2013 (UTC)Reply

As a side note, it's been pointed out to me that Wikitravel's getting overrun with spambots - meanwhile, we've only been getting 1-2 a day. [21] --Rschen7754 07:53, 28 May 2013 (UTC)Reply

Thanks for chiming in en masse, guys! The amount of people commenting in such a short time is one of the better proofs the project is still alive.

That said, I remain concerned. I fully accept that this is one of the more active MediaWiki projects, but still, I am comparing it to the other vibrant communities I have taken part in. For travel-related stuff, it is e.g. TripAdvisor, whose forums are bustling with content and which has recently added features very similar to Wikivoyage. Here, I could often use advice, assistance or put something up for discussion. Oftentimes, I find I am just talking to myself, as I am the only active contributor to the article! I feel that there is not enough editors, and not enough visitors for some of them to convert to editors.

As concerns issues resulting from the influx of "influencers" due to popularity, I would gladly accept them should they mean we become popular. Wikipedia is attacked every day from so many sides by every pressure group or plain vandal imaginable, and they are doing brilliantly with it. With a ratio of one admin per every ten users, we should be fine, and we can easily draw on the experience from Wikipedia. Even if WV got overrun by spambots and such, I truly believe admins and specialists from Wikipedia would be willing to help out seeing that this becomes an important MediaWiki project.

It is good to hear that there is more going on behind the scenes at the Search Expedition. To draw a constructive conclusion from the above musings - how about setting up a bot at WP that would automatically backlink Wikivoyage articles that link to WP? That would help tremendously, not least in the way that a Wikipedia editor seeing the link might go to WV and, seeing an underdevelop article, start contributing. What do you guys think? PrinceGloria (talk) 08:03, 28 May 2013 (UTC)Reply

If the spam becomes really bad, we can always opt into becoming a global sysop wiki, and then we would get the assistance of stewards and global sysops (about 60), but they're of course spread across all the other WMF wikis. But I think we're doing fine as it is, especially since the stewards are good at globally blocking problematic ranges. As far as setting up a bot, just keep in mind that it will have to pass enwiki's bot approval process, which can be tedious.

A thought: will there ever be a Simple English Wikivoyage, or could the articles on Simple English Wikipedia link to here? --Rschen7754 09:20, 28 May 2013 (UTC)Reply

There won't be a Simple English Wikivoyage any time soon, so it's perfectly okay to link to en:Wikivoyage. Currently, Simple Wikipedia links to en: sister projects apart from Wiktionary which also has a Simple version. JamesA ^>talk 09:27, 28 May 2013 (UTC)Reply

If more moderation is needed, we can also increase the number of admins, before calling upon stewards to help more. I can think of several potential admins, for when we need more. Ikan Kekek (talk) 09:41, 28 May 2013 (UTC)Reply

I am just going to chime in and say something if there is the Simple English Wikivoyage – if there was a wiki with that name, and if I was an admin there (like being one of the bosses to that wiki), I think that wiki should only articles about well-known travel guides (eg. Canada, it could be a beautiful place to visit in the snow for a holiday/vacation) with simple text. Also, like the other simple.wikis, 'pedia and 'ktionary, IPs should help improve and create pages and articles. And so on to the simple.wikis. curtaintoad | chat me! 10:06, 28 May 2013 (UTC)Reply

Well, the main issue with creating another wiki is that we already have 500 or so wikis that are only edited by spambots, so we would need to make sure that there's a demand for another one. The Simple English Wikiquote already closed, as did another Simple English wiki. And I am not sure that it is wise to spend resources on that when Wikivoyage isn't quite established yet. But you can always try starting one out at incubator.wikimedia.org if you want. (P.S. Admins aren't the "bosses" of anything; we don't get extra authority in content discussions). --Rschen7754 10:35, 28 May 2013 (UTC)Reply

Actually, a large percentage of Wikimedians are opposed to the "Simple English" projects. And, no, you can't start a test wiki on the Incubator in a "simple" language currently (cf. Requests for starting a test: Simple French Wikipedia, Simple world languages). It's not defined by ISO as a language. Langcom decided to allow new Simple projects, but is there a benefit to having more? PiRSquared17 (talk) 03:30, 31 May 2013 (UTC)Reply

Back to the bot thing - does anybody have the necessary skills and knowledge to set one up? I love debating, I have no reservations to substantiate the argument at en-wiki that we should run one. Anybody? No? Dust? PrinceGloria (talk) 09:47, 28 May 2013 (UTC)Reply

I think the bot could be counter-productive. Firstly, because it's highly unlikely to pass. But secondly, because it may spark a debate about how lenient Wikipedia is about allowing Wikivoyage links. They may turn around and decide that they shouldn't be linking to Wikivoyage off all 32,830 respective Wikipedia articles. If we were able to have a list of which WV-linked Wikipedia articles do not link back to us, it may help in a manual effort to increase linkage. JamesA ^>talk 10:15, 28 May 2013 (UTC)Reply

Using manual AWB may be an easier alternative. --Rschen7754 10:35, 28 May 2013 (UTC)Reply

For the less literate - what is manual AWB and how can we start using it? Thanks! PrinceGloria (talk) 13:40, 28 May 2013 (UTC)Reply

w:WP:AWB. LtPowers (talk) 14:53, 28 May 2013 (UTC)Reply

Yes. Unfortunately it's a Windows-only program, which is a bit annoying. You also need approval on the English Wikipedia to use it there, but approval is by the trustworthiness of the editor rather than related to the task. --Rschen7754 19:04, 28 May 2013 (UTC)Reply

I am editing from a Mac, bummer. At any rate, whenever you get the approval, do provide a link and a short tutorial - I will gladly launch a PC to help out.

Or does every editor using it require a permission? I used to be moderately active on en.Wikipedia, perhaps I left an impression good enough to pass. How do I get the approval? And how do I configure the AWB to do the task? PrinceGloria (talk) 19:40, 28 May 2013 (UTC)Reply

Every editor requires the permission. I personally use w:Boot Camp whenever I need to use AWB, since I get Windows free from my university. The approval page is linked off the link above. --Rschen7754 01:25, 29 May 2013 (UTC)Reply

I've got AWB and have used it in the past. But just to be clear, what list are we generating? I understand we want a list of Wikipedia articles that are linked to from Wikivoyage but do not yet have a Wikivoyage template. But is obtaining such a list even possible? If so, please explain =) JamesA ^>talk 09:51, 29 May 2013 (UTC)Reply

It will probably take some creativity to generate the lists - maybe going through categories of cities and checking for a Wikivoyage template, for example, and then adding the template and the proper article name. --Rschen7754 05:16, 30 May 2013 (UTC)Reply

Note that Wikidata will soon allow linking between Wikipedia and Wikivoyage, so perhaps this isn't necessary yet. --Rschen7754 10:25, 17 June 2013 (UTC)Reply

Why no Capitalize on the launch of a new logo and have a banner campaign at Wikipedia, similar to that when this site was first launched? I imagine such a banner to state something along the lines of "Wikivoyage has launched a new image" 81.178.172.94 23:48, 23 August 2013 (UTC)Reply

I am sure it will get mention in the signpost. Greater coverage would be good but getting consensus may be an issue. Travel Doc James (talk · contribs · email) 23:51, 23 August 2013 (UTC)Reply

I was a bit shocked to read at PageRank:

"Current PageRank (0 to 10, higher is best):

• For the Paris article: Wikivoyage 3/10, Wikitravel 6/10"

It does not matter what the so-called "PageRank" of our Paris article is (or what any other of our articles is).

What matters is whether our article appears in the first 2 or 3 organic search results on relevant searches (and, for those with a competitive streak, whether it appears above or below a WT entry). Google continuously changes its algorithms and there is now only a very tenuous link between so-called PR and search results.

We can not directly change some of our on-page factors but one thing we can do is make sure that the templated inbound links to us at the foot of a Wikipedia article are as attractive a spur to action as possible to most people who might spot them.

Wikivoyage has travel information related to: Nelson, Lancashire

Please edit the text of the Wikivoyage template at Wikipedia so that instead of, the bureaucratic sounding message of the first box,

Wikivoyage has a travel guide for Nelson, Lancashire.

it has the more attractive and relevant message of the second box.

Fortuitously, this will also mean that the more popular search term of guide will appear next to the anchor text.
(Few people include information or related (or synonyms) in their search terms as http://www.google.com/trends/explore#q=travel%20guide%20to%20paris&cmpt=q indicates!) --W. Frank^email_talk 15:32, 13 September 2013 (UTC)Reply

It's probably worth posting here about this as well. We really do need to increase our popularity in as many ways as we can. --Nick ^talk 16:17, 13 September 2013 (UTC)Reply

I think it is better that you do that, Nick, I think you have a registered Wikipedia account? This is for at least two reasons. The first is that I am not a native English speaker and your use of language is likely to be far more punchy, persuasive and lyrical than mine (as you've recently been demonstrating on our Manchester article). The second, I think you can guess.

If there is no opposition, you can then make the edit to the relevant template directly - it's very easy to do. --W. Frank^email_talk 17:05, 13 September 2013 (UTC)Reply

Ok, happy to do so! I'll wait and see if anyone else posts here with any more ideas or thoughts but, if not, I'll suggest it there later on this evening. (Thanks for the Manchester comments - you're very kind! :) ) --Nick ^talk 17:22, 13 September 2013 (UTC)Reply

I like the new link for destination articles, but I would like to see an alternative to use when linking to travel topics, itineraries etc. For example see the WP article Ferry which has a link to Boat Travel. Would it be possible to either allow alternative text with a field or create a second template? AlasdairW (talk) 19:56, 13 September 2013 (UTC)Reply

Wikivoyage has a travel guide for journeys using ferries and other ships.

Please bear in mind that I am not using the actual Wikipedia templates here but rather facsimiles of them, but the actual Wikipedia template allows for a title of choice to display for the link to the actual article, so in your example, you might want to use something like this piped text. --W. Frank^email_talk 20:56, 13 September 2013 (UTC)Reply

I think the colon should be removed and a full stop added to the end of the sentence. Nurg (talk) 03:47, 14 September 2013 (UTC)Reply

I agree with colon removal and I'm agnostic about a final full stop. (Let's hope it can be done on the NHS). --W. Frank^email_talk 11:42, 14 September 2013 (UTC)Reply

The punctuation has now been done, following Template:Commons. Nurg (talk) 20:49, 27 September 2013 (UTC)Reply

Suggested here. Would 'guide to' rather than 'for' sound better? --Nick ^talk 12:16, 14 September 2013 (UTC)Reply

Probably - but it's ever so slightly less appropriate for some of the stranger titles of a minority of our articles, Nick. --W. Frank^email_talk 13:06, 14 September 2013 (UTC)Reply

Fair enough! :) --Nick ^talk 13:40, 14 September 2013 (UTC)Reply

I support the proposal to change "travel information related to" to "a travel guide for". And we appear to have a consensus to proceed. Nurg (talk) 21:01, 27 September 2013 (UTC)Reply

• Support for the change of wording to "a travel guide for". Danapit (talk) 08:14, 28 September 2013 (UTC)Reply

A rough consensus has emerged: It was proposed that "travel information related to" be changed to "a travel guide for" in the WV template at Wikipedia. No unresolved objections have been raised. Follow-up actions: Change made at Wikipedia. Added by: Nurg (talk) 01:39, 1 October 2013 (UTC)Reply

• I may be mistaken, but I think the original wording was to account for situations where we have a page that isn't actually a travel guide... or the linked travel guide does not have a one-to-one correspondence with the Wikipedia article. For example, w:Japanese language now has a box that says "Wikivoyage has a travel guide for Japanese phrasebook." Maybe we need more switches on the template? LtPowers (talk) 13:01, 1 October 2013 (UTC)Reply

The number one strategic task for the English Wikivoyage is to increase our readership. We've done some great work - assisted by our German Wikivoyage colleagues - with introducing dynamic maps and folks are furiously sharpening up the prose of our articles and keeping them up to date, but all this is somewhat in vain if nobody can find us in the search engines.

It's easy enough to get our new articles (such as Nelson (England) and Choum - that have no equivalent at Wikitravel) to appear in the top 3 search engine results.

I have now started an experiment to empirically discover what tactics we can use to make our legacy articles appear in the top 3 search engine results too.

On the same day, 12 September 2013, Somerset, Tasmania was created from scratch here at Wikivoyage and "Somerset (Tasmania) was created at Wikitravel. (I've deliberately not given the Wikitravel version a free hyperlink here, but I'm sure you'll find it over there).

In the Somerset, Tasmania article, I've deliberately practised some of what I preach below - for example, using a different lede for at least the first couple of sentences. I've also been much more radical by NOT using our "traditional" section headings that were inherited from Wikitravel and I'd make a heartfelt plea for nobody to change these if my experiment is to have any validity.

One reason I chose Somerset, Tasmania for my little experiment is that it was very unusual in not having a Wikipedia result on the first page when I searched from google.com, co.uk, ca,

Unfortunately, just two days after I created our Somerset, Tasmania article and linked from it to Wikipedia, our "big brother" is now showing up on the first page (and, at least from Glasgow IP's, Wikivoyage is too. Wikitravel article does not show up in searches from Glasgow yet).

Here are my suggestions:

1) Don't believe all the yarns that Google spins about "no-follow".

A significant self-help move we can make is to stop signalling to their robots that we are only an inferior derivative work every time they spider us. We should immediately junk the two free hyperlinks of relevant anchor text we give those nice, friendly folk at InternetBrands on the vast majority of our article pages (and on their associated discussion pages). We should immediately replace this kind of hyperlinked text

"This article is derived from the article Dresden on wikitravel.org/en in its revision as of 08:30, 6 July 2012 (UTC)

with this minimally compliant but legal attribution:

"This article is derived from the Dresden article on wikitravel.org/en in its revision as of 08:30, 6 July 2012 (UTC).

There was (and is) absolutely nothing in Wikitravel's licensing regime that mandated hyperlinking when giving attribution!

Doc James wrote on Wikimedia that "...since we have moved over the entire history of edits attribution is given... We probably do not need to mention WT at all."

Stefan2 pointed out: "If you download a PDF version ("Print/export": "Download as PDF" in the menu to the left), then Wikitravel isn't mentioned at all. Is this an error? If a clickable link is required, then I'm a bit troubled about Wikivoyage:Offline reader Expedition as you can't include a clickable link in a paper copy of Wikivoyage."

There are some technical tricks that our webmasters can do to mitigate our Google duplicate penalty (and I'm certainly not going to even outline them here in plain text - if anyone is interested they can phone, and those are, presumably the subject of the rather woolly bugzilla report) but there are also three more things that we can do to help ourselves IMMEDIATELY and I have seen no cogent opposition to:

2) Delete the literally thousands of "outline" articles that consist of only an opening "lede paragraph" in the form "x is a city/region in y" and a skeleton of (empty) standard sections. (Oh, I forgot, now thanks to the bot brigade these mostly now have a fine new banner and a link to a relevant Wikipedia article). We are not a Gazetteer and getting rid of roughly 20% of our articles and starting them again (as and when there is someone to take an interest in developing them properly), without duplicate text and the WV attribution to completely non-useful content, would go a long way towards signalling that we are a different (rather than a derivative) site.

In it:voy they deleted long ago all the skeleton articles (more than a thousand). Now (apart from a few hidden exceptions that may still exists) they only have articles with a minimum amount of useful information, and it's not allowed to create skeleton articles.

Indeed, Andyrom75 wrote at Wikimedia: "Now regarding the ranking I have the opposite problem, in some cases google is indexing also the page that I haven't created yet, but just mentioning (with a red link) in existing article. I've noticed that this cases are the ones where an article with that name exist in en:w but not in it:w. It could be good to develope those articles but I can't do all at once... :-( "

3) Change our ledes. Start with the 200+ country articles and high visibility articles like Paris, etc. I think most of us here at this specialised expedition already know that 95% of searchers don't go past the first 3 results. I think you also know that the results for "travemunde travel guide" and "mitzpe ramon travel guide" search terms will vary according to your IP, time of day, google domain and server that you are using, search history and other variables. That said, although the WT articles always appeared higher up the search result page by more than 4 places when I tried some tests, they did do much better than most WV "legacy attributed and hyperlinked articles" and I think I know the reason. In both cases, they are currently wearing Star nomination templates which changes the lede substantially as far as the Google spider is concerned. This gives the clue to another thing we should think about doing: try and drop the formulaic "x is a city in y" introductions in most ledes which almost always duplicates the WT intro.

4) Change all our "traditional" standard section titles. The "old guard" need to decide what they value most - tradition or a large and growing readership.

In many browsers, you can see the HTML that search engine spiders will see by pressing [CTRL] and [U]. Neither WT nor us have optimised our H1 tags and, right now, the important H2 tags signal that we are a mirror (and a cracked one at that) of WT. --W. Frank^email_talk 20:34, 14 September 2013 (UTC)Reply

I may be horribly out-of-date, since the last time I looked at search engine optimisation (SEO) methods was in the late 90s, but my understanding is that adding keyword= tags should be the first thing anyone does in SEO. Trying ctrl-U (thanks, Frank; I did not know about that) on our Yangshou page, I see no keywords. This strikes me as a blunder which should be fixed, pronto.

Comparing to the WT page, I see they have both a keywords= line and a description= line which repeats most of the keywords. It seems possible to me that this alone might account for the differences in search rank.

We need to be a bit cautious about this. My understanding is that Google penalises sites that put in too many keywords in transparent attempts to cheat on search rank. Someone whose knowledge is better & more current than mine should make those decisions. However, unless my info here is utterly out-of-date, adding some keywords should be a top priority task. Pashley (talk) 21:33, 14 September 2013 (UTC)Reply

Doing a Google search for "basic SEO" and looking at the first two sites — 7 Basic SEO Techniques All Webmasters Must Know and 10 Basic SEO Tips To Get You Started — I find both put a heavy emphasis on keyword tags. Pashley (talk) 21:44, 14 September 2013 (UTC)Reply

(Edit conflicts) http://www.homestead.com/blog/05/2013/7-basic-seo-techniques-all-webmasters-must-know#.UjTaRH85zAU isn't wrong, but the other site makes me weep at it's out of date naivety.

searchenginewatch.com has newsletters you can subscribe to if you wish to keep up to date with empirical research.

No harm in having a better description tag, but keywords have been pretty well ignored by most search engines since the last century. The four low hanging fruit are enumerated above (now in order of importance). I do have four fusion weapons up my sleeve, but I'm not going to explain their use to anyone that hasn't signed a confidentiality agreement since they are commercial secrets. Phone or e-mail if you're interested. --W. Frank^email_talk 21:50, 14 September 2013 (UTC)Reply

I've been deleting "Related pages" sidebar links some new users have been inserting, which tend to cover destinations already up the breadcrumb trail for articles and such. Should these links continue to be omitted from the sidebar, with the "Related pages" tab reserved for sister-site links (Wikipedia, Commons, DMOZ)? Ikan Kekek (talk) 21:47, 14 September 2013 (UTC)Reply

"Related pages" and "Related sites" are different. The former seems appropriate for the intra-Wikivoyage links, while "Related sites" is for Wikipedia/Commons. Melbourne is an example. -- torty3 (talk) 06:44, 17 September 2013 (UTC)Reply

Thanks, Torty3. However, that link (to UNESCO Creative Cities) is to a related category, not to a page that's already up the breadcrumb trail. Ikan Kekek (talk) 07:01, 17 September 2013 (UTC)Reply

Thank you for putting so much thought into this Frank! This is a really important issue for WV and something that we need to resolve ASAP.

Firstly, it would be interesting to know just how we would be able to change the keyword tags on here and how much emphasis Google places upon them. From what I gather, as has been said above, sites that overuse such tags are penalised. Do we know if WP uses such tags and whether the WMF has a policy them? If they are permitted and an appropriate number easily judged, it would be good to have them implemented soon.

As for the other points raised, I completely agree that our attribution to WT should be changed, although it will have to go through the WMF, so could take some time. Do we know if the WMF is yet considering such a change?

As I wrote in the pub, I would also be keen to see our lead paragraphs re-invigorated; if this come with an SEO benefit then this is definitely something to encourage.

Your other two suggestions (delete skeletons and change section names) are a little more radical and would worry me a little if implemented immediately. I would prefer to see keywords and a programme of introduction revitalisation implemented before we head down those avenues. If neither keywords or new leads have an impact, we could then consider stronger action. --Nick ^talk 21:59, 14 September 2013 (UTC)Reply

Nick - keywords should be irrelevant, see [22] (directly from Google). My knowledge of SEO is pretty good but not at top levels as I'm not directly responsible for SEO in my day job, but I work with people who are and have to keep it in mind in everything I do as a web developer. Here are some thoughts in no particular order:

1. I remain of the opinion that the biggest issues working against us are that our articles need to develop further so that our content is significantly different from WT, and that our site simply needs to be online longer as Google gives weight to a site's age. Google views us as a mirror of WT because the content is significantly similar, they've been around longer, and their pages historically rank higher.
2. A HUGE factor in some of the weightings is that WT has the word "travel" in their URL, and Google gives massive weighting to domain name. Do a search for a city name + "voyage" and we'll almost always be ranked higher. Overcoming that disadvantage is simply a matter of building up our ranking.
3. Removing the WT links (and ideally text) in the footer should also help, but until legal gives an OK that's not something that can happen.
4. I also believe that changing content solely for SEO reasons is the wrong way to go. Let's change content to make it better, and SEO will follow.
5. For basic SEO, the Mediawiki software automatically does most of what the SEO guides recommend - redirects, canonical tags, etc are all handled by the software.
6. If there are simple ways to increase our incoming links that would be an easy way to increase page rank, although I think we've gotten most of the low-hanging fruit. If people have their own blogs or sites, remembering to link to Wikivoyage when appropriate is important.

-- Ryan • (talk) • 22:37, 14 September 2013 (UTC)Reply

As far as I can tell this initiative is broadly technical in nature and seems concerned about how to improve the competitiveness of WV in search engines. This is all good although it mostly applies to a subset of 'power users' on WV who can discuss and action changes across the site.

Can we provide a shorter list of guidelines that 'standard' contributors can take that will improves the visibility of individual pages that they work on? I recently created an individual page Haeundae that does not exist on WT, and it has a Google ranking of 18. Not bad for a start, however would like to know how to improve... --Andrewssi2 (talk) 03:26, 24 September 2013 (UTC)Reply

Interesting news: Just 10 days after similarly named destination articles were created from scratch both here and over at Wikitravel, our article now always appears on the first page of Google's organic results for permutations of these common search terms "somerset tasmania guide" and always tops Wikitravel in the results. (Expedia still usually beats us).

I've used 12 different Glasgow IP's today to search various national Google domains and (except for Israel), results seem surprisingly regular. I'd welcome reports on my User talk page of any of your local search results using the search terms: "somerset tasmania guide", or permutations.

Obviously, I've only used two of my 4 "action points above ("Change our ledes" and "Change all our "traditional" standard section titles" so that they are not an identical mirror of Wikitravel) but I've also used some other subtle changes from our current MoS, such as Units.

Just as interesting to me - and a clear empirical rebuttal to Google's propaganda about "just write better and more useful text if you wish to rank better" - is that I have deliberately ensured that my experimental guide here is actually slightly weaker in terms of "hard content" - eg: it only has one "Dine" listing, whereas WT's has nine. --W. Frank^email_talk 15:27, 22 September 2013 (UTC)Reply

Did you note the relative rankings before you made your change, by any chance? --Inas (talk) 03:14, 23 September 2013 (UTC)Reply

I'm convinced there is something very odd going on with this. The WT article, has a very low ranking. Interestingly, on Bing, it doesn't even appear in the search results (and neither does WV). However, I think the results speak in favour of a broader experiment, and we should give a similar work over to a prime time article that we know rates lower. --Inas (talk) 03:15, 24 September 2013 (UTC)Reply

It's interesting that the WV article ranks higher in this case where the two articles were created around the same time. I agree with Inas that an experiment on a prime time article would be good. The chosen article should be one that has the same content on both WT and WV though -- based on the Somerset example, I can't determine if the headings and lede made a difference or if it's because most of the content is different. My own experience from re-writing a couple of ledes is that it doesn't help in the search results if that's the only difference in the articles. -Shaundd (talk) 08:02, 24 September 2013 (UTC)Reply

Shaun's right: I've done a lot of work to our Manchester article, but it still fails to appear in the first pages of results. Would it be worth thinking about what the new headings might be if we were to change them? --Nick ^talk 16:19, 24 September 2013 (UTC)Reply

I created a new Picton (New South Wales) article, and the content is pretty ordinary. The article now rates 3rd on Google for me, above much better travel guides for the same town. There was no corresponding article on WT. So, there is some google juice flowing our way, and it really does seem to be just the existence of an earlier, similar WT article that has such a dramatic effect. --Inas (talk) 04:30, 30 September 2013 (UTC)Reply

If I type the WT Somerset article URL directly into Google it is not finding the page, which suggests to me that it is not yet in Google's index. If that is the case we need to wait longer before any conclusions can be drawn. -- Ryan • (talk) • 01:17, 1 October 2013 (UTC)Reply

Typing the WT Somerset article URL into Google is now returning the relevant WT article, so I would suspect it may have just been crawled by Google. We might want to give it a few more days and then revisit the ranking as compared to WV in order to get any sort of accurate information about whether one site's article outranks or replaces the other. -- Ryan • (talk) • 00:03, 3 October 2013 (UTC)Reply

Inas: You wrote "Did you note the relative rankings before you made your change, by any chance?". There were no relative rankings "before change". Both articles were created from scratch on the same day.

It is vital if we are to overcome Google's current site assessment (that in all cases where there are two articles with the same or similar names, the Wikivoyage article should be regarded as an (inferior and inaccurate) mirror of the WT article in organic search results) that all 4 changes proposed above should be made at the same time and not sequentially over a period of weeks. To the original 4 action points, I would add a fifth: Change the naming scheme to that used by Wikipedia eg use Somerset, Tasmania rather than Somerset (Tasmania).

Action plan should, therefore, be:

1) Obtain permission from WMF legal for removal of hyperlinked attribution (to be replaced with human readable minimally compliant attribution - a link to our own imported article history page(s) may be sufficient)

2) Get timetable from technical for this site-wide change: "A-Day"

3) Change article names where necessary as close as possible to "A-Day" - use bot to replace incidences of parentheses?

4) Decide on alternative section titles so they are not the same "legacy" section titles of "Sleep", "Eat", etc as used by Wikitravel. Use bot to change all section titles (except "Go next" and "Contact") as close as possible to "A-Day".

5) Start the long process of changing all our ledes so they don't use the same boring formula as WT. This will take a long time, but start with our most popular articles and include 200+ country articles as priorities. --W. Frank^email_talk 13:57, 1 October 2013 (UTC)Reply

There is no evidence that this will work. The testing you've done on a new single article doesn't really support the strategy/plan. --Inas (talk) 21:08, 1 October 2013 (UTC)Reply

It works for this article. I agree that it is not the best example though as it is not an article that existed previously on both WT and WV. Travel Doc James (talk · contribs · email) 22:10, 1 October 2013 (UTC)Reply

I don't know why we would choose a new article to demonstrate a concept we want to apply to existing articles. I've worked with a existing articles. Yowah I deleted and recreated, because it was 100% written by me. No effect. Broome I've changed the headers, no effect. Before we go and make extensive changes, we need to point to a single existing article on both WT and WV that has been increased in google rank by any of these factors. I'm not aware of any, but lets keep working at it. --Inas (talk) 23:18, 1 October 2013 (UTC)Reply

Agree with Inas. It's a good result that the WV Somerset page is well-ranked, but the test is not all that relevant for our 20,000+ existing articles. Not only are the lede and section headings different between the WV and WT articles and there's no need for links back to a WT page, the content in the individual sections is different. Ryan indicated above that different page content is very important to Google search results and that's consistent with the limited stuff I've read about SEO. If this SEO stuff is as simple as changing headers, page names and ledes, that's great, but let's do some more due diligence to see if changing headings, page names, etc really makes a difference. A number of the changes Frank proposes will be contentious and not worth making (IMO) if they don't help out SEO-wise. -Shaundd (talk) 23:44, 1 October 2013 (UTC)Reply

I agree -- I think changing headers and switching to using long-form article names are both things I would definitely like to avoid at all costs unless it can be proven without a shadow of a doubt that they will substantially boost SEO in a way that absolutely nothing else will. Texugo (talk) 00:50, 2 October 2013 (UTC)Reply

Exactly, we need definitive proof before making any wide ranking changes. And definitive proof that we can all recreat. Travel Doc James (talk · contribs · email) 03:13, 2 October 2013 (UTC)Reply

• 40 DAY UPDATE: I've heard today that it's likely I won't be editing Wikivoyage for quite a few weeks, so I'd like to leave you (what I regard as) the final result of my Somerset, Tasmania experiment.

In the organic search results my article (with sub-section titles experimentally altered from our standard ones and a lede that avoids the formulaic "A is a city in B", and, vitally important, no hyperlinks or other clues signalling it is a sub-standard mirror of Wikitravel) now consistently outranks Expedia, Wikipedia and Wikitravel for the search term I was testing: "guide somerset tasmania" and, in so far as I can test things here in Glasgow always appears in the first three organic (ie not paid for advertisement/results) results positions, usually in first place. (The one exception, of course, is Australian Google if you ask it to prioritise results from Australian websites.)

Since the demise of Scroogle, it has been awkward to get anonymised, non-personalised results out of Google. Here is one website that helps with the US site: https://startpage.com

This tool will give you search results from Google websites in the US, Canada, Australia, Ireland, UK, Germany, etc, (and more, if customised) if you use Firefox as your browser - but beware, it is not anonymised as effectively as the first tool: https://addons.mozilla.org/en-US/firefox/addon/google-global/

Now, I appreciate that some of you may loathe and detest me (and have consistently libelled me as a fraud, liar and a cheat), but let me "put my money where my mouth is". I am so confident that, if you, adopt my five point plan below to boost our readership, I will absent myself for six months if it fails to lift our readership by 400% in the first 2 months of implementation (and one whole year if it fails to lift our readership by 200% in the first 2 months of implementation):

Action plan for a %400 uplift:

1) Obtain permission from WMF legal for removal of hyperlinked attribution (to be replaced with human readable minimally compliant attribution - a link to our own imported article history page(s) may be sufficient)

2) Get timetable from technical for this site-wide change: "A-Day"

3) Change article names where necessary as close as possible to "A-Day" - use bot to replace incidences of parentheses?

4) Decide on alternative section titles so they are not the same "legacy" section titles of "Sleep", "Eat", etc as used by Wikitravel. Use bot to change all section titles (except "Go next" and "Contact") as close as possible to "A-Day".

5) Start the long process of changing all our ledes so they don't use the same boring formula as WT. This will take a long time, but start with our most popular articles and include 200+ country articles as priorities.

I must stress the necessity for ALL these changes to be carried out simultaneously to send the strongest possible signal to Google that this is a site with new content and not a gradually improving (but still inferior) mirror of Wikitavel.

If you find someone willing e-mail me in confidence about implementing some subtle changes to some of our hard coded HTML tags and meta-tags I can promise a further lift of 400-2000% uplift in search engine directed traffic - however, the in-article changes must be done first and before the behind-the-scene changes. --W. Frank^email_talk 16:37, 15 October 2013 (UTC)Reply

I have said it before, and I'll say it again. This experiment has too many variables to know what elements are doing the trick. I want to see the elements tested separately and in various combinations, because if there is not black and white evidence that changing section headers and/or article titles are undoubtedly responsible for an improvement, I do not want to go there. Texugo (talk) 18:52, 15 October 2013 (UTC)Reply

Oh ye of little faith. How many times do I have to say this? My proposals are a package - none of them will work separately or in isolation to the extent necessary. Let me paint an analogy. If I tell you that you will get at least 40 miles per gallon out of an ancient Ford Fiesta if you do all 5 of:

1) adjust the brakes so they do not bind

2) inflate the tyres from 1 pound per square inch to 35

3) put 4 spark plugs in each of the vacant spark plug holes

4)connect all 4 high tension leads to each appropriate spark plug

5)remove the rusty water in the fuel tank, clean the fuel tank out properly, dry it and then fill it with clean petrol

Why ever would you think you can do just one (or even three) of these excellent suggestions and still get a result. Please specify exactly what you think you have to lose. You've wasted nearly a year now of doing things your way and its obvious to any human intelligence that we are the better site. The problem is that there is no human "Site quality evaluation department" at Google. Why ever would you not want four times the readership each day and every day? --W. Frank^email_talk 19:17, 15 October 2013 (UTC)Reply

Luckily, unlike in your analogy, testing each of the 5 points here in isolation and/or various combinations is not going to strand anyone on the side of the road. If we do find that taking all your suggestions together is the only way it will work, I won't stand in the way of bringing greater readership. But I'm not willing to support massive sitewide changes based on just your word and a single trial with a jumble of variables. We need a more scientific set of trials before we decide which direction to go. Texugo (talk) 19:46, 15 October 2013 (UTC)Reply

What exactly do you think you have to lose?

You can only do this once and simultaneously. Do it partially or in a half-arsed way and the chance is gone for ever.

"A-Day" is a bit like D-day. It relies on an overwhelming and (for the robotic Google spiders) an overwhelming and surprising radical change. You can not practice or test it. Right now my test article is the highest ranking of any of our articles. Better than USA for "guide usa" better than Paris for "guide paris france". Better than any other of our thousands of articles. You can't get better than the number one search listing position in Google. --W. Frank^email_talk 20:22, 15 October 2013 (UTC)Reply

"What exactly do you think you have to lose?" I could say the same exact thing to you. What do you think you have to lose by letting other users test your strategy before implementing massive site-wide changes that will take a lot of hard work to implement? Let other users try out these strategies in different combinations, see what results they get, see if they match your claims, and then we go from there. Isn't that the scientific method at work? If you have such confidence in your strategy, why not let others test it themselves? PerryPlanet (talk) 23:41, 15 October 2013 (UTC)Reply

Okay, I'm going to have a go at Broome. Currently, WT rates 3rd, another couple of travel guides above it. I've been through the first 20 pages of results, and there is no mention of WV. So it looks like a good target to me. --Inas (talk) 04:33, 30 September 2013 (UTC)Reply

Yesterday I changed the headings. Today, WT rates 4th. WV article still nowhere to be seen. Not even in the search results. --Inas (talk) 00:23, 1 October 2013 (UTC)Reply

Changing only one factor out of 5 will have little discernible effect. For the best effect, all 5 need to be changed on the same date. --W. Frank^email_talk 13:57, 1 October 2013 (UTC)Reply

Although I'm happy to support experiments based on any hypothesis, there is currently no evidence to support that the timing of the changes matters, or indeed that the changes themselves matter. So, I'm happy to continue with this experiment as is. --Inas (talk) 00:48, 2 October 2013 (UTC)Reply

I will give an article a try. Travel Doc James (talk · contribs · email) 00:24, 1 October 2013 (UTC)Reply

It appears to work. Took 8 days. Travel Doc James (talk · contribs · email) 09:42, 8 October 2013 (UTC)Reply

Stopped working today, both times searches were done while logged out. Travel Doc James (talk · contribs · email) 10:04, 9 October 2013 (UTC)Reply

Might I suggest that we also give moving articles a go? Both Bolton and Manchester/Castlefield - Petersfield have been moved by me, and both now rank on the first page of Google results. Perhaps a good test subject would be Birmingham (England), which we could move to Birmingham, England? --Nick ^talk 00:37, 3 October 2013 (UTC)Reply

First under what search term? Travel Doc James (talk · contribs · email) 05:26, 3 October 2013 (UTC)Reply

They rank on the first page under "bolton travel guide", "castlefield travel guide", "manchester castlefield travel guide" and the district is the first result for "manchester petersfield travel guide". --Nick ^talk 07:41, 3 October 2013 (UTC)Reply

What did they rank before the move? Travel Doc James (talk · contribs · email) 09:45, 8 October 2013 (UTC)Reply

Checking an article with ctrl-U to look at HTML, I see: <title>Xiamen – Travel guides at Wikivoyage</title>. My guess is that a small change in the scripts that turn wikitext into HTML could make that <title>Xiamen travel guide at Wikivoyage</title>, or just <title>Xiamen travel guide</title> since the site name is given elsewhere, and that would improve search results. Pashley (talk) 11:46, 8 October 2013 (UTC)Reply

Google seems to totally ignore commas and parentheses in search, though. Odd it would pay attention to them in titles. --Inas (talk) 22:09, 9 October 2013 (UTC)Reply

Does it ignore the 's' in a plural form? Or presence of an exact phrase? If a user searches for "Xiamen travel guide", as seems likely, will my suggested "Xiamen travel guide at ..." be better than the current "Xiamen - Travel guides at ..."? Pashley (talk) 19:38, 10 October 2013 (UTC)Reply

I just did a search for "Moss Beach travel guide" and "Moss Beach travel guides", and to my surprise the plural form showed up on the second page of Google results while the singular form didn't show up in the first three pages of results. We have control over the content of the title tag (MediaWiki:Pagetitle) so I'd suggest changing the current "$1 – Travel guides at Wikivoyage" to something like "$1 - Travel guide at Wikivoyage" or "$1 travel guide - Wikivoyage" (note: the second title duplicates WT and might cause duplicate content issues). Anyone have any thoughts or other suggestions? -- Ryan • (talk) • 18:07, 31 December 2013 (UTC)Reply

Sounds like a good idea; well worth testing out! We could try '$1 - Wikivoyage travel guide'? --Nick ^talk 18:18, 31 December 2013 (UTC)Reply

With the caveat that in my day job I'm involved mainly involved in SEO decisions that are unrelated to specific keyword choices, my understanding is that the page title is very important and it would be better to use something that is closer to the actual search term, so "$1 - Travel guide at Wikivoyage" would be better than "$1 - Wikivoyage travel guide" since a user would be more likely to search for "$1 travel" or "$1 travel guide". If that understanding is wrong then hopefully someone else can correct me. -- Ryan • (talk) • 18:37, 31 December 2013 (UTC)Reply

Is the title text customizable per namespace? Powers ^(talk) 20:20, 2 January 2014 (UTC)Reply

As far as I know it is not customizable per namespace, but I haven't done any research on the matter (the current title of this page is "Wikivoyage talk:Search Expedition (section) – Travel guides at Wikivoyage"). Did you want to propose an alternate text that would be more generic across namespaces? The danger of that approach is that the title tag is our most valuable SEO tag, so it should really include the words "travel guide" in some way, even if that means that non-mainspace articles will have slightly misleading titles. -- Ryan • (talk) • 06:14, 3 January 2014 (UTC)Reply

No, I think you might be reading too much into my innocuous question. =) I mainly asked because I believe we had such a setup at WT, in which "travel guide" was only appended to mainspace article titles. (Looking now, I see they also managed to avoid appending "travel guide" to travel topics and phrasebooks, while maintaining it for itineraries and destination guides, even though they're all in the mainspace!) I would like to see us find a phrasing that doesn't look like we're copying WT, if possible. Maybe "Wikivoyage - Free $1 travel guide" or something like that? Powers ^(talk) 16:41, 3 January 2014 (UTC)Reply

I'll leave it to others to figure out if we can change things per-namespace (I thought WT did it with custom RDF that Evan wrote, but could be wrong). With regards to the title text, for SEO purposes it is important to have the most important terms on the left (among many others, see [23]), so "Wikivoyage" really should be at the end since it's our least important search term. While using the same title as WT would be unfortunate, the article name ("Chicago", "Asia", etc) really does need to be first since it's the most critical and descriptive term, and "travel guide" should also feature prominently unless there is another set of terms that we feel are important. -- Ryan • (talk) • 17:03, 3 January 2014 (UTC)Reply

Since there isn't objection to changing the title I've changed it to the singular "$1 - Travel guide at Wikivoyage" while we try to figure out if something else would be better. -- Ryan • (talk) • 17:32, 4 January 2014 (UTC)Reply

Excellent decision! --118.93nzp (talk) 19:46, 21 January 2014 (UTC)Reply

Could we change the page description of the Main Page as well? I just noticed that it doesn't contain the words 'travel guide'.... --Nick ^talk 01:19, 6 February 2014 (UTC)Reply

Given that minor tweaks to page titles can make a big difference to how a page is ranked in search results I've changed the page title pattern from "$1 – Travel guide at Wikivoyage" to "$1 travel guide - Wikivoyage". The downside of this change is that the pattern is now the same as WT and it is not ideal for non-travel pages (like user pages), the upside is that it makes "PAGENAME travel" and "PAGENAME travel guide" more exact search matches. If there is objection I would ask that we try out the new pattern for a couple of weeks just to see if it makes any difference before the change is reverted. -- Ryan • (talk) • 23:23, 13 February 2016 (UTC)Reply

Would changing the names like London (Ontario) to match what WP & Commons use, e.g. w:London, Ontario, have a useful effect? I think we should do it anyway & there has been some discussion somewhere, but I cannot find it now. Pashley (talk) 02:48, 14 February 2016 (UTC)Reply

From an SEO standpoint anything that can be done to make Wikivoyage content different from other sites would be valuable, and page URL is one of the most critical ranking factors for a search engine. That said, renaming a lot of pages would be a huge job, so I'd suggest soliciting support for a very controlled test on perhaps 10-20 pages, making sure nothing changes during the test period but the page URL, and seeing if there is a noticeable change in search result rankings. Depending on the result a broader change would be worth considering. -- Ryan • (talk) • 05:02, 14 February 2016 (UTC)Reply

I've reverted the pattern back to "$1 – Travel guide at Wikivoyage" as I'm not seeing any change in the ranking of articles I monitor, so if there was any benefit it seems to be either very small or else offset by the fact that the page title was adding to our duplicate content penalty. -- Ryan • (talk) • 03:34, 8 March 2016 (UTC)Reply

Another way to get links and eyes on to our content, is to encourage additional reuse. The screen scraping required, and risk of being blocked for copying data is a big problem at WT. Here, we have full db dumps. I think a good page on how to reuse our content, the link back that we would like for attribution could encourage other sites to use WV info. I might start work on a How to reuse WV content guide. --Inas (talk) 23:48, 15 September 2013 (UTC)

Excellent point, Inas. Go for it! --W. Frank^email_talk 23:57, 15 September 2013 (UTC)Reply

We already have Wikivoyage:How to re-use Wikivoyage guides. That may need improvement or better links — main page? welcome, business owners? redirects? elsewhere? — or it might need to be advertised outside WV, but we do not need another page. Pashley (talk) 02:25, 16 September 2013 (UTC)Reply

Yep. I'm happy to start there. --Inas (talk) 06:49, 16 September 2013 (UTC)Reply

I've had a first pass. Feel free to join in. I think we rank #1 in google rank for a search on "reusing free travel guide". The word reuse is awkward. --Inas (talk) 05:39, 17 September 2013 (UTC)Reply

Wikivoyage:Travellers'_pub#Hiding_the_page_titles_in_the_main_namespace has people congratulating themselves for getting rid of titles. Checking a random page with ctrl-U, I find we no longer have a <title> tag. I suspect that is a bad idea in terms of SEO — I want to keep it and add <keywords> and <description> tags — but am asking here so people who know more of SEO can comment.

I also notice that there is much javascript involved. How does that work for spiders that presumably do not interpret scripts, or for people that disable javascript for security reasons? (I do for most sites, but have WV whitelisted.) Could we get rid of some scripts without either doing unacceptable damage to functionality or overloading the servers by making them do work that is currently farmed out to client machines? Pashley (talk) 17:34, 19 September 2013 (UTC)Reply

The <title> tag is still there on pages that have page banners, although you're right that the <h1> that displays the page title is being hidden by Javascript. Some refactoring of Template:Pagebanner is probably in order to get rid of the current DISPLAYTITLE mess, although it's not obvious to me that this issue would hurt our SEO - it doesn't help, and the page banner should probably be modified to use an <h1> for the page title, but I don't think we're committing any major SEO sins with the current approach. As to the use of Javascript, what we're doing is fairly benign and is required to overcome deficiencies in the Mediawiki software when it comes to suppressing page titles. I can make an attempt to address some of these concerns later today if there are no objections. -- Ryan • (talk) • 17:50, 19 September 2013 (UTC)Reply

It's much better from an SEO perspective to have relevant text between the <h1> opening and closing tags. --W. Frank^email_talk 19:54, 19 September 2013 (UTC)Reply

Template:Pagebanner has been updated to remove the DISLAYTITLE bits (which cleans up the CSS around the title tag) and some unnecessary bolding. It is not possible to change the title in the page banner to a heading tag without having it show up in the TOC, so I left that part alone. -- Ryan • (talk) • 21:21, 19 September 2013 (UTC)Reply

In light of the discussion above, I think it would be good if we reviewed pages that have undergone significant changes in the past few months and see how they appear in Google search results. I'm assuming that all WV pages started off in the nether regions of Google's search results (which may not be entirely true, but it was my experience that Feb/March I couldn't find any of our pages in search results where I typed "PLACENAME travel guide"). The list below is articles I've worked on plus other guides that I'm aware have been significantly changed in the past few months. If you're know of any others, please add them below and let's see if we can find any patterns.

• Buffalo - guide that has been districtified. Substantial changes to Districts, See, Do, Eat, Drink and Sleep but much of the original text remains in other sections. Result - listed on fourth page of search results; WT article on page one.
• Garibaldi Provincial Park - outline that was deleted and started from scratch. Result - ranks above WT article with same name on first page of results.
• Lions Bay-Britannia Beach - outline guide with a different name from the equivalent WT guide. Substantially rewritten. Result - ranks on first page of Google results, but below WT article.
• North Vancouver - guide status article where I've rewritten the lede, Understand, added a map and added several new listings. Result - now on the second page of Google results (previously not ranked at all); WT guide on first page of results.
• Pemberton (British Columbia) - outline article that's been substantially rewritten. Result - search result depends on how name is searched. Just "Pemberton" lists WT ahead of WV on page two, while "Pemberton British Columbia" lists WV ahead of WT on page one.
• Queenstown-Lakes - outline that had its lede re-written and content added to nearly all of the other sections. Result - ranks on first page of Google results, but below WT article.
• Shanghai - guide article with quite a bit of work done but a fair bit of the original text remains (maybe half). Result - listed on fourth page of search results; WT article on page one.
• Tokyo/Roppongi - guide article with a new Understand section and updated listings (some new, some deleted, some modified). Result - on the second page of Google results; WT guide on first page of results.
• Travemünde - outline article on WT that's been rewritten and now nominated for Star on WV. Result - ranks on first page of Google results, but below WT article.
• Vancouver - guide status article where I've rewritten the lede and Climate section but nothing else. Result - still lost in the nether world of Google search results.
• Yowah - usable article that was deleted and recreated with the same content (with permission). Result - ranks above WT article with same name on first page of results.

Cheers. -Shaundd (talk) 00:20, 2 October 2013 (UTC)Reply

Just a reminder to make sure you turn off personal search results when doing this test! You can do so by clicking the globe icon (as opposed to the head icon) in the upper-right corner of the Google search results page.

• Letchworth State Park — Outline that was almost entirely rewritten in March, with pictures added; Buy, Eat, Drink, and Go Next left as-was. Our article is now number 2 on the list, with WT's article number 1.
• Erie Canal — Outline itinerary with lots of background info but only a few actual destinations. I've since added several more destinations. Our article is number 4 on the search results; WT's is number 2.

-- LtPowers (talk) 16:26, 2 October 2013 (UTC)Reply

I've been rewriting and adding a number of cities in the Netherlands and checking SEO results too. The best results I'm seeing are for articles that were skeletons at the old site, have been deleted here and recreated as full texts. Example: Roermond (result: 5th, after eg tripadvisor, but above WT). JuliasTravels (talk) 20:20, 2 October 2013 (UTC)Reply

Should we make a table of these on the front page? --Inas (talk) 23:08, 2 October 2013 (UTC)Reply

A table is a good idea. I struggle with them so I took the easy way out and just made bullets to save time (sorry!). -Shaundd (talk) 23:12, 2 October 2013 (UTC)Reply

I've created a table now and included Manchester, Bolton and Horwich within it. Manchester's still proving elusive, but Horwich is one below the WT article, whilst Bolton is one of the few articles imported from WT that are now ranked above their equivalents there. --Nick ^talk 23:52, 2 October 2013 (UTC)Reply

I created an article on Sharon (Connecticut) on Sunday, and I see it nowhere in my Google search results. How long does it usually take before articles are indexed by Google? Ikan Kekek (talk) 00:22, 3 October 2013 (UTC)Reply

Nick - nice table!

Ikan - I think it's taken up to two weeks for new articles I've worked on to appear in Google results (but sometimes faster). Significant changes to existing articles have taken 4-5 weeks to be reflected in the search results. It doesn't seem to be a fast process. -Shaundd (talk) 00:45, 3 October 2013 (UTC)Reply

Nick - I'm showing our Bolton article as #2, below WT's at #1. This is with personal search results turned off. LtPowers (talk) 14:15, 3 October 2013 (UTC)Reply

Whoops! Maybe I spoke a bit too soon! I can't work out how to turn off personalised search in Google's new look - any ideas? --Nick ^talk 16:01, 3 October 2013 (UTC)Reply

Are you using google.com or google.co.uk? The switch is prominent on the former but I can't it at the latter. But that could be because I'm not signed in. LtPowers (talk) 23:10, 3 October 2013 (UTC)Reply

I'm on .co.uk, but it's proving a bit elusive! I have tried signing out and changing my supposed location, but I don't know quite how it works. I can only presume that you're right and I'm still viewing personalised results. --Nick ^talk 23:12, 3 October 2013 (UTC)Reply

Also, I notice Shaun had Yowah as an article that ranked above WT. However, my searching previously and now shows that it actually doesn't rate (i.e deleting a recreating an article does nothing). I think we're mixing up WP, accidentally. Can someone else confirm that it doesn't appear for them? --Inas (talk) 04:25, 4 October 2013 (UTC)Reply

I have ours at #2 and WT's at #3, with personalized results off. LtPowers (talk) 12:26, 4 October 2013 (UTC)Reply

I should point out that I'm searching for "Yowah travel guide"; if I search for just "Yowah", WT appears on the first page, but WV does not in the first 10 pages. LtPowers (talk) 12:27, 4 October 2013 (UTC)Reply

Same here, just Yowah gives me nothing for us, but both Yowah travel and Yowah travel guide put us second, above WT (personalized results off). JuliasTravels (talk) 09:51, 8 October 2013 (UTC)Reply

Okay, the difference is between google.com.au, and google.com. google.com.au Yowah travel guide doesn't rate. It's first on google.com. Signed out on both occasions. Okay, I'm interested in this one now, because nothing has changed except the article being recreated. --Inas (talk) 22:14, 9 October 2013 (UTC)Reply

Surely this has been discussed before, but I can't find it just now and I'm wondering...is there serious negative impact of that wt link in the attribution text on the bottom of all imported pages? The text states: This article is derived from the article Amsterdam on wikitravel.org/en in its revision as of 18:28, 29 July 2012 (UTC). There it is published under the Creative Commons Attribution/Share-Alike 3.0 License. More details and the full list of contributors can be found on the associated history page. (Wikitravel contributors are marked by the prefix WT-en.). Would it make sense to lose that direct WT link? The license gives us CC obligations towards the original authors (or their pseudonyms, if applicable). The current WT copyleft/"How to re-use Wikitravel guides" text states that one "must link back", but that notice was added after the fork so isn't applicable for the articles we imported. We could be nice and leave the mention of the old site without a link, or change to wording and link to an explanation page about the fork here on WV, with some extra info on user names and article histories and perhaps a single direct link there. Would it make a difference compared to the situation now? And am I missing a reason why we're not doing this already? JuliasTravels (talk) 20:50, 2 October 2013 (UTC)Reply

I have emailed the WMF regarding this and have not heard back. They are busy with fundraising right now. If we cannot fix it before than I will bring this issue forwards at Wikimania next Aug in London. Travel Doc James (talk · contribs · email) 05:34, 3 October 2013 (UTC)Reply

Thanks for the fast reply, James. So the way this is done now was demanded by WMF as part of the fork agreements? Or why would we need permission? JuliasTravels (talk) 09:12, 3 October 2013 (UTC)Reply

There's been a related discussion at meta recently (here) and a bugzilla request on the subject. Not sure how to interpret the outcome though. It sounds like "let it be" kind of recommendation to me. --Danapit (talk) 09:29, 3 October 2013 (UTC)Reply

That's an easy kind of recommendation to give for a legal manager who has other things to worry about and obviously would prefer not having to deal with a minor issue for a small project :-) It seems just a bit too important to leave it at that, though. Just to clarify: do we /need/ WMF tech support to make this change or can we (in theory) change it ourselves? JuliasTravels (talk) 10:02, 3 October 2013 (UTC)Reply

Looking at the code again, there might be an easier way to remove/delink the second WT link since it's just a constant. The first one is linked to a specific revision from the database and is at least useful. I think it was previously created for the first Wikivoyage fork, then later applied in the second migration, so not really mandated by WMF. I wanted some legal advice because while I don't think the direct link is necessary, it doesn't mean people can't be litigious over it. -- torty3 (talk) 10:11, 3 October 2013 (UTC)Reply

The WMF legal team will not give us advice personally or as a group. If we are able to change it we may as long as we say within the CC BY SA license. We give attribution within the history tab and IMO that is enough. They will only get involved if complaints to the foundation occur. Otherwise we will need to have discussion at Wikimania.

I guess the big question is does this attribution affect our page rank. If it does than we should worry about it. If it does't than we shouldn't. Travel Doc James (talk · contribs · email) 05:23, 4 October 2013 (UTC)Reply

torty3, could you explain how to remove those links that can be removed easily? We would like to give it a try on Russian Wikivoyage, regardless of what it brings in terms of SEO. It is just annoying to see all these links, especially in those articles that we wrote ourselves. --Alexander (talk) 13:29, 4 October 2013 (UTC)Reply

The page to edit is MediaWiki:Creditssource-source-work, which looks like:

This article is derived from the article $1 on $2 in its revision as of $3 (UTC). There it is published under the <a href="https://creativecommons.org/licenses/by-sa/3.0/ ">Creative Commons Attribution/Share-Alike 3.0 License</a>. More details and the full list of contributors can be found on the associated $5. (Wikitravel contributors are marked by the prefix $4.)

We would replace $2 with the plaintext. Since it's easy to change/revert, I don't mind doing it either. -- torty3 (talk) 14:48, 4 October 2013 (UTC)Reply

Something must be feeding those variables into that text; it seems like it would be better to change that source rather than the message. LtPowers (talk) 20:54, 4 October 2013 (UTC)Reply

That would be the proper way, yes. It's filed at Bugzilla: 53942, but I explained at Meta that it would take a week to change/revert, and under the circumstances would prefer a quicker trial, before updating that report. -- torty3 (talk) 02:50, 6 October 2013 (UTC)Reply

Fair enough; if it pans out in the end I would suggest we make the change at the source when feasible. LtPowers (talk) 15:54, 6 October 2013 (UTC)Reply

Wow, it's so simple. Thank you very much! We will see how far we can go. --Alexander (talk) 21:00, 4 October 2013 (UTC)Reply

I was over-thinking matters :) -- torty3 (talk) 02:50, 6 October 2013 (UTC)Reply

Would it be possible to somehow isolate a bunch of articles (say 50 or so) and make this change ourself for those only? If we would have some kind of proof that losing that direct link really does have a serious positive effect, that might strengthen our cause with WMF. JuliasTravels (talk) 20:18, 18 October 2013 (UTC)Reply

Some way of testing would be nice, though I don't know if there is any other way apart from the shortcut I suggested. Somerset doesn't work as an example article except for new pages because 1) it's a new article, 2) with no pre-existing counterpart, 3) with no duplicate content. Deleting and recreating does seem to work well, but can't tell if it's because it's a new article or because of the link. Could Yowah be replicated in terms of copyright? As in, choose a page in the Netherlands for example, of entirely your own work and recreate it exactly as it is. -- torty3 (talk) 14:54, 22 October 2013 (UTC)Reply

I think so. A page that was made back at the WT site, deleted and then recreated straight away or after some time? JuliasTravels (talk) 15:07, 22 October 2013 (UTC)Reply

Have made some changes. We now mention WT once and link to WT once. That is more than sufficient per copyright. We could even remove the bottom bit entirely and simply make a note in an "edit summary" which would than show up under the history tab. The attribution that we as contributors get is all in the history tab. There is no reason why any other entity should get any greater attribution than we get as individual contributors. This latter change though would require a bot. Travel Doc James (talk · contribs · email) 09:32, 2 November 2013 (UTC)Reply

Thank James. Personnally I don't see any reason to credit Wikitravel at all, since we all own the copyright to our own work which happened to have been done there -- 178.197.236.124 20:58, 3 November 2013 (UTC)Reply

Agreed. The fact that WMF Legal declined to require us to keep it is telling. LtPowers (talk) 02:22, 4 November 2013 (UTC)Reply

Good point. For all edits from WT, the users origin from WT is listed in the history tab. If you delete the remaining text here the in article attribution will be gone [24]. I am supportive of someone doing so. Travel Doc James (talk · contribs · email) 12:48, 6 November 2013 (UTC)Reply

I believe the legal issue partly stems from some of the hasty decisions that were regrettably made when this project was copied from the original source. One, all instances of the word Wikitravel being replaced with Wikivoyage without considering how nonsensical or even misleading that would make many edits. Two, the method used to import usernames from the old wiki by appending them with "WT-" creates a big problem. Although many creators of Wikitravel are here now, by far more never came. So for instance a Wikitravel user "John" is known here in page history as "WT-John". But "WT-John" does not exist. John is the author of his content. Attributing his work to someone named "WT-John" is not adhering to the CC/SA license because it attributes to the wrong user (one who in fact does not exist). From what I know of lawyers, independent of any agreement with the Wikitravel owners, I would say this would pose a very big legal stumble block for the project to get rid of the attribution links. Also in my opinion, no follow means exactly that. I doubt Wikitravel is receiving any benefit, nor we any downgrade, due to them. It is because we are still very nearly the same content that they are. That will change in time. But I would suspect google is designed to subvert quick solutions to this issue. As someone said, this is the price of forking. I agree with those saying we should concentrate on content, not on the suppositions that seo tricks are the way to more contributors. --SpendrupsForAll (talk) 02:22, 7 November 2013 (UTC)Reply

The first point (about each Wikitravel name being pre-pended with "WT-") need not be a stumbling block if each and every page edit history that contains an editor name pre-pended with "WT-" has some small print explaining the pre-pend.

I do not believe that the "price of forking" should be eternal damnation to the fires of Google's duplicate penalty. Sorry to be so direct, but it's just Google PR bullshit to believe that "no follow means exactly that". If it really did, then there would not be so many link spammers at WT and attempts made at WP (we're still relatively immune since our tiny readership and Google ranking make us a less handsome target). Please study the SEO literature to see empirical results for ignoring the myth that "no follow means exactly that". Now you can all talk about this till the cows come home, but, the plain fact is that Frank has told you the solution and this delay in implementation is just timewasting.

I confess I am not an attorney. Unless you are, I believe I'll refrain from taking legal advice from you, as will the people whose decision this ultimately is. Although I'm sure your advice is appreciated, simply stating that explaining why we are not attributing content to its actual creators is probably not going to hold up in a courtroom. And given the litigiousness of the parties involved, and the staggering sum already spent, I doubt this will happen at all. Secondly, again though your opinion has been noted (many, many times) you've proven no expert on seo or anything else, and no experiment has yet proven that removing links, deleting and recreating articles, adding keywords to summaries, or any other trick has succeeded in improving a page's rank any more than just leaving it alone would have done. This is about having distinct content, nothing more, and I think the link removal discussion is motivated by spite as much as anything and is a distraction from our goals and values. Also, Frank, please sign your comments and refrain from referring to yourself in the 3rd person in them. It sounds puppety and we don't want to open that worm can again.--SpendrupsForAll (talk) 19:45, 7 November 2013 (UTC)Reply

Please feel free to remove this my entire reply - if you will strike your last sentence as both unfair and inappropriate to both user:W. Frank and I. Relevant commentary is on your own user page. Sorry to have been unnecessarily abrasive before. --118.93.67.66 01:00, 8 November 2013 (UTC)Reply

To your first point: this topic of adequate attribution is not just dealt with in this sub-section, but also above if you would be so kind as to read there too.

To your second point: I think we have comprehensively proved over the course of the last year of Wikivoyage editing that, however much we improve article content incrementally, we will never beat the Google dupe penalty unless Frank's suggestions above are adopted. It's a bit chicken and egg regarding experimentation, isn't it? I do share Frank's opinion that it is best to make one co-ordinated effort to tell Google that we are a NEW site and not a mirror. Failed, Half-arsed attempts will only raise the bar for later efforts. A once-only, co-ordinated effort is required across the entire site. --118.93.67.66 01:00, 8 November 2013 (UTC)Reply

This appears to have finally been fixed. Hooray! Hooray! The cynic in me says "About time". In case, definitely kudos to whoever did it. Pashley (talk) 23:39, 12 March 2014 (UTC)Reply

Okay, this one I find odd and want to share. Our Hlotse article has been deleted 3 days ago as part of the list of outlines I'm recreating. Searching for just "Hlotse" on google gives me a link to our (currently not existing) article at nr. 4 however, while the WT one (with identical content, but actually existing) comes only at the second page. "Hlotse travel" even puts us second, after WP, and WT 3rd. The article is linked from WP, however, with a Wikivoyage-inline template. Not sure how long it'll take google to realise it doesn't exist anymore, but well.. just some food for thought :-) JuliasTravels (talk) 18:59, 16 October 2013 (UTC)Reply

The same is true for some of the other deleted ones in my list. Searching for "Spata travel" puts us 10th, but still 5 places above WT, for example. And unless I'm blind, that one does not seem to be linked from WP. Better yet, the ones I've recreated (e.g Aquiraz travel), also show up now, and above WT. Did you guys change something or what? JuliasTravels (talk) 19:24, 16 October 2013 (UTC)Reply

I suspect that the search engine is detecting a big change from the old page to what it sees after the page has been deleted, which is just the page title plus "There is currently no text in this page. You can search for this page title in other pages, search the related logs, or edit this page.". If you are going to be deleting other pages and then recreating them, it might be worth waiting 2 weeks or more between deleting and recreating, so that the search engine sees two big changes in content. AlasdairW (talk) 21:11, 21 October 2013 (UTC)Reply

If you think it makes a difference, there's no harm in trying that. Still it would be strange that a deleted version gets better results than new articles from scratch? JuliasTravels (talk) 22:12, 21 October 2013 (UTC)Reply

Andyrom75 reported the same thing happening for deleted Italian Wikivoyage articles. I think the empty WT articles don't rank well in the first place, then Wikivoyage wins out in terms of being an entirely new page with no duplicate content. Still very odd, black magic behaviour. -- torty3 (talk) 14:44, 22 October 2013 (UTC)Reply

Swept in from the pub

or get rid of Frank for six!. How can we lose? --92.26.121.21 16:46, 15 October 2013 (UTC)Reply

I don't think anyone is arguing to expel Frank for 6 months, at this point. And speaking personally, I hope that there is never a cause for any more blocks of his account of any length. Ikan Kekek (talk) 17:13, 15 October 2013 (UTC)Reply

Like Ikan Kekek, I would not wish to lose Frank for 6 months, however, I appreciate the confidence with which he has put forward his findings and proposals. If we do implement the suggested changes, we'll have to be very organised and diplomatic about it - these are large changes. Personally, I'm happy to see parentheses replaced by a comma and extended lead paragraphs can only be a good thing. I'm slightly concerned about (although not opposed to) a change of headings, if only because I think the ones we've got are very good and will take some time to better. If we want to do this quickly (and in a single swoop), that could be a sticking point, but let's see how we go. Are we going to go ahead with these proposals? Wikivoyage is feeling a little empty at the moment... --Nick ^talk 18:23, 15 October 2013 (UTC)Reply

Changing our "traditional" section headings are just one (integral and highly important) part of a 5-point plan to kill off our "inferior mirror of Wikitravel" status with Google. Within 4 months of adopting ALL of these linked changes it will be WT that is regarded as an "inferior mirror of us" and then we can change all our section headings back again (if that's the consensus) to ram home WT's inferior status. Job done! --W. Frank^email_talk 18:36, 15 October 2013 (UTC)Reply

Please keep the discussion there if possible.Texugo (talk) 18:54, 15 October 2013 (UTC)Reply

Agreed with Frank. Being "better", while it would be great, isn't really the goal. The goal is to differentiate ourselves from WT, at least as far as Google is concerned. The problem as I see it speaks to what Nick said. Even if they're not improvements, we obviously want any changes we make to section headings, etc. to be at least equally good as what we had before. And what synonyms are there for, say, "Eat", "See", or "Do" that get the point across as well? -- AndreCarrotflower (talk) 19:11, 15 October 2013 (UTC)Reply

Eat: Eating, Dining, Food; See: Sights, Look; Do: Activities, Doing. Are these equally good? Not sure. I think "Food" and "Eating" are as good as "Eat," but "Dining" has other connotations. I think "Sights" is as good as "See," but "Look is inferior. And I think "Activities" is as good as "Do," but "Doing" is a bit strange. Ikan Kekek (talk) 20:50, 15 October 2013 (UTC)Reply

I'm only a recent immigrant to the project but I like your verbs-as-section-titles format and wouldn't want to lose that. I can't think of many obvious alternatives but if you do change the standard sections, keeping to verbs would be nice. (eg. "Get in" could be "Enter", "Get around"->"Traverse" or even "Travel", "Buy"->"Shop", "Sleep"->"Stay" etc). - AdamBMorgan (talk) 21:18, 15 October 2013 (UTC)Reply

Would it work if we just changed some of the sections to present participles? e.g. 'Understand' -> 'About'; 'Get in' -> 'Getting in'; 'Get around' -> 'Getting around'; 'See' -> 'Sights'; 'Do' -> 'Activities' (or controversially, merge 'see' & 'do'?); 'Buy' -> 'Shopping'; 'Eat' -> 'Food'; 'Drink' -> 'Refreshment' (probably the weakest); 'Sleep' -> 'Staying' / 'Accommodatio0n'; 'Connect' and 'Go next' remain. --Nick ^talk 22:20, 15 October 2013 (UTC)Reply

At the risk of drowning people in suggestions, how about adding the Five W's (and one H) before the verb? Get in would be How to get in; Get around -> How to get around; Understand -> What to know; See -> What to see; Do -> What to do; Eat -> What to eat; Drink -> What to drink; Sleep -> Where to sleep; Buy -> Where to shop; and Stay safe -> How to stay safe. This could be a compromise for those who like the classic ones. Now that I'm actually writing these down, though, they seem long. Oh well, it can't hurt to suggest them… Nick1372 (talk) 00:17, 16 October 2013 (UTC)Reply

Yeah, I think they're pretty clunky and not good. :-) Ikan Kekek (talk) 00:35, 16 October 2013 (UTC)Reply

Stylistically consistent, unambiguous and not worse than the current standard should be the minimum requirement. Better would be good, Worse would not. • • • Peter (Southwood) ^(talk): 07:34, 16 October 2013 (UTC)Reply

I've tried some new headings in our Wigan article; please feel free to comment or change them - we could use that page as a sort of test bed, without disrupting the ongoing Somerset, Tasmania experiment. --Nick ^talk 10:12, 16 October 2013 (UTC)Reply

By all means discuss whether you think there are some improved names for section titles, but please don't confuse two entirely separate and distinct issues here.

I wish to increase our readership ten-fold in a few months. This will bring us new problems in the form of increased amounts of spam but it will also bring us new editors and patrollers.

The way to do this is to convince the ever-changing Google algorithms that we are not an inferior mirror site of Wikitravel (which currently beats us in Google's organic search results for most of our articles - the exception being my experimental Somerset, Tasmania article). We only need to do this once, then IB will pull the plug on Wikitravel since their advertising revenue will fall in step with their suddenly diminished readership.

Once we have convinced Google's crawling spiders that we are a changed site we will no longer suffer the crushing duplicate penalties.

Then (within reason) this site can have all the SEO sapping section titles and banners it wants.

This sea change must be a shock and awe, once only occurrence - not a gradual change - for it to work. That is why it is a VERY BAD IDEA to have a series of ad hoc individual experiments now.

By all means discuss what section titles you want to have AFTER "A-Day + 2 months" (maybe you'll decide it's just better to go back to our traditional titles) but, the section titles for "A-Day" must not be synonyms of Wikitravel's existing titles! Remember that "A-Day" is for robot readers - not for human readers. "A-Day + 2 months" when the job is done we can go back to (pretty well) disregarding what the Google spiders read...

I'm going into hospital now so I won't be writing here for some while, so please forgive me that I won't be here to guide your efforts. I wish everyone all the best! --W. Frank^email_talk 10:48, 16 October 2013 (UTC)Reply

I think some of us are getting ahead of ourselves with new section names and such when we haven't established which changes will do the trick, and several of us will be opposed to changing such things if not 100% established that it is what is needed. Again, I would really appreciate it if we would stop here and continue this conversation on Wikivoyage talk:Search Expedition, where this topic is already split into various different threads. We really don't need yet another parallel discussion going on here in the pub when so much has been written about it already there. Texugo (talk) 11:40, 16 October 2013 (UTC)Reply

It's definitely a good idea to keep discussions in one place but please discuss this rationally rather than continue to try and sabotage my experimental article. It's clear that you are implacably opposed to my project to increase readership (why, I'm not even going to hazard a guess at) but surely there can be just ONE exception to the current implementation of the Google search result slaughtering banners in our thousands of articles? I know you hate seeing my empirical results prove what I'm preaching, but try and exercise some judgement and self control, please. --W. Frank^email_talk 12:23, 16 October 2013 (UTC)Reply

I didn't simply revert your removal of the banner - I pointed you to the talk page where User:Torty3 explains how you are wrong about this. And I don't need you to tell me what my opinions are. I am not opposed to increasing readership, though I am opposed to your overly assertive and highly dismissive tactics in trying to force changes through, and I disagree that your results are empirical enough. Texugo (talk) 12:30, 16 October 2013 (UTC)Reply

As important as SEO efforts are, over a year has passed and I'm sure we can allow a few more weeks for some additional testing if that can get us to agree on a strategy. I understand that you're impatient, Frank, and this is our highest priority, but we're not in that much of a hurry. At the same time, I do believe we should allow Frank to have his own "experimental" article without interfering. Let's just agree on a few low-traffic test articles; there's no real harm in that, especially when compared to the gains we're trying to establish. JuliasTravels (talk) 13:35, 16 October 2013 (UTC)Reply

The test articles sound reasonable, but I'd be surprised if "changing everything at once" was hugely different than methodically updating one page at a time for SEO. The search engine is still just going to stumble across updates one article at a time as it isn't pulling the wikivoyage:database dump as one huge block. One option I'd like to see is to allow contributors to create "Whereverville/new" as a complete rewrite of an article on "Whereverville" that doesn't use any descriptive content from the old WT version. Any content added after the move to WV would be put into the new article with attribution, the old article would be discarded once the new one was at least to "usable" and the new page moved into its place. We don't have enough people to do this for every city, but having fresh content for a few of the most popular destinations would improve Wikivoyage. After all, a huge part of the reason why Evan and Maj decided a travel wiki was needed in 2003 is that it can be more current than a printed guide. Anything here that was recycled from WT is by definition anywhere from a year to a decade old, so much of it is stale. K7L (talk) 14:20, 16 October 2013 (UTC)Reply

With regard to Google pagerank, also see bugzilla:52688 and dependency bug (please, no "me too" comments over there). --AKlapper (WMF) (talk) 08:24, 25 October 2013 (UTC)Reply

We have been very successful with a newly created article. But have we had success with a long standing article yet? I tried with Cranbrook British Columbia and while it worked briefly it than stopped working. Travel Doc James (talk · contribs · email) 13:05, 31 October 2013 (UTC)Reply

I started a discussion related to links & search at Wikivoyage_talk:Airport_Expedition#Links_from_WP.3F. Pashley (talk) 16:07, 21 October 2013 (UTC)Reply

I did a Google search for "UK travel guide" and we were nowhere to be seen (WT came third). But a search for "United Kingdom travel guide" we were findable, but not in a great position (19th). I wonder if we should do something to make redirects more visible, maybe creating List of Redirects so that they get indexed by search engines. AlasdairW (talk) 21:02, 22 October 2013 (UTC)Reply

Redirect pages are coded such that search engines should already be paying attention to them. WT doesn't have an index, and they're doing fine. LtPowers (talk) 22:28, 22 October 2013 (UTC)Reply

As an experiment, I am going to add links to UK to a couple of low profile articles (towns in Kent) to see if this improves the ranking. I will wait a couple of days so that anybody that wants to can get "before" results. AlasdairW (talk) 22:25, 24 October 2013 (UTC)Reply

Two weeks ago, I added links to UK to three pages (Chatham (England), Chatham-Kent and Faversham). Now Google puts us in 46th position for "uk travel guide", when previously we were nowhere in 328 links. Not a brilliant result, but I think that it does show that it is worth having links to all (or most) redirect pages. Are there other redirects which were imported from WT which are likely search terms? Should we have a List of Redirects? (I assume that the search engine picked up on the WT redirects just after they were created from Recent changes.) AlasdairW (talk) 22:19, 11 November 2013 (UTC)Reply

Two weeks ago, I changed the headings (and nothing else) in our Wigan article (a WT import) to see if that alone would make a difference to the page's search ranking. At present, I'm yet to see much of a difference, but let me know if you notice anything or have any thoughts! --Nick ^talk 14:19, 31 October 2013 (UTC)Reply

And I did the same thing to Broome, to no effect. --Inas (talk) 23:02, 7 November 2013 (UTC)Reply

I was doing some research just before and wanted to leave this link with you all as a useful measure of how much of the content on one of our articles duplicates the content on WT. Not only does it provide a % score (lower is better), it also provides a list of content it considers to be identical in the two articles.

I popped in Doha, seeing as that's had a massive rewrite over the past few months, and apparently 28% of our article duplicates WT. It's worth noting that our updated and comprehensive Doha article is still nowhere to be seen on Google, while WT sits as Number 1. I'd like to test whether removing all instances of identical content in our article (whether that be by changing or shifting a few words around) improves its search ranking. I might have a go with Doha, attempting to bring the similarity down a notch, and see what happens in a few weeks.

Hopefully, as the founder of this expedition, I can return in a few weeks are provide a bit of focus! James A ▪ talk 05:13, 4 November 2013 (UTC)Reply

Interesting... I tried a compare on Kingston (Ontario) out of idle curiosity and got "3,080 matching words were found: (WV) 7,037 words total, 44% matched, (WT) 5,018 words total, 61% matched". Potsdam (New York) gets "153 matching words were found (WV) 850 words total, 18% matched (WT) 383 words total, 40% matched". Basically, we're adding good new content but stuff that's already there never seems to get updated, removed or rewritten. I wonder what the duplicate content penalty does with pages where the WT text is a mere subset of the WV text? K7L (talk) 17:22, 5 November 2013 (UTC)Reply

Sounds like a very useful test. Travel Doc James (talk · contribs · email) 12:39, 6 November 2013 (UTC)Reply

This edit took Merrickville from (522 matching words, WV 1538 words total 34% matched, WT 1070 words total 49% matched) to (263 matching words, WV 1530 words total 17% matched, WT 1070 words total 25% matched) even though the changes are purely cosmetic. A 2:1 difference, so that might have done as much or more than more useful (ttcf) changes like removing businesses that are out of business. The page was also linking to a WP article that didn't link back here, that's fixed now. Would it be worth watching this little village as a test to see if its search rank changes? K7L (talk) 14:09, 18 November 2013 (UTC)Reply

Links within WV do affect page rank and in many cases they are a good idea anyway, so I think we should make some effort to add them.

One case is major national/state/provincial parks or recreational areas like Lake Tai that serve nearby cities; the "Go next" sections of those city articles should have links. Another is major airports that serve a whole region; for example, if someone who lives Oxford goes to Heathrow for international flights then the Oxford article should link there. I have not checked if it does.

The Erie Canal article is mentioned on the page, so I used it as a test case. I hit "what links here" and the answer was not a whole lot. Even Albany at one end of the canal did not have a link, though Buffalo at the other did. I fixed Albany, but I think someone who knows the area might do a lot more, putting links into the articles for lots of towns along the route, perhaps in state or region articles. Rideau Canal or Silk Road or Route 66 might be candidates for similar treatment. Pashley (talk) 22:37, 11 November 2013 (UTC)Reply

Rideau Canal already seems to be linked from every town Kingston-Battersea-Westport-Smiths Falls-Merrickville-Ottawa on the route, from the parent regions Ottawa-Rideau and Seaway Region, grandparent region Eastern Ontario, great-grandparent Ontario, the list of itineraries, the UNESCO World Heritage List, one town bypassed by the route (Gananoque), from the Westport disambiguation, from Cruising on small craft ‎and from the Windsor-Quebec corridor. Other than the endpoints, the largest town on this route is population 10000, at the midpoint. Unless your intent is to start creating articles for places like Newboro (a tiny place near Westport) or Manotick (which technically is within the ridiculously-bloated Ottawa town line) I'm not sure what else can reasonable be linked for this.

Route 66 might be a different beast, as it was 2448 miles instead of 120 miles. Pontiac (Illinois) mentioned it but didn't link to it until a recent edit, for instance. Some of the other places, like Radiator Springs, have the link right in the lede. w:Category:Communities on U.S. Route 66 might be a good checklist? K7L (talk) 17:24, 3 December 2013 (UTC)Reply

It has produced mixed results as per below. Nothing definitive. No idea what we should do next if anything

• Cranbrook British Columbia.

Sept 30, 2013: Neither one is listed in top 100

Oct 8th Neither one is listed in top 100

Nov 14 WV is 30th WT is not in the top 100

Mar 9th, WV 14th WT 24th

• Guide Cranbrook British Columbia

Sept30: WT is 14th WV is 44th

Oct 8th, 2013 WT is 35th and WV is 25th

Nov 14th WT is 1st WV is not in top 100

Mar 9th, 2014 WV 3rd WT 12

• Fernie British Columbia

Oct 1st Neither in top 100

Oct 8th: Neither one in top 100

Nov 14: WV 48th, WT not in top 100

Mar 9th: WV 89th, WT not present

• Guide Fernie British Columbia,

Oct 1st, 2013: WT is 11 WV not in top 100

Oct 8th: Neither one in top 100

Nov 14th: WT is 25th WV not in top 100

Mar 9: WT 17th WV not in top 100

Travel Doc James (talk · contribs · email) 08:33, 14 November 2013 (UTC)Reply

Can this one be retired from its experimental status, returning it to its proper naming and giving it back its pagebanner and proper headings? It's been almost 3 months, and since WT doesn't even have an article for this destination anyway, I'm not clear on what it's trying to prove. Texugo (talk) 14:55, 2 December 2013 (UTC)Reply

Support from me. -- Ryan • (talk) • 15:43, 2 December 2013 (UTC)Reply

Might be worth checking with whomever created the page first. K7L (talk) 19:46, 2 December 2013 (UTC)Reply

This is one of W. Frank's effort to "prove" to everyone that he already knows the answer to our SEO woes. I don't think checking with him is possible or even really advisable. Texugo (talk) 19:49, 2 December 2013 (UTC)Reply

Go ahead Texugo. It seems he tried to deceive us. No? --Saqib (talk) 20:58, 2 December 2013 (UTC)Reply

No.

On the same day, 12 September 2013, Somerset, Tasmania was created from scratch here at Wikivoyage and "Somerset (Tasmania) was indeed created at Wikitravel. The fact that, a few weeks later IBobi was so embarrassed by the inferior Google results of Wikitravel that he both deleted the article and oversighted and protected it from recreation, does not invalidate the experimental result or turn Frank into a liar. Pashley is still an admin at Wikitravel and should be able to confirm the truth of my words.

Wikipedia assumes good faith; I don't think you should assume the reverse, Saqib, and I think you owe Frank a public apology. Whether he's able to read it or not is moot... - (unsigned) 04:17, 5 December 2013‎ 118.93.239.53

I believe I, too, remember this to be the chain of events. Isn't this documented somewhere in real time? Ikan Kekek (talk) 05:42, 5 December 2013 (UTC)Reply

Its funny stumbling into all this. I was the creator of the Somerset (Tasmania) article on WikiTravel. I had no idea my page was a part of an experiment. When IBobi deleted my article I questioned him about it and he resultingly banned my account, the reason given "Doppelgänger". I appealed the ban through multiple channels explaining that I wasn't a doppelgänger (maybe an ip clash) and that I'd done nothing wrong, but he refused to lift the ban or even talk to me. My 300 contributions to WT meant nothing. I was very annoyed with WT, which I had been using a lot in my travels, and searched for an alternative. That's when I discovered WV, found out about IB and converted to become a strong WV supporter. The rest of my work is still intact on WT, but every trace of the Somerset article is completely gone. I never let it go because I put a decent little bit of work into the Somerset article, hence how I got here (this page). So finally I've found the answers which have been gnawing at me for half a year. If it wasn't for you guys I'd still be using WT, so thanks for getting me banned, and thanks for resultingly getting me onto WV. Dukenemesis (talk) 08:20, 25 July 2014 (UTC)Reply

That's hilarious! I'm glad you're here. Ikan Kekek (talk) 08:34, 25 July 2014 (UTC)Reply

He did show that we are able to create articles that are top ranked on google from scratch. I consider this to be a glimmer of hope. We are not totally black listed. Travel Doc James (talk · contribs · email) 03:41, 5 December 2013 (UTC)Reply

Yes. I think the experiment was worthwhile and did show something. But what it couldn't show is what to do with preexisting articles or even new articles about places already covered on Wikitravel. Let's give the respect due to Frank for the experiment, and then we can go ahead and turn it into a "normal" article, while making sure its record of achievement is clearly noted for future reference. Ikan Kekek (talk) 07:17, 5 December 2013 (UTC)Reply

The article has already been changed back to meet standards. Incidentally, my point was not to claim that it was never created on WT, but that since it was then deleted, this test had long lost its point. As to what it proved, I don't think it proved anything more than what Travel Doc James said above, that we can create top-ranked articles from scratch, but it's actually not necessary to drop the banner and change our naming format and headers to prove that. See Engaru, Schaprode, or South Pelion for recent examples. Texugo (talk) 10:41, 5 December 2013 (UTC)Reply

We should continue testing to see what changes make a difference and what does not. It might be just a question of taking articles, rewording (or removing) everything that http://www.copyscape.com/compare.php identifies as duplicate text, then making sure the page is usable and has inbound links (local and interwiki). Merrickville was rewritten Nov 18 and currently google.ca has WT 1st, WV 5th with most of the intervening results leading to tripadvisor.

Merrickville travel guide - Wikitravel

Open source travel guide to Merrickville, featuring up-to-date information on attractions, hotels, restaurants, nightlife, travel tips and more. Free and reliable ...

Understand - ‎Get in - ‎Get around - ‎See

vs.

Merrickville – Travel guides at Wikivoyage

Merrickville is a village of 1200 people on the Rideau Canal in eastern Ontario. ... from the edge of urban Ottawa, it is a popular destination for summer day trips.

Not sure why Google is hyperlinking to individual section headers from the WT page but not for us? I'd be curious to know whether Google is treating constant bits of text (sidebars, CC-BY-SA copyright notice, section headings, various bits of the MediaWiki interface) any differently from the rest of the text when labelling pages as duplicate. K7L (talk) 18:02, 5 December 2013 (UTC)Reply

One thing that may be worth looking at is the HTML generated for our pages and that for WT. I don't know enough to go into this, but I did notice that WT has Meta keywords and description fields that we are missing. I don't think that the keywords matter, but the description is what appears as the snippet below the title in the search results - "Open source travel guide to...". This text is what attacks people to click on the link, and hence raise Google's ranking of the page. Can I encourage others to look at the HTML (right click - View Page Source in Firefox or View Source in IE). AlasdairW (talk) 23:46, 5 December 2013 (UTC)Reply

From an HTML standpoint the biggest thing I see that we could do is to get an <h1> tag with the destination title back - currently the banner code replaces it via Javascript, but putting it back in via Javascript opens us up to penalties for SEO manipulation, while putting it in via wikitext creates TOC issues. Beyond that, Mediawiki is mostly handling HTML and HTTP headers using best practices. A description would be helpful (as you've suggested) but we would probably gain more from a site map [25] than from trying to get useful descriptions on all pages. Overall, I remain of the opinion that there is no quick fix for the SEO problems, and that it's mostly a matter of continuing to slowly evolve our content and grow our incoming links so that Google no longer sees us as duplicate content and stops penalizing us in search results. -- Ryan • (talk) • 00:03, 6 December 2013 (UTC)Reply

To expand on my point above, it appears that WT uses the same description for all destination pages, with the destination name being a parameter, as everything else looks the same on different pages. Travel topics etc seem to use the start of the page as a description. AlasdairW (talk) 00:19, 6 December 2013 (UTC)Reply

It's fairly standard practice to add meta tags using a configuration string that is the same for all pages, which is what I'm sure they've done. But I don't think it will make much difference - most SEO guidelines these days de-emphasize anything done via meta tags. Our problem clearly remains the duplicate content issue - search for any destination plus the word "travel" and WT is almost certainly a top result, with us no where to be found. Search for the same destination plus the word "voyage" and the situation is exactly reversed. Until Google thinks there is a significant difference between the content of the two sites this seems likely to remain the status quo. -- Ryan • (talk) • 00:40, 6 December 2013 (UTC)Reply

If our article lede is our search engine description (displayed to the user) it might be worth rewording these ledes to something a bit more exciting than "X is a village in Y"? K7L (talk) 18:42, 6 December 2013 (UTC)Reply

If the Somerset guide at WT has already been removed a long time ago I don’t really think the WT article will show up in a Google search at all. So, yeah, now our article indeed has a higher Google rank than theirs. One rainy day in the summer I created a couple of articles for minor Finnish towns that do not exist, and I suppose have never existed, on the English language WT: Jokioinen, Ylivieska, Nivala and Kalajoki (not for SEO purposes but out of boredom). After I noticed this discussion I’ve searched for them using the search term “(town) travel guide” from Finnish IP’s, yesterday from an Estonian IP and now from my regular Finnish one using Google, Bing, Yahoo and Duckduckgo. To sum it up, what shows up seems to depend on the search engine’s temper of the day. Either the Wikivoyage article is among the top 3 results or then it isn’t among the top 50. Our Somerset, Tasmania travel guide is at the top when searching for it on Google and Duckduckgo but sixth among Yahoo results, WT’s guide being at number 4. On Bing our article isn’t among the top 30 hits - Travel’s Tasmania guide shows up at the fourth place here too. Another interesting thing: when Duckduckgo shows the icon of the side on the left hand side of the hit; they still use the old World Trade Organization-like logo for WV. ϒpsilon (talk) 21:56, 6 December 2013 (UTC)Reply

I just tested for two articles I have created here since the move, Lake Tai and Bactria.

Googling the names alone, WP is first for both, most of the rest of the first page is related stuff (reviews of restaurants named after the lake, or history & archeology) and we turn up on page two. Top of 2 for Bactria, down a bit for the lake.

Add "travel guide" to the search string & we are page one, 1st for Bactria and 2nd for the lake. Pashley (talk) 01:43, 21 February 2014 (UTC)Reply

While we mostly cling to search string of "[place name] guide" when we search our articles ranking. I got to know from Google Trends that Karachi travel and Karachi wiki are more popular search term as compare to Karachi guide. Mumbai travel and Mumbai wiki are more popular search term as compare to Mumbai guide. And I'm sure there'll be more popular search strings. --Saqib (talk) 16:44, 17 December 2013 (UTC)Reply

A quick research revealed that search string of [place name] travel is more popular search term as compare to [place name] guide for Dubai, Thailand, Bangkok, Tokyo and Singapore. --Saqib (talk) 16:55, 17 December 2013 (UTC)Reply

I've always used "place travel" as my search when I do authentic travel searches. For the experiment here where I talk about Okayama Prefecture, I tried it with both since I noticed others were using "guide" as their standard. ChubbyWimbus (talk) 05:33, 18 December 2013 (UTC)Reply

I believe "guide" was Frank's suggestion; I've never understood why he thought it was the most likely search term. LtPowers (talk) 18:12, 18 December 2013 (UTC)Reply

"guide" is unlikely to be the most common search term, but it's going to be extremely difficult for us to outrank on searches for "travel" since Google weights the URL so heavily and WT has "travel" in its domain name. On a similar note, searches that include "voyage" will probably always favor us, so searching using "guide" is a compromise where it is more likely that the content of the pages and the incoming links are the differentiating factors, rather than the URL. -- Ryan • (talk) • 18:17, 18 December 2013 (UTC)Reply

Both Mohenjo-daro travel and Mohenjodaro travel ranks us ahead of WT. --Saqib (talk) 18:24, 18 December 2013 (UTC)Reply

I suspect that may change after Google next re-indexes the WT page - the current Google index contains only a skeleton article [26], but the WT article has been significantly updated in the past week [27]. If the rankings don't change once the Google index updates then this article may be a useful test case. -- Ryan • (talk) • 18:41, 18 December 2013 (UTC)Reply

Incidentally, that diff you just posted betrays some unattributed copying from WV to WT. Look at the sleep listings, for instance... Texugo (talk) 18:50, 18 December 2013 (UTC)Reply

Texugo, this is not new. Previously, WT copied content from our Karachi article as well and didn't attributed. --Saqib (talk) 19:51, 18 December 2013 (UTC)Reply

I would suggest that whomever wrote the bulk of the WV article contact IBAlex about this and see what she says. (I have more faith in her than in IBobi.) Powers ^(talk) 01:02, 19 December 2013 (UTC)Reply

Ryan, I realised Google re-indexed the WT article and we're still beating them. --Saqib (talk) 09:09, 19 December 2013 (UTC)Reply

Interesting data point: Rochester (New York), our article, doesn't show up in the first 12 pages of Google results for "Rochester travel guide", but World66's article does. WT's is of course first. Changing the search term to "Rochester voyage" puts our article at the top. I think it's clear our URL alone puts us at a disadvantage right off the bat. Powers ^(talk) 18:05, 21 January 2014 (UTC)Reply

The movement had some issues with its page view counter recently. [28] While we have increased slightly since april it is not much. Travel Doc James (talk · contribs · email) 17:36, 19 December 2013 (UTC)Reply

Swept in from the pub

One way or another, I shan't be editing on this project for a while (Wikivoyage:User_ban_nominations#User:W._Frank]) so before I go, I'd like to draw your attention to Wikivoyage_talk:Search_Expedition#Action points to boost our readership. All the Best! --W. Frank^email_talk 12:49, 23 September 2013 (UTC)Reply

Thanks Frank. Travel Doc James (talk · contribs · email) 00:22, 1 October 2013 (UTC)Reply

I do think it is important to remove the hard-coded active hyperlinks to (the very censorious and dysfunctional) Wikitravel site on the majority of this site's pages if we ever want to get any visitors from search engines like Google. Most searchers don't go past the first two or three results you know.

Ikan Kekek suggested to me: "Unfortunately, unless Wikimedia Legal changes their minds about what kind of notification needs to be on the affected pages, there may not be much we can do about it." and I then asked him: "Where may I read this (mythical?) legal opinion, please? On the face of it, this (mythical?) legal opinion seems to fly in the face of the requirements of the cc-by-sa-3.0 license used by Wikitravel, Wikivoyage and Wikipedia unless this was a (secret) clause in the out of court settlement with Internet Brands. Do they realise that this (mythical?) strange decision dooms the English language Wikitravel to be inferior to Wikitravel in most search engine results forever (or at least until search engines value "mirror" sites as highly as "original" sites - which probably amounts to forever!). " to which he replied "I seriously don't think this is mythical. I understand that it was a part of the out-of-court settlement. I don't have the info at my fingertips right now and probably won't have time soon to research where you can read this verbatim. My suggestion is, if you can't find it in the Travellers' Pub, please post there and ask.".

Sorry if I've missed this elsewhere but this page is quite big and has already taken a while to skim. --118.93.47.31 23:26, 29 October 2013 (UTC)Reply

I don't know either, perhaps you should ask WM legal. Attribution is necessary, but we have the full history, so should not need to link to WT for history, and attribution is of the contributor, not the site the material was hosted on at the time, so a link to WT seems unnecessary. You may have a good point, and I think it is worth following up with someone who is better aware of the details of the licence. As I understand it, a link to the original site is a substitute for the full history, for uses where providing the full history is inconvenient or impracticable. In the history, all contributors are identified by username on WV, or IP address, which is as full an attribution as you can reasonably hope for. Users who did not migrate to WV are identified by their WT usernames and have an imported user page. I think this exceeds the minimum requirements of CC-by-sa licencing, but I am not an expert. I would also like to see that link removed. Probably all of us would... • • • Peter (Southwood) ^(talk): 07:37, 2 November 2013 (UTC)Reply

@Peter (Southwood) - according to my reading of what Ikan said above, there's a sharp distinction to be drawn between a) what's legally required in general for any hypothetical situation where CC-BY-SA is involved, and b) what Wikivoyage in particular is required to do per the terms of its out-of-court settlement with Internet Brands, which Ikan seems to be saying is more stringent. If that is the case - and absent an official clarification from WMF Legal, I wouldn't take anyone's word as holy writ - we're hamstrung.

If what Ikan says is true, that begs the question of why the WMF shelled out hundreds of thousands if not millions of dollars of donors' money in legal fees only to sell us out and all-but-ensure they'd never recoup their investment, but then again, I'm neither a businessman nor a legal expert so what do I know? -- AndreCarrotflower (talk) 08:17, 2 November 2013 (UTC)Reply

The question is really down to who has the authority to remove the WT link. Every conversation that I've seen points to the WMF legal department. People giving opinions about it on various pages on WV isn't going to effect that at all. The user PriceGloria said that they would contact this department directly which is probably the best proactive step to take. Andrewssi2 (talk) 08:32, 2 November 2013 (UTC)Reply

I have no special knowledge. I guess it's more accurate to say that WMF Legal refuses to give advice to anyone but the Foundation, and leaves the rest of us to edit at our own risk. See meta:Wikivoyage/Lounge#Google PageRank issues:

"[W]e are ethically obligated to represent only the Foundation. We cannot unfortunately give legal advice to the community at large, particularly in cases like this one that may be complex and where we may not know all that facts. With that in mind, we'd urge you to carefully consider any changes you make, on this issue or any other - you should avoid placing yourself in situations beyond your individual tolerance for legal risk. I have written about the responsibility of contributors to the Wikimedia projects elsewhere, which you may wish to review."

The linked legal advice is:

"An editor is legally responsible for his or her edits and contributions on Wikimedia Projects. The Projects are only hosting venues: the Wikimedia Foundation generally does not edit, contribute to, or monitor the content on the site. For that reason, the Wikimedia Foundation is not responsible for the edits or contributions of the editors.[2] See 47 U.S.C. 230

Editors should be advised to exercise caution and avoid contributing any content that may result in criminal or civil liability, including infringing material, defamatory statements, and privacy violations.

The Wikimedia Foundation strongly supports free speech throughout the world. The Wikimedia Foundation accordingly does not agree with many countries' restrictive laws on content. That said, local courts may try to impose local norms. For that reason, the Wikimedia Foundation should advise editors and administrators who do not comply with their local laws that they do so at their own risk."

So the problem here is that if we decide to do away with the links to WT on scads of articles and get individually sued, we have no indication of whether the WMF would represent us or leave us twisting in the wind. Ikan Kekek (talk) 08:33, 2 November 2013 (UTC)Reply

It shouldn't be utterly impossible to come to some sort of determination of whether hyperlinks are required per the terms of CC-BY-SA 3.0, even without the aid of WMF Legal (perhaps the folks at Creative Commons could shed some light on it?)

As far as I can see - and again, I'm no legal expert; I'm just brainstorming here - the only missing puzzle piece is the additional terms that may or may not be contained in the settlement between IB and the WMF. I'm assuming that those documents are sealed and inaccessible to garden-variety editors. That being the case, would individual editors be legally responsible if they violated the terms of an out-of-court settlement the details of which they were not privy to? Wouldn't the culpability for such a thing fall back on the WMF in that case? The WMF can disclaim responsibility for individual editors' contributions till the cows come home, but if said editors were violating not the law itself but an agreement between two third parties that they had no part in, I imagine that might be a whole other issue.

-- AndreCarrotflower (talk) 08:45, 2 November 2013 (UTC)Reply

I don't know, and without legal advice, I don't think any of us can know. If there are any good civil lawyers who happen to be Wikivoyagers, perhaps we could get some advice on law from them, but it would really be much more reassuring to get advice from WMF Legal, and they have so far refused to give any such advice, claiming that they are ethically obligated to refrain from so doing. It seems like a Catch 22. Ikan Kekek (talk) 08:57, 2 November 2013 (UTC)Reply

Exactly. We clearly need legal expertise, but it seems like it needn't be WMF Legal in particular. I repeat that Creative Commons might be the next people to ask. -- AndreCarrotflower (talk) 09:01, 2 November 2013 (UTC)Reply

I don't think CC attribution is the issue. I think the issue is what's in the settlement. But I wish WMF Legal could at least tell us unambiguously that we cannot get rid of the notices, if that's true. Ikan Kekek (talk) 09:04, 2 November 2013 (UTC)Reply

The results of the suit WMF brought against IB are public, and the only issue for WV was that we need to include Template:WMFLEGAL Disclaimer in a prominent location for three years - see Template talk:WMFLEGAL Disclaimer. That said, while it seems pretty clear from a reading of the license and from standard practice that attribution on the history page should be sufficient, AND since "WT" is not an author of any of the content in question (the contributors are the ones who hold copyright on their contributions), any editor who makes a change to the attribution template would be responsible for that change and thus become a target of any legal action IB chose to take, hence the desire to get WMF legal to vet any change and (ideally) for someone with WMF to actually make the change. -- Ryan • (talk) • 15:57, 2 November 2013 (UTC)Reply

If I'm reading that talk thread correctly, the legal disclaimer needs to appear on only one policy page. But then why do we have an attribution template on so many pages, if the legal settlement didn't require it? I think something's missing here. Ikan Kekek (talk) 19:19, 2 November 2013 (UTC)Reply

The footer attribution has been there since Wikivoyage launched, preceding the conclusion of the WMF's suit against IB. I assume that the text used was meant to try to be overly cautious in order to prevent further legal complications, since IB had sued two individuals prior to the site launch. -- Ryan • (talk) • 19:28, 2 November 2013 (UTC)Reply

See also Wikivoyage talk:Main Page Old#Old revisions, which is the last time legal got involved over the footer link. -- Ryan • (talk) • 19:36, 2 November 2013 (UTC)Reply

Well, that's as clear as mud to me. Before an edit conflict, I was going to post this: So there is nothing in the legal settlement requiring the footer? If that's the case, is the only reason that the footer has been maintained a lack of confidence that the WMF would support anyone who's sued as an individual for getting rid of it? Ikan Kekek (talk) 19:42, 2 November 2013 (UTC)Reply

Does the requirement to display the WT link in a 'prominent location' for three years mean that there is no issue whatsoever in having it removed in (I guess) 2015? --Andrewssi2 (talk) 10:55, 3 November 2013 (UTC)Reply

That seems to be a logical deduction. I guess the 3 years started from the date of the settlement. Ikan Kekek (talk) 11:00, 3 November 2013 (UTC)Reply

My guess is that the footer was added by someone at WV:de during the initial migration, as a super-conservative approach, and now we can't remember why it was done. Maybe User:Hansm or User:RolandUnger would remember. I also guess that it is not required by the settlement terms, and if no-one can show that it is required by the settlement terms, I would be willing to remove it once I find out how to do it. • • • Peter (Southwood) ^(talk): 16:13, 3 November 2013 (UTC)Reply

The "prominent location for three years" applies only to Template:WMFLEGAL Disclaimer, which does not mention WT by name, let alone link to it, and has nothing to do with the footer. The issue with the footer is not time-bound and is entirely to do with not breaching the license, which I presume is CC-BY-SA 3.0. Forget about the settlement of the suit and focus on interpretation of the license. If it says we need a hyperlink to the original, we need a hyperlink. If it doesn't say that, but says we need to display the URL, then perhaps an unlinked URL will suffice. An unlinked URL would be better than a hyperlinked URL. Nurg (talk) 20:23, 3 November 2013 (UTC)Reply

┌─────────────────────────────────┘

When we started Wikivoyage in 2006 we decided to have a conservative solution to fulfill the regimentation of the CC-by-sa license to name both the author's names and the source of the derivative work (at that time Cc-by-sa 1.0 was the current license) because we had no war chest for a lawsuit. Within some years the German Wikivoyage outnumbers WT. -- At the time of the WMF migration two association members including Hansm were in San Francisco at the WMF. At this time both WMF and our team decided to continue the procedure of mentioning of the source work for the derivative work on legal grounds. About one and a half year ago I emphasized this problem in discussions with James and other admins and named it a huge ad campaign for IB and made several proposals like starting from scratch and to delete all articles without content. But this problem was ignored by the community. One year ago at SF Hansm and DerFussi repeatedly proposed to kill all meaningless content including talk pages. But our proposals were ignored by the community again. As I said the decision to keep the footer was supported by the WMF's legal department for the same reason. Of course I repeat: what should we do furthermore in that case of the appeasement or naïvety of the English community? --RolandUnger (talk) 20:12, 3 November 2013 (UTC)Reply

Oh, we seem to be behind. Purge any page and see. Kudo's for User:Jmh649 :-) JuliasTravels (talk) 21:09, 3 November 2013 (UTC)Reply

Now I've read more about the free advertising for Internet Brands' Wikitravel website on almost every page of this English language site, it seems that the "Germans" (Frank, Hans, Roland, etc) have been right about this all along. I think it's great that Peter Southwood has volunteered to remove the hyperlinked, clickable text on most of our pages and we should educate him this very week so that he can get the job done and start welcoming a tonne of new readers. Now if we can implement the other suggestions about units, section headers, and lead paragraphs, etc, that should be enoguh of a sudden sea change for Google to drop the site duplicate penalty and for this site to finally take its rightful place in the search rankings. Now don't lose this momentum by more anally retentive introspection or personality politics - go for it! --118.93.67.66 21:51, 3 November 2013 (UTC)Reply

Oh, I was dreaming. There's one last hyperlink to deal with. What seems important to me from Ryan's link, is that WMF legal was already involved in this matter. It was just unfinished, it seems?. Their argument for not engaging is not quite as convincing then: it seems that Philippe intended to get back to them on it but nothing came out of it since, right? Can we ping him again and ask if he can perhaps persuade them to dedicate a little bit of their time to it once more? I imagine the main reason they don't really want to (at Legal) is that they're swamped already wit other things. @Roland, I'm sorry you're frustrated with our community. I missed that conversation back then but let it please you that indeed all empty outlines have been deleted since and others are being recreated to lose the attribution. JuliasTravels (talk) 22:37, 3 November 2013 (UTC)Reply

I, too, am sorry that the German Wikivoyage folks who were so helpful to the English-speaking community were dissed this way. Ikan Kekek (talk) 22:42, 3 November 2013 (UTC)Reply

I'm sorry, but dissed? It may be Roland's opinion that we "ignored" certain issues, but if so, I doubt it was intentional and it certainly doesn't rise to the level of active disrespect. LtPowers (talk) 02:27, 4 November 2013 (UTC)Reply

It looks like they felt dissed. Ikan Kekek (talk) 05:08, 4 November 2013 (UTC)Reply

Then you may want to look at Wikidata:Wikidata:Requests_for_comment/VCards_for_Wikivoyage, because they are feeling dissed or ignored there too. -- torty3 (talk) 13:21, 4 November 2013 (UTC)Reply

I was totally unaware of this. What's the best way for us to be aware of these initiatives? Ikan Kekek (talk) 19:04, 4 November 2013 (UTC)Reply

The best way is to read Summit and to participate in it. As far as I remember, it was one of the Summits, where Stefan linked to this Wikidata proposal. --Alexander (talk) 20:41, 4 November 2013 (UTC)Reply

I don't see anything at that Wikidata discussion that indicates disappointment in the response to the proposal. I do see several legitimate criticisms raised and not much attempt to address them. LtPowers (talk) 22:14, 4 November 2013 (UTC)Reply

Have summarized the attribution to WT. Here I have proposed we move the last link to WT from the mainspace to the history tab. [29] All we need is a bot. Any bot makers? We simply do it and if legal has issues they will revert. They will not give us approval ahead of time. Travel Doc James (talk · contribs · email) 12:38, 6 November 2013 (UTC)Reply

What's the summary here? Are we go? --Inas (talk) 00:27, 18 November 2013 (UTC)Reply

Swept in from the pub

Here is an idea on how to get more visitors/contributors.

• Start with a list of our ten most visited articles. Let's say they are "City 1", "City 2" and "City 3" etc.
• We can then assume they are also amongst the most common searches for travel guides on google. So "City 1 travel guide" is probably one of the most popular searches on google, much more common then "Any_other_city travel guide".
• We should now make sure these ten articles makes it to the top in a google search. If "City 1 travel guide" has wikivoyage as the top result, that will make a greater impact on the number of visitors than if we are the top result for "City 374 travel guide".
• Then we also make sure these ten articles are of the best quality possible, make them star articles. This means most of the new visitors will find wikivoyage a useful travel guide and recommend us to others. Link to wikivoyage in blogs, forums etc.
• I am sure this will make a difference. We will get more visitors and more contributors, and hopefully it will make a snowball effect.

Traveler 712 (talk) 11:25, 23 October 2013 (UTC)Reply

For one related list, see Wikivoyage:World_cities/Large. For discussion of search issues, see Wikivoyage:Search_Expedition and the talk page for that article.

Currently the title text is "City1 - Travel guides at Wikivoyage". Would it make any difference to change it to "City1 travel guide ..."? How should we handle phrasebooks, itineraries and travel topics if we make that change? Pashley (talk) 11:40, 23 October 2013 (UTC)Reply

It seems to me that the competition for the top-searched destinations is fierce - there are thousands and thousands articles, guides, webpages et al. about Paris or New York. Our forte seems to be that we also do cover Menzies, Hilversum and thousands of other less popular destinations, some of them really well. Some day Wikivoyage will be the default search result for all travel-related searched, much as Wikipedia is for almost everything, but until then what can win us viewership and recognition is catering to travellers who really can't find much info elsewhere - and their searches will probably also be ones we have much less competition for, plus it will be easier to be formally (i.e. via an actual hyperlink) referred to by other sites, blogs et alia. PrinceGloria (talk) 13:16, 23 October 2013 (UTC)Reply

I suppose it depends where in New York... filling some gaping holes in places like Oswego, Ogdensburg and Massena may get us traffic with less competition than if we were vying for the top spot in search results for New York City - which many sites already cover well. K7L (talk) 14:52, 23 October 2013 (UTC)Reply

If there are people interested in working on top search cities like Paris, New York, Tokyo (which really could use it!), etc. then by all means plunge forward! To me, it's more difficult to start with pages that are already good. For example, Chicago was a star long ago, so it would take a lot of thought to change the guide enough to make it show up on top of searches. On the other hand, cities that are a mess and have a lot of room for improvement, like Tokyo, Sao Paulo, etc. and pages that have little or no content would be much easier and probably more fruitful to focus on. Anywhere in Africa would give us a strong and quick boost I imagine, as well, since our coverage continent-wide is rather weak. ChubbyWimbus (talk) 01:51, 24 October 2013 (UTC)Reply

By the way, there's a list of most visited articles here –sumone10154^(talk) 03:45, 29 October 2013 (UTC)Reply

Nice. Looks like our top five cities are London, Paris, L.A., Singapore, and ... Sedona?! LtPowers (talk) 21:51, 30 October 2013 (UTC)Reply

It looks like that statistics page stopped being updated in April. Texugo (talk) 22:01, 30 October 2013 (UTC)Reply

What is more relevant is our atrocious search engine rankings.

If you go to https://startpage.com
(which features results from Google, but anonymised so that the results are not skewed by personalisation, browsing history, geographical location, etc)
and key in a typical search term for any of these "top destinations" (such as "guide london" or "guide paris" or "guide L.A." or "guide singapore" or "guide sedona"),
in every case you will find our "parent" site Wikitravel listed above us.
In most cases we do not even appear on the first page of results!

Why oh why are none of the movers and shakers interested in adopting the action plan proposed here?

Is it because, in a very few weeks, they would then have to cope with a flood of spambots or is it because some name-callers would then have to eat humble pie? Surely this scandal is going to cause some waves with the WMF if their investment remains so obstinately obscure? --118.93.47.31 23:09, 30 October 2013 (UTC)Reply

1) Frank, please log in when you edit. Using an IP makes it look like you're hiding something.

2) You continue to unfairly impugn the motives of large segments of the editing community. This will stop immediately. LtPowers (talk) 00:09, 31 October 2013 (UTC)Reply

Please let's not have a discussion about identities here in the pub. User talk:118.93.47.31 is a better place for it, even if disregarding that 118.93.47.31 is now blocked for one day and cannot respond from that IP in the pub. thanks. Nurg (talk) 00:38, 31 October 2013 (UTC)Reply

I have tried this on one article and while it worked briefly it than stopped working. If we can get this to work for a long established article and work for a significant period of time you will have my complete support to institute it all across WV. Travel Doc James (talk · contribs · email) 09:04, 2 November 2013 (UTC)Reply

Swept in from the pub

I see there is a multitude of discussions everywhere - here, at Expedition talk pages and elsewhere, and I believe a majority of them are not just W.Frank's ASBO parades. I have a limited amount of time for WV, and I am just a humble editor, so I admit it is beyond me to follow all of them and make sense of their tangents.

I am seeing our Alexa positioning and editor count essentially flatlining (ok, let's call it "levelling"), so I was wondering if I should treat it as a sign of our dismay at a lack of solution to move WV from where it is to where it belongs (i.e. as the no. 1 go-to site for travel guide information), or just a calm before a storm which will elevate WV one level ahead again, thanks to an ingenious solution that is brewing or just small concerted efforts of whose combined significance we are unaware?

So, what is the scoop? What plan do the WV boffins have? Thanks for letting us casual editors know! PrinceGloria (talk) 06:06, 31 October 2013 (UTC)Reply

I'm not aware of anything that isn't already on Wikivoyage:Search Expedition, but there may be actions underway at WMF or by others (see Bugzilla: 52688). There have been a number of proposals made for improvement, but most of the low-hanging fruit (links from WP, getting the word "travel" on pages, in titles, and in incoming links) has already been taken care of. The value of some of the remaining proposals aren't entirely clear - changing headings would signal to Google that one of the most important page structures has changed, but it isn't clear how much effect that will have on rankings and would require agreement on new headings, which is contentious. Removing or modifying the WT footers is something everyone wants to do, but given past history that isn't something that seems plausible without a legal OK. Changing URLs from "city" to "city, state" has been suggested, but that's another big change whose value isn't clear. Given the magnitude of those changes, unless one or more individuals take on the task of providing some measurable way of determining their value it may be tough to push them through - I don't think anyone wants to go to the effort of finding consensus for new headers, changing every article on the site, and then fail to see any SEO benefit.

In lieu of all of the above changes, one thing that will clearly help us is continued evolution over time. Having pages that have been in Google's index for a longer time helps, having more incoming links help, and getting additional contributions that make our articles different from WT will help. If any further proposals are made that can be implemented without major site changes then those would probably be implemented immediately, but at least for the moment my impression is that we're mostly just working on an article-by-article basis to try to make sure our content evolves and is distinct from WT, thus signalling to Google that this site isn't just a clone. -- Ryan • (talk) • 07:04, 31 October 2013 (UTC)Reply

Thanks for the roundup, so I am right in believing we are not actually DOING anything at the moment, we now have many proposals and are unsure if they will work and if anybody has the willingness to put the effort in those. Did anybody try to do a reasonably convincing test on any of those - even including stuff mentioned in your second paragraph (they all probably do work with Google, but how effective are they vs. other stuff)? Is anybody in charge with getting the OK from legal to stop linking to WT? PrinceGloria (talk) 07:08, 31 October 2013 (UTC)Reply

WMF Legal stated that they can't provide advice on the footer issue: meta:Wikivoyage/Lounge#Google PageRank issues. You would need to read through Wikivoyage talk:Search Expedition for updates on your other questions. -- Ryan • (talk) • 07:14, 31 October 2013 (UTC)Reply

Hi PrinceGloria, I think it is great to want to make actions to improve the rankings of WV however my observation is that this is task is not straightforward and actually discussion is causing friction within the WV community. My view is that as 'casual editors' we can make a real difference for now by just improving content, and even by creating new articles that have no links back to WT. It is frustrating to wait, however I believe that by playing the 'long game' the site will eventually solve its ranking issues. Andrewssi2 (talk) 07:42, 31 October 2013 (UTC)Reply

In other words, forget it. If we want to change the footer, we do so at our own legal risk, which no one wants to take. If we want to make major changes to section headings, we need consensus, at the exact time when our lack of ability to form consensus anymore is frustrating longtime editors off the project. The best way to effect change is to continue to evolve the site, yet the process of evolution is glacially slow and getting slower as editors leave. -- AndreCarrotflower (talk) 07:46, 31 October 2013 (UTC)Reply

It's only "causing friction" because some admins are too thin-skinned and precious. All that counts is what works - not who is for it or agin it. Frank's told you exactly and precisely what needs to be done and more fool you if you don't suck it and see. As other IP's have said (and been censored for saying it) Wikivoyage has nothing to lose collectively and everything to gain. If it doesn't work within a matter of weeks, all you have to do is go back to your old "dead museum curator" mentality when it comes to the sacred lead format and sub-section titles.

Don't be so defeatist, AndreCarrotflower. The resultant ten=fold boost to our readership will boost our editor ranks too - and don't forget that some of the players that "took their ball off the pitch" have already returned with brand, shiny new accounts. What does " W.Frank's ASBO parades" mean, PrinceGloria? Would you translate, please? --203.173.196.38 07:57, 31 October 2013 (UTC)Reply

I plan on writing an essay about my thoughts regarding the future of Wikivoyage... whenever I find the time, things have been crazy the last few weeks :/ In the meantime, I'm still around, just keeping quiet :)

I am kind of disappointed by WMF Legal's response (among other things) but oh well, we gotta make the best of what we have. --Rschen7754 08:07, 31 October 2013 (UTC)Reply

Yes, the legal conversation has to be directed towards the lawyers and discussing it here between ourselves will only cause unnecessary grief. Keeping the WT link was the cost of forking the project, and only the layers have the means to correct that, if they can. Andrewssi2 (talk) 08:33, 31 October 2013 (UTC)Reply

As a total sidenote, can we pause for a brief moment and consider banning edits by unregistered users altogether since we seem to have so few editors anyway? It is a unique feature of MediaWiki projects among all other "social" and "crowdsourced" media around. I'll get round to formally suggesting that sometime later, just some food for thought now. PrinceGloria (talk) 08:19, 31 October 2013 (UTC)Reply

I myself was an unregistered editor for 2 years or more. Many good edits are made by unregistered users. If you're suggesting we should require people to register before they edit, I'd suggest that's exactly the wrong thing to do if we want to be user-friendly and welcome more editors. Ikan Kekek (talk) 08:29, 31 October 2013 (UTC)Reply

The WMF would never allow that; they've rejected lesser measures on other wikis before. --Rschen7754 08:35, 31 October 2013 (UTC)Reply

And rightly so. What would the slogan be then? "The travel guide that anyone can edit...once you've registered, been fingerprinted, and had a blood sample taken"? :-) Ikan Kekek (talk) 08:47, 31 October 2013 (UTC)Reply

For the record: while I think it goes without saying that Frank is the author of the IP edits in question, which is troublesome given the accusations of sockpuppetry that have dogged him since as far back as I remember, that doesn't negate the fact that the proposals he has put forth are interesting and, I think, worthy of far, far more consideration than has been given to them thus far. -- AndreCarrotflower (talk) 08:37, 31 October 2013 (UTC)Reply

I have always respected Frank's proposals. My only objection to Frank, ever, has been his sometimes off-putting delivery. Ikan Kekek (talk) 08:47, 31 October 2013 (UTC)Reply

[Back on topic] We sure should all do our best and focus on the core of Wikivoyage, i.e. editing, improving, expanding and creating articles. And I am quite sure everybody does, as much as time allows. We are really changing this place around massively, it looks nothing like the garbage dump we inherited from "big brother" not so long ago.
HAVING SAID WHICH, I believe we need to push forward harder. I believe we need to be proactive and, being a relatively small and cohesive group of editors still, act in unison. I do not consider what was linked to above as a formal answer from WMF's legal team, and I will pursue one in clear writing, preferably with a stamp, signature and on company paper. This is a WMF site, and we need help from WMF. This way or another. And we need clear answers. I urge you to continue pursuing those regarding the active hyperlinks to Wikitravel for until a solution is found or the inability to do anything about them is made clear to us in a way that leaves no other options. I believe we are far from that.
I also believe we could make a list of proposed "search engine optimization" actions and start formalized experiments e.g. on a test group of 10 articles each, over a predetermined time and under predetermined circumstances. If we need some tech help from WMF to better frame those tests and their circumstances, let's ask for it. If we believe we can do that ourselves, let us just start a subpage at the Search Expedition, list and schedule the tests and put timely updates on results there.
Any other actions, apart from just editing the articles, trying to remove WT links and "search engine optimization" that have been suggested and I missed? PrinceGloria (talk) 08:19, 31 October 2013 (UTC)Reply

make a list of proposed "search engine optimization" actions and start formalized experiments

Yes, this is what we should do. I don't think anyone has a problem with testing the proposed changes individually to find out what works. All I had a problem with was the assertion that all the proposed changes should be implemented immediately without testing the individual effectiveness of each. Texugo (talk) 10:11, 31 October 2013 (UTC)Reply

Yes, a more systematic approach is much needed, one new article with no pre-fork counterpart and absolutely no duplicate content is not representative of 99% of all articles, and this is not even considering the raft of variables involved. Anyone and I really mean anyone with the time and inclination, could step up to help organise it. It isn't in an admin's remit alone (what I realised with the dynamic maps, for better or worse). I'd like to see less stark lines drawn.

One important thing is that Hazard-bot at WP still has not added links to some pages, I'm not sure whether it's because of the placement of templates? Otherwise, Wikidata has superseded the need to compare data dumps, and someone could go through the articles with AWB like how Texugo did.

Another angle to pursue more aggressively is social media like Twitter, Google+ and Reddit. Places like Manchester still have WT links in their sidebar, and some Reddit sub-communities have gladly changed them over to WV - they just like Wikipedia a lot. This comes pretty close to stealth marketing though.

For good news, regardless of all the over-obsession with WT, did you know that the number of very active contributors have been higher than any point of history? Pages views have also been growing 20% a month. -- torty3 (talk) 10:54, 31 October 2013 (UTC)Reply

Thanks torty3. This was a point bothering me today during this conversation which was although WV obviously has problems in many areas (social, technical, whatever), the community has actually created a far superior 'product' to WT in a very short period of time. 20% page view growth is great, and even if resolving some issues is moving at a snail's pace we are still nevertheless going to get there. Andrewssi2 (talk) 11:15, 31 October 2013 (UTC)Reply

Thanks for your thoughts torty3 and Andrewssi2!

• I've just contacted the mods of that Reddit page to ask if they'd be prepared to change the link.
• Two weeks ago, I changed (just) the headings in our Wigan article to see what effect that would have on its own. From what I can discern, it hasn't had a great SEO impact as yet, but I suppose that could change.
• Twitter followers have been rising rapidly of late and, as I haven't had quite as much time as I used to, I've shared the credentials with the WMF's social media team who are also posting on our behalf. It would, however, be good to push social media as a whole a bit harder.

It's great to read that things are improving - I hope that by the time the site's 1st WMF birthday comes, we'll have pushed our way to the top (or near the top) of the pack. --Nick ^talk 14:06, 31 October 2013 (UTC)Reply

If you don't mind, when you do tests like that at Wigan, please formalize them somewhere at the Search Expedition so people know what's going on. Texugo (talk) 14:10, 31 October 2013 (UTC)Reply

I did mention it at the time in Wikivoyage:Travellers'_pub#Boost_our_readership_by_at_least_400.25_in_two_months, but I'm not sure anyone noticed. I'll post about it in the Search Expedition now. --Nick ^talk 14:16, 31 October 2013 (UTC)Reply

Yeah, we actually need to go through all the discussions so far and consolidate everything to make a table to track all the experiments with the titles, dates, and what was changed. Texugo (talk) 14:23, 31 October 2013 (UTC)Reply

(indent) Having an attractive homepage would certainly help, too. Right now we still have the white captions that are mostly unreadable on the pictures. Having a sloppy-looking homepage is not doing us any favors, especially since the possibility of having a front page feature is supposed to be a way to encourage users to improve articles. Right now, the incentive is rather low in my opinion, because the features are not presented well. I've lost track of the discussions (if any are still occurring) because I myself can't do that sort of thing, but I have to admit that for me, the current front page actually makes me NOT want to make a page featurable. If I work hard enough to get an article up to standard, I wouldn't want it featured as an image with illegible writing over it. ChubbyWimbus (talk) 16:07, 31 October 2013 (UTC)Reply

I don't have as much trouble with the homepage as you do, but it does sound like both of us would like for there to be the option of black text in the captions on the images for the featured articles. How hard is that to do? It doesn't seem like it should be hard. Ikan Kekek (talk) 17:44, 31 October 2013 (UTC)Reply

This seems like a thread that could easily go in a lot of directions and become hard to follow, so would it be OK to move discussions about the main page to Talk:Main Page? In particular, Talk:Main Page#Main page readability is a discussion I started a long time ago but haven't had time to follow-up on. -- Ryan • (talk) • 17:56, 31 October 2013 (UTC)Reply

Manchester on Reddit now links to the WV article in its sidebar - yay! As for the Main Page, I'm sure it would be easy to add a black text option, although Mark would know more about that. --Nick ^talk 00:23, 1 November 2013 (UTC)Reply

Well done Nick! I believe that such tireless inbound-linking, allied with content improvement, is the most efficient way to improve rankings. This can be done by anyone at the local level. Share links about a recent trip on Twitter, blog about a place you've been, embed links, add POIs :-) Nicolas1981 (talk) 06:55, 1 November 2013 (UTC)Reply

Google hasn't been updating it's PageRank service since early 2013. Google's Matt Cutts had explained that Google is relying less on publishing its PageRank calculations because web masters rely too heavily on PageRank rather than developing a quality website with quality content. However, it appears that Google has finally updated its PageRank service. My site has a PR2 and now I see most of Wikivoyage with a PR6, which puts at a tie with the red-headed step child Wikitravel. -- Sapphire (talk) 17:31, 7 December 2013 (UTC)Reply

Swept in from the pub

There are a few items currently open on VfD requesting an existing WT-import page be deleted (Tsushima, Tsushima (disambiguation), Pusan, Pyeongchang) so that it may be replaced by a new page (with no WT text and no WT attribution footer) on the same topic.

This looks like a good idea which should be encouraged for SEO reasons, so that we're not just an updated copy of some other travel wiki, but there should be an easier way to do this without going through VfD each time. I propose:

• Users should be able to create a new version of placename at placename/new (so Elbonia would be rewritten as Elbonia/new)
• The page would start as an empty skeleton (such as {{smallcity}})
• Any useful contributions made after the WT/WV split could be preserved by taking one user's text at a time and inserting it into the new page with that contributor's name cited in the edit summary ("Importing contributions from User:Soandso...")
• The imported WT text would be discarded in its entirety and new text written at placename/new
• The old article would be tagged for speedy deletion, maybe by creating a template {{rewritten}} "This page has been rewritten, the new version is at [[{{PAGENAME}}/new]]. Administrators, please delete this page and move the new one into its place".

Would this be a worthwhile addition to our deletion policy? K7L (talk) 18:21, 2 January 2014 (UTC)Reply

Sounds like a very reasonable solution for nearly empty articles or articles nearly empty at the time of migration, if any editor can prepare the recreation. I'd happily support that! Danapit (talk) 20:23, 2 January 2014 (UTC)Reply

Count me in. I can get myself involved in the recreation task. --Saqib (talk) 21:09, 2 January 2014 (UTC)Reply

• Before we jump right into this, we should decide how long is too long for an article to be if are going to delete and re-create it. Doing it for empty or nearly empty articles is fine, but I question the practical value of spending time paraphrasing huge chunks of text solely to remove the attribution (especially given that our SEO problem is a long way toward being solved, given this information on our Google PageRank). Some of the articles linked to at VfD seem to be approaching that threshold. -- AndreCarrotflower (talk) 23:38, 2 January 2014 (UTC)Reply

I saw the deletion of skeleton and redirect articles with WT attribution as a 'low cost, quick win' activity. I wasn't thinking to rewrite any articles that had any significant WT content and in that respect I am in agreement with AndreCarrotflower. It would however be good to have an easy mechanism, as Danapit suggests, to delete and recreate articles that are either skeleton or near-skeleton. Andrewssi2 (talk) 00:24, 3 January 2014 (UTC)Reply

There is also an ethical & legal issue here. Deleting an essentially contentless article — just an empty template with no more text than "Left Overshoe is in in Manitoba" — seems obviously OK to me but I'm not sure about anything beyond that, not even a simple redirect since that does contain information. If we end up paraphrasing text it seems to me we must then keep the attribution. We need to be very careful about following CC license terms, including attribution, for several reasons. It is the ethical thing to do, we want others to do that with our text, and WT are known to be litiginous even when they have no case

However, I am not a lawyer and I may be wrong about any of the above. Perhaps we should ask WMF legal for advice here? Pashley (talk) 01:03, 3 January 2014 (UTC)Reply

I don't want to get lost in legal arguments. The rationale is to have a mechanism to delete skeleton articles and redirects. As long as no original text is copy and pasted I really do not think there is a problem. Andrewssi2 (talk) 01:11, 3 January 2014 (UTC)Reply

If "imported WT text would be discarded in its entirety" there's nothing to attribute, regardless of length before or after. There's no reason why our article on X should even use WT's article on X as a source of information if we can get the same facts elsewhere. K7L (talk) 03:55, 3 January 2014 (UTC)Reply

As we start 2014 it has become apparent that many of the discussions in articles that took place originally on WT (2012 and before) have often been archived.

On Talk:Iran we actually removed the WT attribution from the talk page because of this.

My question is whether this action will have any impact to our search rankings? If so then there are many other talk pages that could benefit from this as well. Andrewssi2 (talk) 13:16, 9 February 2014 (UTC)Reply

I'd say it should be a general policy to (make that: "Where possible, we should") archive all talk page discussions from before the move, of course retaining WT attribution there, and create new pages for post-move discussions, of course without WT attribution. Pashley (talk) 15:01, 9 February 2014 (UTC)Reply

Some of those discussions are still active, and others are still being referred to in other discussions. Making that a general policy could complicate things significantly. (Archiving is bad enough when it breaks internal links.) Powers ^(talk) 16:32, 9 February 2014 (UTC)Reply

Links can be fixed, though I admit it is often a major hassle. Exceptions might also be made where required.

As a general policy, though, I think this would be worth doing. Pre-move discussion is all now well over a year old; I do not think all that much of it is still relevant. Pashley (talk) 16:48, 9 February 2014 (UTC)Reply

On the other hand, checking a few articles I know well, I find some unexpected difficulties. Talk:Xiamen, for example, is too short to make archiving sensible and several sections that began before the move also have more recent comments so they could not reasonably be archived. Hence the strike-through above. Pashley (talk) 16:59, 9 February 2014 (UTC)Reply

This is worth doing, as something discussed more than a year ago on another project (WT is not WV) is most often quite irrelevant by now. It would avoid having users today inadvertently reply to comments left in 2008 (on WT not WV) by users who are currently no longer active on any of WT, WP, WV. K7L (talk) 17:42, 9 February 2014 (UTC)Reply

It is not the attribution that is killing our search engine visibility.

It's the hyperlinked attribution that helps search engines regard us as an inferior mirror site of Wikitravel.

If anything, creating new archived discussion pages that don't have any new, Wikivoyage content on them will just make our search engine visibility worse since it reinforces the mirror label.

Ephemera that provides no illumination or explanation of current article contents can be deleted - if anyone is really interested, the comments will still be there in the article history. However, there are many topics which need to remain - such as the reasons for choice of a particular currency format or a country's regional division.

These discussions are not just conversations with users who may now no longer be around to reply, they often explain the current structure and content of our guide. --118.93nzp (talk) 20:07, 9 February 2014 (UTC)Reply

Thanks for your comments everyone. Please bear in mind that my question was not about archiving policy (and you are free to create a separate discussion around that should you wish) but more specifically around removing WT attribution on discussion pages that no longer have WT content.

I get the impression that this action, where applicable, would be of some benefit. It should also be emphasized that archiving should not be undertaken solely to facilitate this. Andrewssi2 (talk) 02:27, 10 February 2014 (UTC)Reply

I believe that Wikilinks within the site do affect search results, though I am not clear on details. It therefore seems worth asking whether it is worth developing more index-type articles with many such links as an SEO method.

We already have indexes like UNESCO World Heritage Sites and Wikivoyage:World cities with many links. Itineraries link to places along the route and travel topics to suitable destinations. I have created tables at Wikivoyage:World cities/Large and Retiring abroad/Table and extra-hierarchical region articles like Bactria or Lake Tai; these also have many links.

Should we consider additional articles of these types to improve linking? Of other types? Would expanding the Go next section of articles with more links be useful? Loosening the 7+2 rule in other sections to allow more links? Could we automatically generate some sort of site map with many links? Pashley (talk) 22:35, 12 February 2014 (UTC)Reply

Internal links definitely help with SEO (it's one more clue to the search engine about which pages are most important), but we should avoid creating index pages solely for SEO purposes; any attempt to game the system is likely to eventually backfire, and it's something that would be done at the expense of our readers - content should be created because it is a useful travel resource, not because we want Google to find more links. What we CAN do that would definitely help is ensure that articles have routeboxes, that the "Go next" section links to neighboring destinations (as you've suggested), and that region articles include all of the cities within that region. In addition to being more valuable for readers, these sorts of links will have more SEO value than the "one big page of links" approach, which search engines often de-value as attempts to cheat. -- Ryan • (talk) • 22:44, 12 February 2014 (UTC)Reply

Now you've started talking sense. But all of that is extremely secondary in effectiveness to removing the hyperlinks to Wikitravel, changing our naming scheme to lose the brackets (eg: Wendover, Utah rather than Wendover (Utah), changing the lead paragraph of articles and changing section titles to rid ourselves of the duplicate penalty of being a "mirror" site of Wikitravel. --118.93nzp (talk) 01:03, 13 February 2014 (UTC)Reply

Our Kirthar National Park article mentions that the park is on the "United Nation's list of National Parks". We do not appear to have an article for that, but it has a UN homepage. This might be an index to add. Pashley (talk) 15:05, 15 February 2014 (UTC)Reply

The UNESCO World Heritage List article's opening paragraph links to several other UNESCO lists. We have articles on some but I don't think they are complete and some aren't there at all. Pashley (talk) 15:09, 15 February 2014 (UTC)Reply

I've just updated about twenty articles that mentioned LAX to make sure they include a wiki link to Los Angeles International Airport. At the time of this writing I was seeing us fifth for a Google search of "Los Angeles International Airport travel guide", so let's see if (or how much) that changes in the coming weeks. -- Ryan • (talk) • 03:44, 17 February 2014 (UTC)Reply

I did a similar exercise with Seoul Incheon airport by updating the internal links of Korean articles to the WV airport article. A search for "incheon international airport travel guide" actually places us 4th with Google, although "incheon international airport" still sends us way down to position 43. Andrewssi2 (talk) 03:55, 17 February 2014 (UTC)Reply

Another example is Gimhae_International_Airport that I did the same exercise on. "gimhae airport travel guide" comes first in Google and "gimhae airport" comes 10th. Andrewssi2 (talk) 05:51, 17 February 2014 (UTC)Reply

A search for "Los Angeles International Airport travel guide" now has Wikivoyage as the second result behind the airport's official site. I kind of doubt that the additional internal links are solely responsible, but they seem to have helped, and it might be worthwhile tracking down similar articles where we should be internally linking but aren't. -- Ryan • (talk) • 05:09, 3 April 2014 (UTC)Reply

As of today, using an incognito search, I'm seeing Wikivoyage as the #1 result for a search on "Los Angeles International Airport travel guide". Again, I don't think this can be solely attributed to the additional internal links, but clearly it didn't hurt and shows that it is worth the effort to update our existing articles to make sure that we are using internal links when an article destination is mentioned. -- Ryan • (talk) • 00:11, 25 April 2014 (UTC)Reply

118.whatever claims above that "changing our naming scheme to lose the brackets (eg: Wendover, Utah rather than Wendover (Utah)" would be a useful SEO technique. I have no idea if that is correct, though clearly it would make our titles different from WT which would perhaps reduce any duplicate content penalty the search engines apply. Other opinions?

This would be more consistent with WP usage; for example their article is at w:Wendover, Utah. Also with Commons which has, for example Phoenix, Arizona. I'm therefore inclined to think it may be a good idea.

There would be complications. One I can see is where a city and the surrounding area have the same name. I think how this should be handled varies from country to country. For example, we have New York City and Quebec City but New York State and just Quebec for the surrounding areas, and I think those are the right choices. In China, though, for most places I'd have e.g. Fuzhou for the city and Fuzhou Prefecture for the surrounding area. Shanghai has Shanghai Municipality as a redirect, which I also consider correct.

At some point, discussion of this topic might belong on Wikivoyage_talk:Naming_conventions. However, I think we should try to settle the question of whether it is useful for SEO here first. Pashley (talk) 15:37, 15 February 2014 (UTC)Reply

I like this idea for a number of reasons, not least because I think it's more aesthetically pleasing. I'm no expert in the world of SEO, but I can only think that any type of differentiation from WT's content will only do us good. --Nick ^talk 16:08, 15 February 2014 (UTC)Reply

I don't know how much difference it would make for SEO purposes - likely there would be some benefit, particularly for people searching for exact terms like "Phoenix, Arizona", but I don't think it would be significant. I think it would be an improvement from the standpoint of making our naming more consistent with other projects, and for making disambiguation clearer - for example, Córdoba (city, Argentina) could be renamed to the more logical Córdoba, Argentina (city). However, this would be a lot of work, would obviously require broad agreement, would need a bot that could accurately handle different names, and we would need to comb through Wikivoyage:Naming conventions to make sure we've thought through everything - I think district naming, and any other naming that does not currently include parentheses should stay as-is, for example. -- Ryan • (talk) • 16:36, 15 February 2014 (UTC)Reply

Córdoba, Argentina (city) wouldn't make us consistent with WP, which uses just Córdoba, Argentina for the city. Nurg (talk) 00:25, 16 February 2014 (UTC)Reply

I agree. I will not have time to do any of the work, but making our naming more consistent with the other projects seems a good thing. JuliasTravels (talk) 17:12, 15 February 2014 (UTC)Reply

User talk:Sbmeirow#Article naming is at least somewhat relevant to this discussion. More to the point is w:Wikipedia:Naming conventions (geographic names). Whatever we do, we don't want something as long and complicated as Wikipedia's policy on nomenclature, and we should never adopt a policy that always identifies what US state a city is in but does not always identify what province, departement, Land, etc. a city is in another country. Such a policy smacks of being intolerably US-centered, and it's unnecessary, too. I'd suggest maintaining our policy of never using the name of a state, Land, departement, province, et al., unless it's needed for disambiguation, but understand that that will not make our naming consistent with the names used for most US cities and towns on Wikipedia. Ikan Kekek (talk) 20:09, 15 February 2014 (UTC)Reply

Consistency across projects is a shifting target. I don't see any compelling reason to chase it, especially not with Wikidata doing the work of keeping them linked. Powers ^(talk) 22:08, 15 February 2014 (UTC)Reply

Is Cordoba experimental? Shouldn't Córdoba (city, Argentina) actually just be Córdoba and the province would be Córdoba (province), the same way that the city of Tottori is Tottori, while the prefecture is Tottori (prefecture)? That was our policy as I remember. ChubbyWimbus (talk) 08:16, 16 February 2014 (UTC)Reply

We can't do that because there are other Cordobas. Powers ^(talk) 15:03, 16 February 2014 (UTC)Reply

If we do change the naming scheme I'd recommend changing titles of that kind to Córdoba, Argentina (city) which I think is clearer. --Nick ^talk 18:18, 16 February 2014 (UTC)Reply

Additional discussion at Wikivoyage_talk:Naming_conventions#Commas_instead_of_brackets. Pashley (talk) 23:44, 12 March 2014 (UTC)Reply

Per the evidence here [30] what it did to the article Cranbrook, British Columbia worked, what I did to Fernie, British Columbia did not really work. Travel Doc James (talk · contribs · email) 06:37, 13 March 2014 (UTC)Reply

Here are the changes [31] and they included changing the headings. Travel Doc James (talk · contribs · email) 06:40, 13 March 2014 (UTC)Reply

This is the google search I did and WV comes up second [32] after hellobc.com Travel Doc James (talk · contribs · email) 06:43, 13 March 2014 (UTC)Reply

That's great. Congratulations! Unfortunately, I don't like some of the new headings you used. Not sure what to say about that. Ikan Kekek (talk) 07:02, 13 March 2014 (UTC)Reply

I still see WT as the #1 result for "Cranbrook travel guide" and #3 for "Cranbrook guide". The fact that the WV has "British Columbia" in the article name while WT does not makes it more likely that WV will outrank WT when that additional term is included. Is there another comparison available where the article names are similar (with the potential difference of just comma vs parentheses)? -- Ryan • (talk) • 07:51, 13 March 2014 (UTC)Reply

Yes you are right WT is still high for those strings. Only a partial success. And WV does not come up for either of those search terms. Not enough success to change everything. Travel Doc James (talk · contribs · email) 09:58, 13 March 2014 (UTC)Reply

Since it's been 16 months since the Fernie experiment started, I'm assuming it has run its course now and have moved the page back to a name that follows WV:Article naming conventions. -- Ryan • (talk) • 21:05, 12 December 2014 (UTC)Reply

Swept in from the pub

in having an already established article come second on Google per [33] Travel Doc James (talk · contribs · email) 06:42, 13 March 2014 (UTC)Reply

Swept in from the pub

We seem to have managed the worst of both worlds.

Most of our pages still carry the legally unnecessary HTML hyperlink to Wikitravel's main page which boosts Wikitravel's Search Engine organic results.

If this wasn't bad enough, at the foot of most of our guides we also still have another legally unnecessary HTML hyperlink to a Wikitravel article with a similar title.

However, the legally required attribution is now defective in there being a defective (ie red) link at the foot of most of our articles to its history which ever so slightly saps our own Search Engine organic results.

When you add to this the fact that we preserve a less than optimal article naming scheme and almost all of the old H2 section titles used at the site we forked from, it's no wonder that most important search engines still regard us as an inferior johnny-come-lately imitation of Wikitravel and assign us a dupe penalty! --118.93nzp (talk) 20:13, 23 March 2014 (UTC)Reply

It is not clear to me what you are talking about. I just checked two articles — small city Xiamen & big city Shanghai — and neither shows the problems you describe. Neither links to WT and both have working links to a history that shows contributions made on WT. Pashley (talk) 20:36, 23 March 2014 (UTC)Reply

I enclose a current screenshot of the foot of the Shanghai article to try and make things clearer... --118.93nzp (talk) 21:23, 23 March 2014 (UTC)Reply

I've had a look at Shanghai as well and I see what Pashley describes. Is it possible that you're somehow viewing a cached version of the page, 118.93nzp? --Nick ^talk 21:36, 23 March 2014 (UTC)Reply

You're right, Nick and Pashley! I've just been to a colleague's machine and now I see a (much better, from the SEO perspective) shortened sentence: "This article is partly based on Creative Commons Attribution/Share-Alike 3.0 Licensed work from other websites. Details of contributors can be found in the article history." Thank you Doc James and Ryan! This is great news to start my week!

Now if we can just change those H2 section headers and article naming scheme, our organic search engine results should improve by leaps and bounds... --118.93nzp (talk) 21:44, 23 March 2014 (UTC)Reply

Show it works and you will have my support. Travel Doc James (talk · contribs · email) 09:09, 24 March 2014 (UTC)Reply

I would advise fellow Wikivoyagers not to encourage 118's forum shopping by engaging him in any way on discussions of this nature. -- AndreCarrotflower (talk) 09:41, 24 March 2014 (UTC)Reply

Many of you already know that Google is not king in every country and know that Baidu dominates in Chinese and Yandex in Russian (what will happen in Arabic speaking countries, I wonder).

Here in Uganda (and, funnily enough in the francophone DRC & Rwanda, too) I've noticed that Ask is surprisingly popular.

Imagine my shame to find that whenever I search here (without logging on and on a variety of different hotel and internet cafe machines using their default browser) that a search for "Dolphin Guest House, Plot No. 26 William St, Kampala" always puts Wikitravel in pole position, followed by our Hebrew language version (http://he.wikivoyage.org/wiki/משתמש:ויקיג'אנקי/Kampala) followed by http://www.triposo.com/poi/T__3537056916f8. Our own, English language, Kampala article never appears on the first page of results.

Especially chastening is that the third ranking result (Triposo) has this text:

"Description

There аre quite а few clean аnd cheap hotels along William St, іn walking distance frоm the "Old Taxi Park" matatu station. There аre аlsо quite а few bars оn the same road.

Sources: Wikivoyage"

(Ducks to avoid the traditional brickbats and defamatory accusations from those who wish to keep their lights firmly under bushels/ ignore the Elephant in the room/ shoot the messenger...) -- Alice^✉ 22:33, 27 May 2014 (UTC)

There is extensive discussion of SEO at Wikivoyage talk:Search Expedition and I'd say that is where any new discussion belongs. Pashley (talk) 22:48, 27 May 2014 (UTC)Reply

Generally I'd say you're spot on with that comment, Pashley. Far better to keep discussions focussed in one forum so that they can reach a definitive conclusion.

The difficulty that I have is that I do care passionately about this guide and I want it to succeed.

If we are genuine about our prime directive, it's not just enough to write a readable, up-to-date, comprehensive, easy-to-use, accurate and good-looking travel guide.

We need people to read our efforts too, and not waste their time on outdated or second hand abbreviated versions without knowing about the real McCoy. It's a virtuous circle too; the more readers we have, the more contributors we'll get - some active, prolific and long-term - others just updating a closed restaurant or a defunct ferry service.

Now unfortunately, the discussions ground to a halt many months ago at Wikivoyage talk:Search Expedition#Action_points_to_boost_our_readership and we have not had the sea-change in readership that we deserve. Obviously in my opinion, some good advice was given there but the fact is that it was poo-pooed and ignored. Now more than half a year has gone by and I see no substantive SEO progress.

So, the purpose of my post here is not to provoke discussion here.

Presumably AndreCarrotflower posted a new sub-section of Some new pagebanners just above (rather than at Wikivoyage:Banner Expedition) not because he was fishing for compliments nor because he was trying to start a discussion here - but rather because he wanted merely to draw attention to an important issue that some of our community may have missed.

Unlike pretty banners, which lots of our editors take an interest in, many of our editors (both recently joined and old hands) may have missed the discussion at Wikivoyage talk:Search Expedition.

That's why I'm posting here at the pub: to raise an awareness of the English language edition's lamentable performance so far in growing our readership in the hope that new blood may have new ideas and/or wiser counsels may prevail. -- Alice^✉ 17:13, 28 May 2014 (UTC)

And the considered reaction to the above is -

shoot the messenger. —The preceding comment was added by Alice (talk • contribs)

Have you asked someone active on he.WV if they've performed any of the actions you are suggesting we should? If they have not, then the reason for coming second among the search results is the result of something else that they've done - or abstained from doing. ϒpsilon (talk) 11:34, 29 May 2014 (UTC)Reply

When I search "Dolphin Guest House Kampala", neither WT nor WV show up, but I actually didn't expect them to. Instead, hotel booking sites appear, which is typical when searching for hotel names. When I search for "Kampala travel" WT is number one and WV is number 3, so the work done on our Kampala guide may not have got it on top yet, but it's very visible. My personal experiments in simply improving articles have been mostly successful, contrary to what these doom and gloom SEO posts always try to claim. ChubbyWimbus (talk) 12:45, 29 May 2014 (UTC)Reply

I know the post is about Ask SEO, but just four your information, typing the exact same thing as you did on Google puts our guide as number 1. ChubbyWimbus (talk) 12:50, 29 May 2014 (UTC)Reply

(unindent) That's what I experience too, with several browsers and search engines. Typing "Kampala travel guide" into Startpage with Torbrowser (ie. nothing is "rigged") ranks WT as number one, then Tripadvisor, Visituganda, Lonelyplanet and then us at number five. ϒpsilon (talk) 13:06, 29 May 2014 (UTC)Reply

I'm just a nosy passer by, but here are my observations. I just Googled "World Cup 2014 travel guide" WV was the last result on the first page of Google results, WT was nowhere to be seen, but WT has a finished and well polished article on the subject. in contrast the article here is not much more then a stub, a half hearted article has been created but it amounts to nothing more then a list, yet the event is just about to happen. Therefore google are perhaps not treating you so badly, but you have to develop individual content, it's not suprising that if you create faximilies of WT articles Google only links to one site and too much of your content is the same as over at WT. even this page the "Travellers'_pub" uses the exact same name as over at WT, why? Can't you even create a diferent name for a forum? There really is no point in a seperate travel guide if you ape another guide so completely. The problem here really is lack of individuality and origional content, sorry just my opinion 86.31.250.212 17:51, 30 May 2014 (UTC)Reply

Swept in from the pub

It appears for some searches such as "Cranbrook Travel Guide" one need to click on "In order to show you the most relevant results, we have omitted some entries very similar to the 195 already displayed. If you like, you can repeat the search with the omitted results included." Not sure what if anything can be done about this. Travel Doc James (talk · contribs · email) 18:04, 12 April 2014 (UTC)Reply

In my experience, pages that have been "omitted very similar to the ones already displayed" are almost always websites whose content is completely or almost completely identical to other websites, i.e. TheFreeDictionary Encyclopedia vs. Wikipedia. Google probably has an algorithm that detects identical text, so the answer seems to lie in the fact that Wikivoyage's Cranbrook article must be essentially the same as Wikitravel's. Thus, there are two possible answers: 1) this problem will gradually solve itself, both through development of Wikivoyage and the degradation of Wikitravel through unchecked spamming and touting, and 2) I hesitate to bring this up given the identity of the past proponents of this idea, but perhaps altering section headers, etc. would be of assistance. -- AndreCarrotflower (talk) 01:48, 14 April 2014 (UTC)Reply

Perhaps the text itself needs to be reworded a little and updated periodically to distinguish WV from competing products? We've already changed two of the section headers ("Connect" and "Go next") but descriptive text can get stale if it hasn't been edited in a long while. K7L (talk) 12:05, 14 April 2014 (UTC)Reply

I can't easily use Google any more in China so I am relying on bing more for searching. I searched for 'travel Nanjing' and saw that:

• WT is still number 1
• wiki.travel.com which is a verbatim copy of WT is number 6
• WV is down on the next page at number 12

Is the problem really just down to our avoiding the word 'travel'? Seems unfair that perfect copies of the WT are ranked higher than our highly developed material. Andrewssi2 (talk) 05:35, 25 July 2014 (UTC)Reply

Seems unlikely, but I don't know. If that would be the case we should probably rename our main space from "wikivoyage:<destination>" to "travel:<destinationa>", and put our destination articles there ;) JuliasTravels (talk) 18:44, 26 July 2014 (UTC)Reply

Yes adding travel to the name of our articles may help. Travel Doc James (talk · contribs · email) 01:03, 27 July 2014 (UTC)Reply

Would it be worth asking WMF to change http(s)://en.wikivoyage.org/wiki/<destination> to http(s)://en.wikivoyage.org/travel/<destination>? It'd be awkward as this changes a server config file, but it wouldn't be a change to the MediaWiki code. K7L (talk) 04:18, 27 July 2014 (UTC)Reply

I think changing the namespace is a bad idea - it sacrifices readability and a sensible naming scheme for SEO value - but K7L's suggestion would be more feasible and would give us a noticeable boost for searches containing the word "travel". However, such a change is probably something that would need to be coordinated with other language versions so that our URLs aren't different from other languages, and would also require that proper redirects be set up for all of the old URLs. From an SEO perspective it would be worthwhile, but the process of getting agreement and making that change successfully is likely to be difficult unless someone wants to personally take charge of making it happen and coordinating with all affected parties. -- Ryan • (talk) • 06:08, 27 July 2014 (UTC)Reply

I do not like the idea of renaming a namespace, let alone the main one, at all.

When I go to a test article and look at source, I see <title>Chongqing – Travel guide at Wikivoyage</title> and <h1>Chongqing</h1>, so I do not understand why people think "travel" is missing or anything needs changing to fix that. I also see large amounts of rather complex HTML which strike me as unnecessary cruft, but quite likely I am wrong about that; the people who wrote it may have good reasons for doing it that way.

On the other hand, I do not see a "keywords" tag and would be inclined to add one and automatically populate it with the city or region name plus something like "travel, guide, hotel, hostel, accomodation, restaurant, trip, sights". I realise that many search engines now ignore that tag, but I'm not sure all do; it would cost almost nothing to add it and it might help in some cases. Pashley (talk) 16:16, 27 July 2014 (UTC)Reply

WT has both a "description" tag and a "keywords" tag, both with lots of search-bait content. Lonely Planet has multiple tags like <meta content='Top travel tips: Chóngqìng, the best advice for your Chóngqìng trip' name='description' /> Pashley (talk) 16:44, 27 July 2014 (UTC)Reply

Would changing our section headers to use more words people might search for help? That would be relatively easy to do with a bot.

Get in -> Transport to and from

Eat -> Restaurants

Buy -> Shopping

Sleep -> Hotels and other accommodation

Drink -> Bars and entertainment venues

Those examples are just first-cut; we could quite likely do better with some discussion. In addition to facilitating search, I think the right choices here might both make the site more readable and help differentiate our text from WT as search engines see us. A problem is that such changes would break many internal links that point to article sections. Pashley (talk) 16:27, 27 July 2014 (UTC)Reply

According to Google, keywords are no longer used in weighting search results. Page title, h1 tags, incoming links, and URL all have a large effect on search, however, particularly the domain name - for example, if you search for "Alaska travel", the third result is alaskatravel.com, even though that web site isn't particularly notable and almost certainly has few incoming links (try searching for any combination of words, and the domain containing those words is almost always at the top of the results list). Adding "travel" in our URLs would weight us much more heavily for searches that contain the word "travel" than just having it in our page titles, although I'm not sure that the SEO benefit is worth the work required and the downside of having URLs that are inconsistent with other Wikimedia sites.

Better section titles would help (and would also distinguish us from WT), but it has been next to impossible to gain consensus for changing them in the past. -- Ryan • (talk) • 16:31, 27 July 2014 (UTC)Reply

Two section titles were changed — "Go next" and "Connect" — so it's not impossible. I'd suspect "hotels" is already in use as a subsection in a few destinations (to distinguish from bed and breakfast, motel, caravan park, cabins, cottages or other accommodation). K7L (talk) 16:45, 27 July 2014 (UTC)Reply

What would it take to change the H1 header text, e.g. from current <h1>Chongqing</h1> to <h1>Chongqing travel guide</h1>? How would we avoid absurdities like "Pickpockets travel guide" among the travel topics?

I know Google now ignores keywords, but do other search engines? And what about "descripion" tags? Pashley (talk) 16:56, 27 July 2014 (UTC)Reply

On this talk page there have been discussions about improving WV's page rank by changing standard headings in the article. If this is something that is supposed do the job, would it not be even more effective to get rid of the headings altogether? Now and then I stumble upon maps and travel brochures where headings for different isn't written out in text but the heading for restaurants is an icon of a fork and a knife, a bed represents the sleep section etc. Would it be a good idea to try the same here? I'm talking about level 2 headings which are the same for pretty much every article we have. ϒpsilon (talk) 19:04, 12 August 2014 (UTC)Reply

Some concern in one of the latest discussions on changing headings involved the idea that in order to help our ranking, the headings ought to be relevant search terms. I don't personally know if that's true, but obviously icons wouldn't add to our searchability. It's an interesting idea, but I don't know how it would work. It might be hard to come up with icons which provide an intuitive distinction between Get in, Get out, and Go next, or which intuitively represent things like Understand and Cope. Texugo (talk) 19:15, 12 August 2014 (UTC)Reply

Isn't one of the concerns that we are "too similar" to WT, this would at least fix that problem to some extent. Why not just include search terms in the file names of the icons? If I'm not mistaken, search engines not only look for text that is visible to the reader but everything in the page's code.

Disclaimer: Just got the idea when I stumbled upon the collection of unused icons in Template:Poisym. I don't know much of anything about SEO and such, therefore I'm just posting the idea here for people to read who are more familiar with the topic and can tell if this is something useful or not (or in the worst case something that would even lower our search ranking). ϒpsilon (talk) 20:47, 12 August 2014 (UTC)Reply

Well, it's certainly not an outright horrible idea to play around with. There are also the See/Do/Buy/Eat/Drink/Sleep/Listing icons used for the template insertion buttons above the edit window on the edit screen. Anyway, if you can come up with a complete set of acceptably intuitive matching icons for all the level 2 header sections, you could certainly set up an experiment. Texugo (talk) 21:03, 12 August 2014 (UTC)Reply

Oppose. In fact I want to go the other way in some cases, e.g. replacing the map icon up at the top of many pages (which I think readers may not notice or click on) with a big red word MAP. i also want to put likely search terms into headings, and this moves the opposite way.

I think using icons in listings might be a fine idea, though it is not clear that, for example, a knife and fork would work everywhere. Use chopsticks for China? Pashley (talk) 21:15, 12 August 2014 (UTC)Reply

Replacing the map icon would at least be a great improvement for usability, as of now most new readers wouldn't even know to look for that icon. BTW elsewhere we recently discussed changes to the embedded maps. Perhaps a good replacement would be a "MAP" or "OPEN MAP" written in red at some prominent place. ϒpsilon (talk) 10:13, 13 August 2014 (UTC)Reply

My initial feeling would be a 'no, however I would be interested to see an experimental page demonstrating this. Could we use both an icon and text together in a heading? Andrewssi2 (talk) 10:19, 13 August 2014 (UTC)Reply

This is what it looks like. I would've liked to have just the logo, but in that case the Table of contents becomes unusable, so we'll have to use some sort of text headings anyway. This wasn't apparently such a good idea after all.

Some smarter person perhaps could develop the idea into something better. ϒpsilon (talk) 13:24, 13 August 2014 (UTC)Reply

It's generally a bad idea for accessibility reasons. There are workarounds of course but they all lead to less aesthetic guides. Powers ^(talk) 23:57, 13 August 2014 (UTC)Reply

In Spanish Wikivoyage an user made this proposal to change from traditional headings: see it at es:Islas Salomón. Maybe it is not what is needed here. (It currently is not being used in eswikivoyage). --Zerabat (talk) 19:07, 6 February 2017 (UTC)Reply

I had thought that switching Google to use non-personalized results (the globe icon in the upper right of the results screen) was equivalent to using an anonymizing service like Start Page... but I'm getting different results between the two. Powers ^(talk) 15:20, 24 August 2014 (UTC)Reply

I would think that often people enter in search engines the term - travel guide to city name. Would it be useful to have this phase on every article page? Propose changing the wording of status templates from This is a usable article to This is a usable city travel guide to place name. See bottom of Aachen article for example I have set up. Would this help in search engine hits? --Traveler100 (talk) 20:37, 8 September 2014 (UTC)Reply

I don't know how Aachen ranked before and how it compares to our other guides (maybe there are other important factors), but it seems to work. Since this is apparently a quick edit to the template, I'd go for it and see how it pans out. PrinceGloria (talk) 21:27, 8 September 2014 (UTC)Reply

Good idea. Go for it. Pashley (talk) 22:23, 8 September 2014 (UTC)Reply

OK started to make the edits, will try and vary the text for each type of template. --Traveler100 (talk) 06:53, 9 September 2014 (UTC)Reply

All the article classification templates should be reworded as much as possible to make them as different as possible from the old site. It is only a little bit of text on each page, so will only make a little difference to the duplicate content penalty for each page, but they are used on every article, so the cumulative effect across the site will be significant. Because the main templates are transcluded on so many pages, this is probably the biggest bang we would get for a relatively limited amount of work. Nurg (talk) 10:44, 9 September 2014 (UTC)Reply

My anecdotal experience suggests that while we tend to rank high when the word "travel" is used in search, "guide" lands us far lower. Perhaps we need to use "guide" more often. Wikitravel also seems to use a standardized template to display under their search result, which probably contains the proper keywords - maybe we should consider this as well, or at least leave the option for editors to compose one, as the first sentences of an article sometimes do not make for a good summary in search results. PrinceGloria (talk) 11:04, 9 September 2014 (UTC)Reply

I've edited Template:Usablecity to strip out the disambiguator from the article title (just as Template:Routebox does), so it shows "This city travel guide to Bowling Green has the status usable." instead of "Bowling Green (Kentucky)". I can see an argument for keeping the disambiguator, though; I'm not sure which is preferable from an SEO standpoint. So I didn't change any of the other templates. Thoughts? Powers ^(talk) 15:36, 21 September 2014 (UTC)Reply

I would think it is quite common that people type into search engines the phrase, travelling to place name or restaurants in place name or hotels in place name. Would it help if we added these sentences after the Get in, Eat and Sleep sections? The place name obviously being substituted by the page name. This is something that could be done on mass with a script. --Traveler100 (talk) 20:42, 8 September 2014 (UTC)Reply

We could just as well put "travel" and "traveller" in every other sentence and start a network of interlinking webpages. Or put "boobs", "porn" and "Kim Kardashian" somewhere in white script as people search for those as well.

I believe we're not trying to start a SEO campaign here (which is bound to fail under Google's new rules), we're just trying to make sure we get a somewhat more prominent role in search results. Seems like efforts linked to improving ACTUAL CONTENT started to work out quite well, and the old truth about Google that time is the greatest contributing factor seems to also work in our favour - we have proven to be a reliable site that is genuine and filled with original, relevant content that people click to access. This is why we are gradually advancing.

I do believe we could rethink the way we present our content, but I'd hate a robot insert a mindless phrase everywhere. Perhaps there is a way to do it in hidden script / comments, but then one would have to be careful about editing it and every editor should know why not to touch it. Regarding robots, how about articles where the relevant sections actually DO have some introductory comments, how would that be handled. And especially if they already contain a similar, but not identical, phrase.

I would find it much more worthwhile if a robot could provide a list of destination articles, perhaps grouped by number of hits or searches (maybe even not from our own data, but Google or anything else), where "Eat", "Drink" and "Sleep" have NO introductory text so that we could fill that in and perhaps drop in some keywords. PrinceGloria (talk) 21:21, 8 September 2014 (UTC)Reply

Better yet, just replace "eat" with "restaurants", "sleep" with "accommodation", "drink" with "bars and entertainment", etc. That would get useful search terms into every article without requiring convoluted & unnecessary changes to text.

It would require some changes, e.g. the discussion at China#Tea fits under "drink" but not in a "bars" section and a discussion of a country or region's cuisine does not belong under restaurants. Those could be worked out; my guess is that a couple of new optional sections would be needed in the country template. Pashley (talk) 22:47, 8 September 2014 (UTC)Reply

I'm with PrinceGloria . The fact is that although the team here do put a lot of work into those articles where we have an interest, the overall site actually has many articles that have changed only a little since the fork. Trying to second guess Google and other search engines is unlikely to be productive and just use up our collective energy.

In the interest of differentiating ourselves with actual content, I would suggest drawing up a list of destinations that perhaps the regular contributors are less familiar with and start to work on improving (not just changing) those. There should be a tipping point where the search engines no longer believe that we are a content farm. Andrewssi2 (talk) 23:48, 8 September 2014 (UTC)Reply

I wonder if it would be feasible to download the database and have an automated process compare the current version of each page to the 2012 versions (that huge batch of edits with "LtPowers replaced 'Get out' → 'Go next'") to spot articles which haven't been manually changed in any substantial way. Those will be the outdated entries which are incurring the duplicate content penalties; any popular or often-visited destinations on that list would need to be "first against the wall" for a rewrite "when the revolution comes" if we are to differentiate ourselves from WT. K7L (talk) 00:33, 9 September 2014 (UTC)Reply

Definitely feasible. The link you gave only seems to provide database dumps for 2014, so we would need to find the original. (I'm sure it can be found, just didn't see after a quick look)

Then we need to rebuild the databases side by side and do a diff comparison and filter out any automated edits. This is not trivial, but can be done. Does anyone have an easier way to do this? Andrewssi2 (talk) 00:55, 9 September 2014 (UTC)Reply

Actually yes, there is an easier way :) I forgot for a moment that the latest database has the entire edit history.

So is there an existing query tool we can leverage to tell the differences between the original WT state and the current state of the articles? Andrewssi2 (talk) 01:14, 9 September 2014 (UTC)Reply

Good points made, was just an idea. I think it would be good to have introductory text for each section, but can see it would pay to be individual for each article (not going to get round hard work). Think it would be useful to have a list of articles that have text before a listing line; check if real sentence or just not using standard listing format. Then once got rid of all bad listing styles, create a list of articles where sections start with a listing so we can add text to the pages. Yes long job I know. --Traveler100 (talk) 06:57, 9 September 2014 (UTC)Reply

When I google for "Xiamen travel guide", first five are travechinaguide, tripadvisor, WT, xiamenguide & WV in that order. Checking the pages with ctrl-U, I find both Tripadvisor and WT have a keyword-loaded "description" section in their headers, and Google displays that on the search page, rather than the actual page text. Travelchinaguide and xiamenguide use "meta name=description" instead; Google does not display that text.

That is, all four sites that beat us have something here. I'm strongly inclined to think we should add it. Pashley (talk) 19:30, 11 September 2014 (UTC)Reply

Would it be possible to have a user override of the keyword "description" if we set up a site one? Standard sentence like the other sites might be useful but it would be an advantage if contributors could change it via a parameter in the pagebanner template to something more specific and hopefully more interesting in the search results. --Traveler100 (talk) 21:14, 11 September 2014 (UTC)Reply

I've never liked seeing the same text, regardless of article, in WT blurbs on Google. I prefer how it displays an actual excerpt for Wikivoyage. But if we're missing a keyword opportunity here, that needs to be fixed. Powers ^(talk) 21:31, 11 September 2014 (UTC)Reply

I see no value in having a user override here. Just write a strong introductory sentence in the actual article. Use "meta name=description" so Google displays that, not the description text. Pashley (talk) 18:46, 30 September 2014 (UTC)Reply

Can you explain more about how to use "meta name=description"? Powers ^(talk) 19:30, 1 October 2014 (UTC)Reply

These are headings within the <head> ...</head> block of one HTML page:

• <meta name="keywords" content="Xiamen, China, Xiamen holidays, Xiamen tourism, Xiamen packages, Xiamen holiday packages, articles, guide books, review, travel, reviews, hotels, attractions, advice, deals, offers, vacation rental, vacation rentals, Xiamen, China holiday packages, Xiamen, China Overview"/>
• <meta name="description" content="Xiamen Tourism: TripAdvisor has 33,390 reviews of Xiamen Hotels, Attractions, and Restaurants making it your best Xiamen travel resource."/>

Other sites differ in detail, but most seem to have something like this. These are part of the wiki text but material automatically inserted by software during the wiki->html translation. Pashley (talk) 20:18, 1 October 2014 (UTC)Reply

I probably could have figured that out if I'd stopped to think. But I guess the better question is if you know how we can insert that tag into the generated html. Powers ^(talk) 01:54, 2 October 2014 (UTC)Reply

mw:Special:Search/html_meta brings up a mess of extensions, none of which are currently installed here. I suppose it would be possible to request one be installed if we need it? K7L (talk) 03:52, 2 October 2014 (UTC)Reply

I'm wondering if we should be using keywords in /* Go next */ links, so

Go next

Toronto

Montréal

becomes:

Go next

visit Toronto

tour Montréal

It looks spammy, but it would give better internal links for SEO. K7L (talk) 16:04, 30 September 2014 (UTC)Reply

Ryan is more well versed in these matters than I, but I believe I've read that this sort of edge-tinkering isn't usually worth the effort. Powers ^(talk) 18:01, 30 September 2014 (UTC)Reply

It looks like new Wikivoyage articles (where nothing previously existed on WV or WT) are being indexed by Google in as little as one day, if they're linked from Wikipedia. I kludged together an article on Tweed, a pop-1500 speck-on-a-map village north of Belleville (Ontario), on Saturday and brought it up to "usable". I dropped the {{wikivoyage}} tag on the corresponding Wikipedia articles. On Sunday, I tried pasting some of the text into a Google search query and, while the page didn't get top rank, it was indeed already in Google's index. (The same is likely not true of other search engines... is Google following w:special:recentchanges on WP?) Clearly, if we post something other than the two-year-old leftovers which already plague WT, Google takes notice. K7L (talk) 16:04, 30 September 2014 (UTC)Reply

Maybe not. I created Helensville 4 days ago and added a link from WP 3 days and it is not indexed by Google yet. Nurg (talk) 07:31, 1 October 2014 (UTC)Reply

Interesting. If anyone's interested in another test, take Tiel. For some weird reason, I can't even see it in search results when I search for Tiel wikivoyage... Any thoughts on why that would be? Can someone check this (just to make sure I'm not crazy)? Then we can add a link on Tiel's wp page and let's see if that changes anything. JuliasTravels (talk) 17:22, 3 October 2014 (UTC)Reply

Well "Tiel Wikivoyage" surely shows up at the top of the Google search results page. --Saqib (talk) 17:32, 3 October 2014 (UTC)Reply

Well, for clarity, I get that too when I put it just like that in Google in my browser, but not if I use any anonymous search site like startpage.com. JuliasTravels (talk) 17:45, 3 October 2014 (UTC)Reply

Tiel wikivoyage on Startpage gives me the Wikipedia article as the first result, Tiel at German Wikivoyage second, Eastern Netherlands as the third... and Tiel itself isn't among the first ten hits. ϒpsilon (talk) 17:55, 3 October 2014 (UTC)Reply

4th on list when I use Tiel travel guide --Traveler100 (talk) 18:14, 3 October 2014 (UTC)Reply

Helensville created a week ago (less 7 hrs) and still not indexed AT ALL by Google or Bing. The WP article has been reindexed by Google since the link was added there. Google seems to be deliberately indexing WV only infrequently. This is the opposite conclusion to the first post in this thread. Nurg (talk) 01:21, 4 October 2014 (UTC)Reply

Search engines are slow to update. Tweed, created Sept 27, was on Google relatively quickly but is only just starting to turn up on Bing or DuckDuckGo now, at least half a month later. Before then, nothing except on Google. K7L (talk) 03:22, 17 October 2014 (UTC)Reply

Swept in from the pub

A proposal: Should the subheadings under "get in" and "get around" be changed to match Wikivoyage imperative style, ie:

• By car → Drive
• By air → Fly
• By train, by bus, by camel → Ride (by train, by bus...)
• By horse → Trot, Jump, Gallop
• By bicycle, by motorcycle → Bike
• By sea, by ferry → Sail
• By public transit (subway, taxi, streetcar, rickshaw) → Ride

It would make us look a little more different from other travel guides, but I'm unsure if there's a better word than "Ride" for the numerous public conveyances. K7L (talk) 18:47, 31 July 2014 (UTC)Reply

The enforcers of British English will hate "Ride," because they change phrases like "cab ride" to "taxi journey" all the time. But I disagree with putting all these disparate forms of transportation under "Ride," too. Ikan Kekek (talk) 20:45, 31 July 2014 (UTC)Reply

Should we collect basically all forms of land transportation where someone else is controlling the vehicle under a single "Ride" heading? I don't think it would be a smart idea at all. ϒpsilon (talk) 20:58, 31 July 2014 (UTC)Reply

We should not. The heading for bus and train should be "Take the bus" and "Take the train" if they need to be imperatives for some reason. 2001:5C0:1000:A:0:0:0:100F 22:52, 31 July 2014 (UTC)Reply

A 'ride' to me implies something done for fun, rather than utilitarian transportation. Linking motorcycles and pushbikes is also unhelpful. I like the idea, but the fact seems to be that there aren't imperatives to match each of the modes of transport. --Inas (talk) 23:03, 31 July 2014 ( TC)

I don't support any change at this time. I don't think any of the suggestions above would do anything but make the headings more cumbersome. Texugo (talk) 23:05, 31 July 2014 (UTC)Reply

The imperative is in the heading "Get in", and "by ..." is the right qualifier eg "Get in by train". Also "ride" is a verb I am more likely to use for a bike than a train. It would be far better to spend the time improving the information in "Get in" in some articles - too many articles have only one way of getting in when others are available. AlasdairW (talk) 23:18, 31 July 2014 (UTC)Reply

What is the actual benefit in changing subheadings to imperatives? Copying WP style for the sake of it doesn't actually help us. I recall there was some discussion before that changing subheadings 'may' improve SEO results. Andrewssi2 (talk) 01:32, 1 August 2014 (UTC)Reply

<indent reset> But really, there are so many different things to put our attention and effort to. This particular proposal doesn't hold water for the reasons mention above, and our current headings do serve us well. New, better, original content not only "may", but does help our SEO, and it helps bring in and retain readership and generate positive buzz with a snowball effect. There are still many articles on important destinations that are quite shambolic. PrinceGloria (talk) 15:14, 1 August 2014 (UTC)Reply

For SEO, it would be far better to change headings so that they contained likely search terms. get in->transport to & from, get around->local transit, eat->restaurants, sleep->hotels, etc. There would be problems with that, & details to be worked out, of course.

From that viewpoint, using "ride" would be appallingly stupid, replacing several fairly common search terms with one no-one is likely to use. Pashley (talk) 18:11, 1 August 2014 (UTC)Reply

Although I agree that is what 'should' happen with SEO when changing articles headings, do we have any actual evidence that it would? Andrewssi2 (talk) 00:36, 4 August 2014 (UTC)Reply

Here is a link to an experiment last year. As far as I can tell it didn't succeed, although the sub headings that were chosen were probably not the wisest choices if they were looking to improve SEO. Andrewssi2 (talk) 00:44, 4 August 2014 (UTC)Reply

That experiment was, I think, mainly to see if making section headers different from WT improved search engine rankings, presumably mainly by avoiding the penalty the engines apply to duplicate/copied content. User:W. Frank who created the page thought it showed evidence of success.

As I see it, the headings chosen there were remarkably bad for SEO, probably because he was trying to stay with our policy of using verbs. Changing Eat to Dine. for example, is dumb but harmless. Things like Buy to Spend up! or Drink to Quench are worse; they remove somewhat likely search terms and replace them with even less likely ones.

I'd say drop the policy of using verbs (which was an arbitrary, if reasonable, choice anyway) and make changes aimed mainly at getting likely search terms into headings. Pashley (talk) 01:58, 4 August 2014 (UTC)Reply

I'd happily run an experiment at Xiamen if others think that is a good idea. 02:03, 4 August 2014 (UTC)

There's nothing inherently magical about using verbs. We used them originally because we thought it'd make us look different from other travel guides, supposedly? If so, their existence on both WT and WV (with just "connect" and "go next" as minor differences) defeats that objective. We're no better off with these than with the stock (attractions, activities, shopping, restaurants, nightlife, accommodation) standard header set from any printed guide. K7L (talk) 02:36, 4 August 2014 (UTC)Reply

If the only reason to do this is to 'look different' in a minor aesthetic sense, then I'd say the change isn't particularly compelling. High effort, near zero benefit. Andrewssi2 (talk) 02:45, 4 August 2014 (UTC)Reply

If there's the suspicion that this might help SEO, and anyone prefers to put their energy in this rather than one of the countless other tasks on this site, I see no reason to not try. Why can't we just repeat the test Frank tried to do with more likely search terms? What's the harm? JuliasTravels (talk) 11:32, 4 August 2014 (UTC)Reply

I support Pashley for his proposed experiment article above. Andrewssi2 (talk) 11:58, 4 August 2014 (UTC)Reply

I don't oppose an experiment with likely search terms, but I don't think Pashley's suggestions above (Ride, etc.) are quite worthy of even testing. I don't think testing will produce respectable, actionable results unless we:

a) use likely search terms - "Ride" or "Gallop" are less likely search terms than we currently have

b) use terms which could be substituted without tedious reorganization of the content of every article - i.e. neither "Bars" nor "Nightlife" could be used in place of "Drink" because that section often contains other things as well (cafés, info on water quality, local specialty beverages, etc.); neither "Restaurants" nor "Cuisine" are a good replacement for "Eat" because it contains info on both, etc.

c) run the same test on a group of articles - Since SEO rankings tend to vary over time and depend on many other factors as well, an arguably positive result on a single article is not going to convince me anyway. The test, the exact same changes need to be implemented on a number of articles, say at least 10 or so, including destinations of different types (popular searches, less popular ones, different sizes, region/city, etc.).

If we do all the above and the overall results demonstrate a quite significant improvement across the board, only then might I be inclined to think this type of change to be worth the trouble. Texugo (talk) 12:22, 4 August 2014 (UTC)Reply

That makes sense. Any ideas on which search terms would be best? We have over 25000 articles and these experiments don't make them less useful for a traveller to any specific destination. We could even have 2 or 3 groups of articles to test different headings, if we'd want. I'd say as long as we document the experiments well, so they don't end up forgotten, it's all no problem. And then how long should we wait to have results that can mean something? JuliasTravels (talk) 12:38, 4 August 2014 (UTC)Reply

Personally, I'm rather attached to the headers we have and skeptical that such a change will make a difference that's worth the effort, so while I don't necessarily oppose such testing, I don't feel very compelled to push it forward either. Texugo (talk) 12:51, 4 August 2014 (UTC)Reply

I did not suggest Ride; in fact I pointed out that it was a bad choice for SEO.

I will go ahead with an experiment; see Talk:Xiamen for details. Pashley (talk) 14:56, 4 August 2014 (UTC)Reply

Oh, sorry. My mistake. But what headers do you plan to use? And why bother with such a necessarily long-term experiment on only a single article, knowing that results from only a single page will not be reliable or scientific enough to convince everyone of its effectiveness or lack thereof?

Why not do this:

• first, come up with a workable set of replacements
• implement them as suggested above in a variety of different articles, at least 10 or so
• choose another equal number of comparable articles as control articles (which don't implement the proposed change)
• put all the starting stats in one place for tracking, say, a subpage of Wikivoyage:Search Expedition
• track the stats for all, monthly, for a few months
• see how it goes

Yes, it's more work, involving 20 or more pages etc., but at least I might believe you when you come back months hence and tell me whether it works or not. Texugo (talk) 15:13, 4 August 2014 (UTC)Reply

I agree we probably need to do all that eventually. Talk:Xiamen#Headings experiment is initial exploration. 15:57, 4 August 2014 (UTC)

Swept in from the pub

Has any authority (related to Wikivoyage) noticed and attempted to resolve search results that should but don't reflect Wikivoyage content? Moments ago, on Google and Yahoo Advanced Search, queries for "Quebec City" (only) resulted in no "hit" for Wikivoyage, but one or more hits for Wikitravel. Only if "Wikivoyage" is made a required parameter do hits for Wikivoyage appear. This appears the case for many destinations. Does this not reflect an error in search provider algorithms that adversely affects us? Hennejohn (talk) 18:26, 5 September 2014 (UTC)Reply

This is a well known problem for us, although not really an 'error' as such by the search engines. See Wikivoyage:Search Expedition and the many discussions on Wikivoyage talk:Search Expedition. Andrewssi2 (talk) 23:45, 5 September 2014 (UTC)Reply

This really needs to be adressed. WikiVoyage is far too irrelevant to Search Engines, in contrast to clumsy WikiTravel. If we don't overcome these issues, we might just as well go back to editing Wikipedia articles. Cheers, Horst-schlaemma (talk) 09:29, 6 September 2014 (UTC)Reply

Well, Wikipedia articles are for a different audience and have different content. There is no harm working on both.

The links I provided should demonstrate to you that people are proactively addressing this and are seeing some minor victories, it is just a rather difficult thing to fix. You are also welcome to get involved. Andrewssi2 (talk) 09:35, 6 September 2014 (UTC)Reply

Yes, and for the record, there are many in our community who started out at Wikivoyage and only began editing Wikipedia later; still others who've never edited Wikipedia at all. -- AndreCarrotflower (talk) 14:27, 6 September 2014 (UTC)Reply

I don't intend to pit the Wikis against each other. I'm just saying that people are (way) more likely to end up at Wikipedia if they look for information about some place, rather than at Wikivoyage - which may cater to their interests far better. We should make sure Wikipedia links back to Voyage more. I'm also rather good at SEO mechanisms, but not at Wiki scales I fear. And frankly I wouldn't know how to make my way into the code. When I got to know more about how it actually works I could come up with crucial advice to improve the results. Cheers, Horst-schlaemma (talk) 20:59, 7 September 2014 (UTC)Reply

Just a few random thoughts on getting more visitors to WV

1. We started out with a lot of content from elsewhere - since most of the information we started with was imported. I believe that search engines can detect that and put WV at the bottom of pile.
2. Have good quality content.. (as few errors as possible) - which we are striving for and doing very well at. Regularly edit and update article pages as well as provide new content - valid links etc. Quality means more much more than quantity - at least in my mind... "Build it and they will come"
3. Meta tags - Meta tags I understand are not as important as they used to be (meta title, description and keywords)
4. Links - backlinks from MW
5. Perhaps some kind of useful tool or tool(s) built into WV to assist the traveller. -- add audio file for various phrase books (do we have that ability), currency conversion tool (could be a bit tricky)
6. I don't know if an RSS feed would be useful, or share Star articles on social media? Do any of the Wikis have a blog somewhere or is that even a valid suggestion?
7. Format - in most instances WV has an article format and structure that we follow. What do our visitors think? - pages too long or too short - is the layout great or boring - are we well organised or not? Does the visitor have a way of providing that input -- did visitor find information useful or not or find what they were looking for... perhaps a survey page? Cheers! Matroc (talk) 22:58, 7 September 2014 (UTC)Reply

Matroc, the last item on your list intrigues me. For all the efforts we've made lately to attract new readers and take a bite out of the lead WT has over us on Alexa, one thing we've inexplicably never tried doing is engaging our users - not just the regular contributors and community bigwigs, but the average folks who happen upon our guides after a Google search or stumble in from Wikipedia - and asking them directly, on pages that are easily accessible to newbies who don't know their way around, how we can better do what we do. When it's the same old names doing the deliberating, it's easy for the decision-making process on how to present content to become stale or entrenched. We should reach out to readers who may not ever chime in on a policy discussion and find out what they would like to see on Wikivoyage that's not already here, what we aren't doing well and should be rethought or abandoned, etc. -- AndreCarrotflower (talk) 23:57, 7 September 2014 (UTC)Reply

Regarding social media, Wikivoyage currently has a Facebook page and a Twitter feed which are updated when a new DotM/OtBP/FTT goes into effect, and less frequently when other important events transpire. (Those are the two that I know of; there may be others as well.) See Wikivoyage:Social media. -- AndreCarrotflower (talk) 00:00, 8 September 2014 (UTC)Reply

For all I can see, there's loads and loads of Wikipedia users/editors that primarily want to cater to the interests of tourists. Any idea how to get them here? Perhaps we could have a general message at the top of Wikipedia for a few days to get more attention. Something along the lines of "You want to provide tourist information about a place? Come to Wikivoyage!" with a starter link. There could also be a reminder at the index page that stays a little longer. Just a thought. Cheers, Horst-schlaemma (talk) 17:27, 13 September 2014 (UTC)Reply

Providing links to decent pages here might attract the attention of a few (and it's a long-term way of helping people find us when they are looking for travel information). Banners (sitenotices or centralnotices) would presumably require consent from the target wiki. When we have a project going, we could leave messages at related pages. For example, we could tell w:en:WikiProject India about the proposed India expedition. Is there a newsletter? If so, it could be delivered to off-wiki accounts (if anyone wants to sign up for it). Have any of these been tried in the past? WhatamIdoing (talk) 21:14, 13 September 2014 (UTC)Reply

The India Expedition is up and running and needs more hands on deck. I'll have a look at that WP project page. Ikan Kekek (talk) 21:30, 13 September 2014 (UTC)Reply

WhatamIdoing, that WP link didn't work. Please try again. Ikan Kekek (talk) 21:32, 13 September 2014 (UTC)Reply

Try w:en:Wikipedia:WikiProject India. As a more general point, it is worth noting that there are projects covering many countries and major cities. Try entering Wikipedia:Wikiproject and then a country/huge city name in the WP search box. AlasdairW (talk) 23:06, 13 September 2014 (UTC)Reply

There's also a WP:TRAVEL Wikiproject with subprojects like WP:HOTEL, but it seems to be dead as a doornail. A WP:ROUTE66 project is almost a dead, although there's an active project updating US National Register of Historic Places (NRHP) entries and projects for road/highway articles. The city/region and country projects look to be the best bet, if they're active, as we actually do link to WP for same-topic cities. K7L (talk) 02:31, 14 September 2014 (UTC)Reply

Thanks a lot, AlasdairW. I'll try posting a message on that project's talk page. Ikan Kekek (talk) 02:47, 14 September 2014 (UTC)Reply

Odd Google result. All results using U.S. Google with personalized results turned off:

• "Palmyra travel guide" has Palmyra (New York) as the fourth result (behind WT's articles on Palmyra, Palmyra Atoll, and Palmyra (New York)).
• "Palmyra wikivoyage" has it second behind Palmyra.
• "Palmyra New York travel guide" has it sixth.
• "Palmyra New York wikivoyage" has it first, as expected.

but...

• "Palmyra NY travel guide" doesn't list Palmyra (New York) at all -- even though WT's article is first.
• "Palmyra NY wikivoyage" also doesn't list it at all.

(Note: if I use Startpage.com I can't get our article to come up at all, even with a search for "Palmyra New York wikivoyage"!)

Any idea what's going on? This seems like a significant hole in our search coverage.

-- Powers ^(talk) 11:49, 28 October 2014 (UTC)Reply

#Naming scheme? above deals with a related issue, "Palmyra (New York)" versus "Palmyra, New York". I'd say move the article to the latter and add "Palmyra, NY" as a redirect. Pashley (talk) 13:18, 28 October 2014 (UTC)Reply

We can't move the article without consensus to update the naming conventions, and I don't think it'd solve the problem anyway. But adding redirects might work; we should probably consider systematically adding them for all U.S. destinations. But that still leaves unanswered the question of why WT's article comes up with an "NY" search; they have no redirects in place, nor do they use the comma convention for disambiguation. Powers ^(talk) 14:08, 28 October 2014 (UTC)Reply

I would guess WT probably has incoming links from one or more third-party websites that use the phrase "Palmyra, NY" in the text of the link. Not much we can do to replicate that. I also can't get our Palmyra article to show up on startpage and I have no clue why that might be, but I've gone ahead and added the Palmyra, NY redirect. I've also added redirects for a few other destinations as a test and kept records of what the search results looked like before I started. All searches done through startpage:

• Salt Lake City, UT redirect added. Our Salt Lake City article ranked 14th for the phrase "salt lake city travel guide" and 22nd for "salt lake city ut travel guide".
• Billings, MT redirect added. Our Billings article ranked 12th for "billings travel guide" and 23rd for "billings mt travel guide".
• Chicago, IL redirect added. Our Chicago article ranked 28th for "chicago travel guide" and 37th for "chicago il travel guide".
• Branson, MO redirect added. Our Branson article ranked 20th for "branson travel guide" and 25th for "branson mo travel guide".

Thatotherperson^talkcontribs 00:18, 30 October 2014 (UTC)Reply

Three weeks later and there's no overall pattern of improvement. The Palmyra issue appears to have been fixed, but now our Billings article doesn't show up when the letters MT are used. All four articles have gone down a few spots when the state abbreviation isn't used; there's also been a slight decline for the phrase "salt lake city travel guide", but no change for "branson mo travel guide" and a slight improvement for "chicago il".
Thatotherperson^talkcontribs 02:24, 21 November 2014 (UTC)Reply

It looks like many of our articles are not linked FROM Wikipedia. I've recently been creating and updating articles about Madagascar and was surprised to discover that w:Madagascar did not have any links to Wikivoyage, nor did many of the city and park articles for that country. I've added several such links to Wikipedia using the w:Template:Wikivoyage and w:Template:Sister project links templates (the latter requires the "voy=Article name" parameter) and would encourage others to do so as well. As an experiment, here are the incognito Google search results I saw when searching for several articles that I have linked from Wikipedia today using the search term "Article name travel guide" - I would expect these to improve in a week or two once Google re-indexes the Wikipedia page:

• Madagascar - not in the top 100 results
• Isalo National Park - #1
• Ranomafana National Park - #6
• Antananarivo - #5
• Taolagnaro - #1
• Andasibe-Mantadia National Park - not in the top 30 results, but I suspect Google hasn't indexed this Wikivoyage article yet
• Moss Beach - #11

-- Ryan • (talk) • 06:09, 29 November 2014 (UTC)Reply

I suppose the real question is how many usable articles are missing WP → WV links. The stubby "X is in Y {{subst:smallcity skeleton}}" rubbish we typically don't link from WP, lest the links end up more spammy than useful. K7L (talk) 06:35, 29 November 2014 (UTC)Reply

Yet we link from WV articles to WP articles that are stubs. Ikan Kekek (talk) 06:38, 29 November 2014 (UTC)Reply

Is that why Madagascar wasn't linked from Wikipedia, because it's still tagged at "outline" status? Do we have a bot that is still running to update Wikipedia automatically? Or is there some other process for doing so? And should I not be tagging WP articles until the WV article is more fully developed? I thought once we have an article here we link it to WP, and vice-versa. -- Ryan • (talk) • 06:49, 29 November 2014 (UTC)Reply

Wikivoyage:Links from Wikipedia#When to link answers the question re: interwiki links for outline articles, although the questions about bots and such are still valid. -- Ryan • (talk) • 06:53, 29 November 2014 (UTC)Reply

I use a lower threshold than that guideline page. I'll link to an outline that includes several key attractions and/or activities, even if there is not much in the other sections. For example, I recently linked to Coromandel (4 attractions listed), but decided against linking to Whitianga as the See and Do sections are rather thin. I think we should loosen up our guideline a bit. In the end, WP policies and practices will rule, not ours. Nurg (talk) 08:51, 29 November 2014 (UTC)Reply

And Ikan Kekek has opened discussion at Wikivoyage talk:Links from Wikipedia#Not linking to WV articles that are Outlines. Nurg (talk) 08:59, 29 November 2014 (UTC)Reply

Yes. I had forgotten that the discussion was here, rather than at the Pub. However, it's probably good to try to come to some kind of consensus on policy there. Ikan Kekek (talk) 09:16, 29 November 2014 (UTC)Reply

Here's an update on the stats above (obtained by searching for "Article name travel guide" on Google using an incognito browser window with Chrome):

Article	Search result on 29-Nov	on 13-Dec	on 04-Jan
Madagascar	not in the top 100 results	not in the top 100 results	not in the top 100 results
Isalo National Park	#1	#2	#8
Ranomafana National Park	#6	#7	#6
Antananarivo	#5	#5	#5
Taolagnaro	#1	#1	#1
Andasibe-Mantadia National Park	not in the top 30 results (Google probably hasn't indexed this article yet)	#7	#4
Moss Beach	#11	#7	#8
Marina del Rey	not yet linked from Wikipedia	#6	#4
Oudtshoorn	N/A	#5	#7
Bar Harbor	N/A	N/A	#19

In the case of Isalo National Park, Wikipedia jumped us, apparently because the link to Wikivoyage in the Wikipedia article now includes the words "travel guide", and I assume the same probably happened with Ranomafana. The only article that showed a noticeable improvement was Moss Beach. This test was clearly inconclusive - we know that links from Wikipedia help, but they don't appear to be a magic bullet, and appear to be just one more tool to incrementally improve SEO. So far it seems like the best way to improve search result rankings has been to update article text to ensure it is significantly different from WT, while things like links from WP and internal links from other WV articles have also proven to have smaller effects. -- Ryan • (talk) • 16:33, 13 December 2014 (UTC)Reply

For comparison, Oudtshoorn is currently the fifth Google result in an incognito search for "Oudtshoorn travel guide". The article was already linked from WP, but today I've expanded the article and suspect that may improve its ranking in a few weeks once Google has re-indexed it. -- Ryan • (talk) • 18:25, 13 December 2014 (UTC)Reply

Thanks! Incremental improvements are great. I look forward to see future updates. --Andrewssi2 (talk) 22:14, 13 December 2014 (UTC)Reply

I've been watching a couple of articles too, and came to more or less the same conclusion. Adding individual WP links doesn't seem to have a huge impact on specific articles (although there is likely a general effect). Completely rewriting stubs leaves us buried at page 5 or so, but creating or re-writing rather well-developed articles brings us high up in the ranks. The absence of a wp link doesn't seem to stop that. I obviously not mean to say that it's not useful to add those links; of course it helps interested people to find the articles. What's interesting, is that we seem to perform better for searches after "[Place] travel guide" than we doe for "[Place] guide" or "[Place] travel". E.g. Bolton, which lists us 1st for "Bolton travel guide" but 19th and after WT for "Bolton guide". JuliasTravels (talk) 22:42, 13 December 2014 (UTC)Reply

These findings do raise the question again if we should rewrite some of the world's main destinations...JuliasTravels (talk) 22:44, 13 December 2014 (UTC)Reply

Is there anyway to compare WV and WT articles for similarities and have them ranked somehow? For example Paris may have much more similar text than New York? (I just made the example up with no basis)

I think we must have a large number of articles of big cities barely touched since the founding of WV. --Andrewssi2 (talk) 22:50, 13 December 2014 (UTC)Reply

You can compare pages with http://www.copyscape.com/compare.php, which is listed in the Tools on the project page. Nurg (talk) 00:41, 14 December 2014 (UTC)Reply

Interesting experiment with Eyam today. This is an English village that I created in August 2014, and has no corresponding article in WT.

A google search for 'Eyam Travel Guide' yesterday put us on a respectable 5th place.

I then went to the Wikipedia article w:Eyam and added the WV article as a sister project.

I just tried the same search a few hours later and we are second in the Google results for that article.

I wouldn't say it was highly scientific, but I'm wondering if we put more effort into adding WP links to our distinct articles (i.e. those not shared with WT) then our SEO rankings increase?--Andrewssi2 (talk) 08:53, 28 December 2014 (UTC)Reply

I a currently updating the Derbyshire articles, so that may also be a factor. Actually was planing on linking from WP Derbyshire once I had that page up to usable status. --Traveler100 (talk) 09:07, 28 December 2014 (UTC)Reply

An improvement in a few hours seems extremely odd, since search engine results typically take at least a week or two to update (search spiders don't reindex most pages daily). Google search results are personalized based on user search history, so were you using an anonymous browser window when you did this test? If not, the result might simply be due to the fact that you clicked on a Wikivoyage result previously. To set up an anonymous window in Chrome choose "New incognito window" from the Chrome menu; other browsers offer similar anonymous browsing features. -- Ryan • (talk) • 16:37, 28 December 2014 (UTC)Reply

That must be it. Searching anonymously for Eyam Travel Guide I unfortunately don't see us on the first 3 pages at all (haven't checked further). JuliasTravels (talk) 21:06, 28 December 2014 (UTC)Reply

That was probably it. I went into incognito mode with my Chrome browser and it appears fifth place again. (Not sure why JuliasTravels can not see it at all) Andrewssi2 (talk) 22:21, 28 December 2014 (UTC)Reply

I'm not sure incognito mode in Chrome provides anonymous results? I think it just leaves no cookies or history on your system, but providers or websites you use can still monitor your activity. I use startpage.com for anonymous browsing when testing wv article rankings. JuliasTravels (talk) 22:32, 28 December 2014 (UTC)Reply

According to w:Privacy_mode, the incognito mode should hide everything except my IP address (which may influence Google, but I thought it would not). I am consistently getting 5th place in other browsers as well.

I can only assume that startpage.com is using a different type of search than I would otherwise get through google.com . I will keep trying on different wifi hotspots though. Andrewssi2 (talk) 02:51, 29 December 2014 (UTC)Reply

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘

The table above has been updated with latest results, most of which make no sense to me - we seem to be bouncing around a bit with no real pattern. Best option seems to be to continue to follow best SEO practices (updating articles with good content) and expect things to slowly improve. -- Ryan • (talk) • 06:03, 5 January 2015 (UTC)Reply

I thought I would separate this out since it has little to do with the Wikipedia links above.

Using the compare tool I was happy to see that our Hong Kong article matches only 2% of text from the WT original.

However Google searching for the term puts 'Hong Kong' in a respectable 7th place for WT. For WV I could not find even by clicking far down the results.

Google search for the term 'Hong Kong Travel Guide' put WT at 2nd place, and WV at 35th !

Does this mean that rewriting an article has very little impact on SEO, and external links are more important? Andrewssi2 (talk) 01:37, 14 December 2014 (UTC)Reply

Bing search for 'Hong Kong Travel Guide' still puts WT at number 2, but WV way down at 63. --Andrewssi2 (talk) 01:41, 14 December 2014 (UTC)Reply

I don't know if this is the sole problem, but I think the copyscape tool is failing us here - there is way more than 2% text shared between those two articles, but copyscape is ignoring numerous paragraphs that contain exactly the same text. -- Ryan • (talk) • 01:42, 14 December 2014 (UTC)Reply

Andrewssi2, have you looked at the table of articles at Wikivoyage:Search Expedition#Article Rankings? Nurg (talk) 04:18, 14 December 2014 (UTC)Reply

Hi Nurg, yes I am aware of that table. I have to say that I am having difficulty determining anything conclusive from it, unless the conclusion is that working on lesser known articles will raise SEO rankings, but not 'main destination' articles? --Andrewssi2 (talk) 06:27, 14 December 2014 (UTC)Reply

Well frankly, I think you can't even draw that conclusion :-) As far as I'm aware, we've rewritten and expanded a good number of small destinations, or outlines at WT. If our "new" article is far more extensive, we rank high and on top rather consistently. If our "new" article is unique but equally small as it was on WT, we don't do so well, and WT usually stays ahead. If we're talking about almost empty skeletons, both WT and WV typically don't show up on the first pages at all. For smaller destinations, it all seems to come down to the amount of the unique content.

We have however not completely rewritten any well-developed articles on main destinations, so we can't say what the effect would be. It would be a lot of work. I'd be willing to work on a single experiment just to check, mostly out of curiosity, but only if it would be a joint effort.

For main destinations we must keep in mind that WT probably still has the benefit of loads of historic links pointing to it from all over the internet. No-one knows how Google operates, and incoming links may not be as important as they once were, but it's rather likely that they still play a role. At least for popular destinations that have many websites dedicated to them and plenty of unique content to choose from. It would make sense that this is less true for outline articles on smaller destinations, as there's less competition in terms of relevant content and there aren't many links to them anyway. It's not unthinkable that Google would ignore Wikipedia links to a sister project. JuliasTravels (talk) 13:41, 14 December 2014 (UTC)Reply

I'm happy to help re-write a major article. Any article candidates? :) I guess ideally one that hasn't had that much done to it since the beginning of WV. Andrewssi2 (talk) 21:11, 14 December 2014 (UTC)Reply

More-or-less any page ranking algorithm depends, at least in part, on incoming links, and links from highly ranked pages or sites count for more than other links. Links from Wikipedia are therefore of considerable value to WV. I was under the impression that a bot had long since added them for all the articles with matching names, but that does not seem to be the case. There is plenty of discussion, e.g. at #More prominent link from Wikipedia but it looks as though the action was not taken.

I noticed this in w:Dumaguete and checked three other articles, w:Cebu, w:Toronto, and w:Taxila without finding any WV links. Can this be fixed? Obviously individual articles could have w:Template:Wikivoyage added manually but a bot would be a better solution; also obviously any bot that runs on WP would need approval there.

An aside: I did find a link on the Taxila page to a World66 page (another wiki owned by Internet Brands who control WT) and deleted it since that page contained only copyright violation text copied from a Pakistani gov't site. Search on WP shows 116 hits for "World66"; I'd like to see them all removed. I checked two of them and both were text that is word-for-word identical on many sites, in one case including WP, but I did not delete those since I could not be certain who had copied whom. Pashley (talk) 13:08, 15 December 2014 (UTC)Reply

About the incoming links part, you may want to also read the past days' discussion two sections up: Wikivoyage_talk:Search_Expedition#New_article_links_FROM_Wikipedia. JuliasTravels (talk) 13:29, 15 December 2014 (UTC)Reply

A link to an external page which contains text already present verbatim in WP is of no use to WP. I'd be inclined to remove it per w:WP:CIRCULAR if nothing else. K7L (talk) 16:39, 15 December 2014 (UTC)Reply

It isn't always clear to me when Google is filtering one of our articles from its results because it thinks our article is a duplicate of one found elsewhere. For example, if you do a Google search for "Madagascar travel guide" then we are nowhere to be found, but it was unclear if that was due to our article ranking poorly or being filtered from Google's results.

To determine if our article is being filtered as a duplicate (as opposed to just ranking very low), the following seems to work:

1. Search for "Article name travel guide" on Google. Remember to use a Chrome incognito window or a tool like startpage.com to ensure that you are seeing results that aren't affected by your search history.
2. Manually modify the Google search results URL and add "&filter=0" - for example:

   www.google.com/search?q=madagascar+travel+guide&oq=madagascar+travel+guide&filter=0

3. The new search results should now show content without filtering out duplicates. If the Wikivoyage article shows up and it didn't before then the problem is that Google is treating the Wikivoyage artcle as a duplicate.

I don't know how reliable this approach is, but when I did it for the Madagascar search above Wikivoyage is result #18, while it isn't in the top one hundred in a normal search; the same was true for a search for Hermanus. In cases where an article doesn't show up in searches unless filtering is disabled it indicates that updating the article content should have a significant impact on Google ranking, as opposed to another article that may just be low ranking but not getting filtered as a duplicate. -- Ryan • (talk) • 01:07, 9 January 2015 (UTC)Reply

So would it be a fair assumption that rewriting Madagascar will result in the filter being removed for that article ?

Would it be worth to attempt a 'partial rewrite' first? i.e. if 30% of the article is different that may be sufficient? Andrewssi2 (talk) 01:46, 9 January 2015 (UTC)Reply

I'm not suggesting we start over from scratch - from what I've read no one seems to know how different content needs to be in order for a search engine to remove the duplicate filter, and it's not even clear whether certain things such as headings, images or links count "more" towards determining if an article is a duplicate - but it's clear that we don't need 100% original content for the duplicate filter to be removed since we do have lots of articles showing up in Google results. My goal in bringing the "filter=0" switch up was solely to share a way of determining if an article isn't showing up in the results due to a duplicate filter, in which case updating the article content sufficiently for Google to no longer view it as a duplicate is the only way to resolve the issue.

As an example, I've slowly been updating our Culver City article, but so far haven't done enough to escape the duplicate penalty (we're #16 with the filter=0 switch). I would expect at some point enough will have been changed that Google will stop filtering us, but I'm not sure how much needs to be done for that to happen. -- Ryan • (talk) • 01:56, 9 January 2015 (UTC)Reply

Confirmed that this trick works with Rochester (New York) and that it's being filtered despite two years of periodic updates. It looks like Niagara Falls (New York) is being filtered but Niagara Falls (Ontario) isn't. Powers ^(talk) 02:26, 9 January 2015 (UTC)Reply

Update: recent re-writes to the "Understand" section and elsewhere were sufficient for Google to no longer see Madagascar and Hermanus as duplicate content (Madagascar is result #11 in an unfiltered search for "Madagascar travel guide", "Hermanus" is result #13). Unfortunately, despite some rewrites, Culver City is still being filtered, although for some reason if you use the "filter=0" switch we've jumped from #16 to #8. I verified that the version in the Google cache is for a version that was spidered after the rewrites were done for all three articles. You can look at the diffs of any of those articles to get a sense of how much has changed since the split from WT, and also how much changed in the past two weeks in order to push us over the edge and escape the "duplicate content" filter. -- Ryan • (talk) • 03:07, 16 January 2015 (UTC)Reply

Another update: I'm now seeing Culver City appearing in Google search results as the #7 result in an incognito search for "Culver City travel guide", despite the fact that the Google cache indicates that article was last spidered on January 13 (three days prior to my last update). Whatever the reason, recent rewrites to the introductory paragraph and Understand sections seems to have pushed us over the threshold and out of the duplicate content black hole that we were previously in. -- Ryan • (talk) • 03:08, 20 January 2015 (UTC)Reply

The following should be useful to anyone considering the amount of text that must be updated in order to get a lengthy Wikivoyage article to not be treated as duplicate content by Google...

I've been spending some time lately updating, expanding and rewriting content in the Yosemite National Park article in the hopes of escaping the Google duplicate content black hole (we are currently between 28 and 30 when using the &filter=0 search option, but do not appear in a normal, unmodified search for "Yosemite travel guide"). In recent weeks I've completely rewritten the "Understand" and "See" sections (Special:Diff/2734969/2750549) and would have expected that would be sufficient to get us appearing in the Google results (in addition to more modest changes others have made), but despite the article having been spidered and updated in Google's cache four days ago on 18-March, we are still getting filtered out of search results. Apparently the bar to clear on an article with significant text is a very high one before Google's duplicate content algorithm will let us through. I'll continue updating the article until it shows up in Google results, but this example should give anyone who is interested an idea of the amount of text that must be unique in order to have Google treat other lengthy articles as original content. -- Ryan • (talk) • 06:07, 23 March 2015 (UTC)Reply

I have the feeling that if the humans at Google knew the WV situation and knew what the machines at Google were doing, we would get some improvement through human intervention. I remember 11 or 12 years ago (it seems) when WP links started appearing high in Google results. I figured (maybe wrongly) that the Googlers had decided WP was worthy and that they should boost it in the results. How can a little non-commercial, public-service site get the Googlers' attention? Nurg (talk) 09:28, 23 March 2015 (UTC)Reply

phab:T54688 was previously created to solicit some help from the WMF folks, but based on the discussion in that ticket it didn't seem like anyone had a direct line to Google. While I know they will assign a punishment that drops the ranking of sites that attempt to cheat, I haven't heard of Google tweaking their algorithm to boost specific sites (see also #Google diddles results for an explanation of why they would want to avoid doing so). If anyone knows differently, by all means let's do whatever we can to boost our rankings, but for now I think the best thing we can do is to ensure plenty of internal links, incoming links and unique content, and to be patient - it takes time, but in the end Google usually figures out what content is best. -- Ryan • (talk) • 14:44, 23 March 2015 (UTC)Reply

Maybe we should rewrite Mountain View (California) so it's not so similar to the other version. Nurg (talk) 08:56, 24 March 2015 (UTC)Reply

Update: I'm not 100% sure of the exact date that it made it into the index (it seems to have been within the last two weeks), but after rewriting, updating and expanding the article significantly Google is no longer relegating our Yosemite article to the duplicate content black hole and we now appear as the #16 result in an incognito search for "Yosemite travel guide"; based on past efforts I expect that position may slowly improve in the coming weeks (Culver City debuted in a similar position and slowly rose to the #4 result without significant additional updates). The diff of all changes made since the launch of Wikivoyage is skewed by the conversion of XML listings to templated listings, but I would estimate that at least 40-50% of the article text is unique to Wikivoyage, which is a distressingly high bar to have to clear, albeit one that won't be quite as hard for other articles that weren't so complete at the time of the fork. For anyone looking to do a similar exercise, here are some things to consider:

• Rewrite, expand, and update content. It is important that entire sections of text are actually rewritten to be unique, instead of simply rewriting sentences using similar wording, as Google apparently red flags the latter as an attempted SEO cheat (i.e. do not just change "The scenic vistas are inspiring" to "The beautiful views will inspire you"). Also note that there is anecdotal evidence that changing text at the beginning of an article will have a greater effect than changing text elsewhere in the article.
• Make sure that there are internal links from other Wikivoyage articles where appropriate. In this case, any article that mentions Yosemite but doesn't yet do so with a wiki link should be updated. Similarly, in the article being rewritten, make sure that the first mention of any destination or travel topic is a wiki link since outgoing links may also make a difference.
• Where appropriate, replace images. If superior images exist on commons, or images that are of equal quality exist, swap them out in order to provide an additional signal to Google that our content is unique.
• If it does not affect usability, update secondary and tertiary headings - for example, in the Yosemite article I changed "Skiing" to "Winter activities". Search engines give extra weighting to content within heading tags, so changing headings may have an outsized effect when Google's algorithms try to figure out if pages are duplicate content or not.

Hopefully at some point Google will recognize that Wikivoyage is the more active and up-to-date site and all of our articles will get bumped to the top of the indexes, but until then the continued evolution of our articles will naturally get those articles out of the duplicate content black hole, albeit it a slow rate. If there are articles that people are particularly interested in seeing in Google results that are currently getting filtered, copying the exercise that was just done for the Yosemite article is an option that should also work. -- Ryan • (talk) • 16:10, 12 April 2015 (UTC)Reply

A brief follow-up to the above comments, as predicted we have slowly risen in search results for "Yosemite travel guide" from #16 to #11. I've added a few more internal links and have made a few minor updates to the Yosemite article in the past two months, but I don't think those changes would be enough to bump us up five spots, so the pattern of articles slowly creeping up the results list once they finally escape the duplicate content filter seems to be a consistent one. -- Ryan • (talk) • 18:43, 18 June 2015 (UTC)Reply

If you just type "Yosemite travel" it appears on the 5th page of the search. ChubbyWimbus (talk) 00:42, 20 June 2015 (UTC)Reply

Strangely, as of a week ago the Yosemite article is back to being flagged as duplicate content (as identified using the "filter=0" param in the Google results URL). The only changes to the article were the additions of warnings about closures in the park (Special:Diff/2814361/2819523), so I'm suspicious that the problem is due to a change in Google's algorithms rather than any changes in our article content, although if anyone has other insights they would be appreciated. -- Ryan • (talk) • 16:13, 13 July 2015 (UTC)Reply

We're now back in 13th position on a search for "Yosemite travel guide" after some additional rewrites and the removal of some duplicate links (some SEO sites indicate that too many links in an article can negatively affect SEO): Special:Diff/2819523/2829233. I suspect that the rewrites were almost solely responsible for getting us back in the index and that the link removal played little or not part, but that's pure speculation based on the fact that it seems like too many links would be more likely to drop our ranking rather than trigger Google's duplicate content filter. -- Ryan • (talk) • 15:25, 28 July 2015 (UTC)Reply

Swept in from the pub

I remember vaguely that a few weeks or months ago there was a discussion in Engvoy that focused on SEO in which one of the participants pointed out that in order to know the current correct/exact Google search engine ranking of specific articles one should not use Google.com (as the results on Google.com are skewed to produce higher relevancy according to each of it's users profiles) but instead use a different website that helps figure out the Google search engine unskewed search results, and thus help figure out the real ranking of any specific article or page in the Google search engine. Does anyone remember what the website is? ויקיג'אנקי (talk) 21:30, 15 January 2015 (UTC)Reply

Please refer to Wikivoyage:Search Expedition and use https://www.startpage.com. --Saqib (talk) 21:38, 15 January 2015 (UTC)Reply

Thanks. ויקיג'אנקי (talk) 13:08, 16 January 2015 (UTC)Reply

Swept in from the pub

Readership/pageviews appears to be increasing per Similarweb

• 4M for WV (internal data gives 9.7M) [34]

• 11.4M for WT

Congrats to all involved. Still some work to do to attract greater readership... Travel Doc James (talk · contribs · email) 19:32, 1 March 2015 (UTC)Reply

Nice! We certainly have a sizable readership, let's continue adding content :-) Nicolas1981 (talk) 02:35, 2 March 2015 (UTC)Reply

Did you notice that the headline number is actually -24% readership (February 2014 - February 2015) ? I'd be interested in more data that would provide yearly averages. --Andrewssi2 (talk) 02:54, 2 March 2015 (UTC)Reply

Andrewssi2, the data in Doc James' link shows that there was an unusual spike in readership in February 2014. Comparisons between last month and January 2014, or December 2014 vs. December 2013, both demonstrate the increases in readership James is talking about. -- AndreCarrotflower (talk) 03:55, 2 March 2015 (UTC)Reply

The range for 2013 is 3 to 5 million a month. The range for 2014 is 2.7 to 9.3 million, but take out the anomaly of the spike in July/August 2014 (not since repeated) and we get a range of just 2.7 to 5.4 million, which isn't really a significant change either up or down.

I would like evidence of a consistent growth in readership statistics (which is just not demonstrated by the stats provided), and it would be great to analyze the underlying data. --Andrewssi2 (talk) 20:50, 2 March 2015 (UTC)Reply

I've found the raw data here. Will take a look sometime this week. Andrewssi2 (talk) 21:32, 2 March 2015 (UTC)Reply

Perhaps the spike in July/August 2014 isn't as anomalous as you seem to suspect, Andrewssi2. It seems equally likely to me that it's due to normal seasonal variation. There was a similar, though smaller, bump in the high summer of 2013, and it bears mentioning that July and August are arguably the busiest travel months in the Anglosphere. It should be interesting to see what happens to those numbers in summer 2015. -- AndreCarrotflower (talk) 04:59, 3 March 2015 (UTC)Reply

I called it anomalous because the rise and drop was very dramatic. I tried to do some analyse on the page view numbers, but unfortunately the data dumps only give viewing numbers by the hour.

I did compare the view page numbers for the hour of 09:00 on August 1st and September 1st, and for those two files the read levels are very similar. I would have expected the numbers to be higher for August to be generally higher and therefore the origin of the spike will be elsewhere in the large number of files. The analysis does not prove an anomaly, but it also doesn't support a general higher rate of readership yet.

Is there a good place n Wikivoyage to further this discussion? --Andrewssi2 (talk) 07:03, 6 March 2015 (UTC)Reply

(indent reset) I do not see any data on our stats.wikimedia that would merit any conclusions as to readership / visits, unfortunately it seems we only have a stable group of editors (i.e. not growing). BUT similarweb suggests that our site's readership is growing counter to intuitive trends (more now than in summer months), and wikitravel's is falling, albeit in line with expected similarity. The data at Alexa, however, show Wikivoyage falling, especially in terms of search engine generated traffic, and us growing, albeit with unstable and very low search engine leads.
WHAT IS IMPORTANT, however, is that e.g. our edits to the article on Hamburg have moved it from the third page when searched for "Hamburg travel" via startpage.com to the second page. And we're far from done, and it's been only a few days. So Google appreciates our work.
I expect the readership to suddenly boom when we reach the critical mass in searchability and content quality. People will simply start referring to us as a good source. According to the social media data at similarweb.org, they already are. PrinceGloria (talk) 11:04, 6 March 2015 (UTC)Reply

Completely agree. We also have a potential change later this year with the Google algorithms that may start ranking our articles on a (hopefully) more advantageous basis.

I'm also thinking that continued work on the 'traditional desktop web browser' format is running counter to the worldwide change to mobile usage. I think more focus there will help promote us as a 'good source' in audience mind share. Andrewssi2 (talk) 21:20, 6 March 2015 (UTC)Reply

Hello I'm new here, but used to use Wikitravel. Another way to look at this data, that says (Similarweb) WikiTravel gets 6m visits/month and Wikivoyage gets 1.3m = since Wikivoyage spooled off of Wikitravel 2.5 years ago, that's around 140 million travelers who have been deprived of the work of the editors here. Sad but true. Maybe we should be working on both sites? I am Harlan. Harlan888 (talk) 01:50, 7 March 2015 (UTC)Reply

Well, 2 issues I guess. There were various reasons why we split off from WT, (obnoxious advertising, insanely slow response times, lack of positive engagement by new management, etc) that we wouldn't want to return to. The other is that working on both sites will work against us in terms of search engine results since we would no longer be distinctive. There are a small amount of contributors who do work on both sites which we should allow but I wouldn't want to encourage. --Andrewssi2 (talk) 06:32, 7 March 2015 (UTC)Reply

They like launching lawsuits against their volunteers and the people who run the site are clueless with respect to wikis (I met them). They have run many communities into the group. Travel Doc James (talk · contribs · email) 17:00, 7 March 2015 (UTC)Reply

Ok thank you I had read most of that here. But I am thinking of the readers, who are being served mostly at Wikitravel and are missing our edits here. Harlan888 (talk) 19:56, 9 March 2015 (UTC)Reply

Well at some point in time they will notice that one page is better. If too many people continue keeping the Zombie that is that other page alive, that point in time gets put further and further into the future. But that is just my two cents. Hobbitschuster (talk) 23:17, 9 March 2015 (UTC)Reply

WT is too far behind, trying to maintain it is hopeless. WV has much better tools (like dynamic maps to check locations for instance) making editing enjoyable. Duplicating our work over there would be a pain, and as Hobbitschuster points out, it would only delay the death of WT by a few years. Nicolas1981 (talk) 07:20, 10 March 2015 (UTC)Reply

I think we also need to stop fixating on WT as the competition, and just wish it well on its own very separate journey. It is unfortunate that it continues to attract a large proportion of our potential readership and casual editors, and maybe disappointing that our readership isn't actually rising markedly, but WV is getting better every day and hopefully we are all in for the long haul. Andrewssi2 (talk) 09:03, 10 March 2015 (UTC)Reply

Interesting article on a new search ranking system to be potentially released by Google.

http://www.newscientist.com/article/mg22530102.600-google-wants-to-rank-websites-based-on-facts-not-links.html#.VPOoN_mUfwk

It will rank pages around 'facts not links', which I understand will be more judgmental about the quality of a page (determining 'facts') rather than the number of incoming links.

If this is true, it may potentially have a massive positive impact on WV at the expense of WT. Andrewssi2 (talk) 00:05, 2 March 2015 (UTC)Reply

Probably not massive, as most Wikivoyage articles are still quite similar to their Wikitravel counterparts, aren't they? Ikan Kekek (talk) 00:40, 2 March 2015 (UTC)Reply

I just thought 'massive' would be WV achieving a similar ranking as WT, considering that on our current growth we still have some way to go before we reach that milestone. Andrewssi2 (talk) 00:47, 2 March 2015 (UTC)Reply

I see your point. What's interesting is that since there are non-wiki sites that mirror Wikitravel, they will probably be highly rated, too. Ikan Kekek (talk) 05:57, 2 March 2015 (UTC)Reply

I doubt this affects us much, but it seems worth pointing out. A Slashdot story reports that the Federal Trade Commission says "Google boosted its own services for shopping, travel and local businesses by altering its ranking criteria and "scraping" content from other sites. It also deliberately demoted rivals."

I cannot read the Wall Street Journal story that links to; I'm currently in China without a VPN and the Great Firewall currently blocks WSJ. Perhaps someone else could have a look. Pashley (talk) 01:45, 21 March 2015 (UTC)Reply

Swept in from the pub

I have been interested for a long time in knowing more about the extent of increase to an article's page as a direct result of starting to appear in the first page of Google search results.

Earlier today I decided to go over a handful of prominent star articles on Hebvoy, which have been significantly expanded over time, and check if there is a correlation between the page in which the link to these articles appear in Google's search results and the actual amount of page views each of these article got in March 2015.

To my surprise, I discovered that there was no difference what so ever between the amount of page views each of these articles got - whether links to them appeared in the first page of Google search results, or whether they did not appear at all in any of the first 20 pages in Google's search results.

Article name	Article size (as of April 2, 2015)	The page number in Google's search results in which the link to this article appears (to check this I used the startpage.com unskewed and non-personal based Google results)	Actual page views article had during March 2015
Alps	89,596 bytes	Prominently appears in the first results page when one searches the name of the location in Hebrew at startpage.com (which is actually unskewed and non-personal based Google results)	259
Andorra La Vella	32,288 bytes	Prominently appears in the first results page when one searches the name of the location in Hebrew at startpage.com	89
Haifa	92,317 bytes	Still is not included in the first 18 results pages in startpage.com	173
Tel Aviv	57,053 bytes	Still is not included in the first 18 results pages in startpage.com	233
Jerusalem	65,949 bytes	Still is not included in the first 18 results pages in startpage.com	316

How do you explain this? Shouldn't the links that appear on Google's first search engine result page sent a significantly higher amount of readers to those articles? ויקיג'אנקי (talk) 18:51, 2 April 2015 (UTC)Reply

Just because an article appears in the first page of Google results doesn't actually mean people are actually typing that in and searching for it. (i.e. the amount of readers actually seeing that Google search screen are also probably quite low) --Andrewssi2 (talk) 22:25, 2 April 2015 (UTC)Reply

My 2¥ theory: 90% of the people who enter "Jerusalem" into Google are interested in news or history or religion, because the city is famous for all of this. So Google shows links for that. On the other hand, Andorra La Vella is not famous for news/history/religion: it is known mostly by people who travelled between Spain and France. So Google shows more travel-related links higher. That explains for the difference in page rankings. But when we look at the total number of travellers, much more people travel to Jerusalem than to Andorra La Vella, thus the larger number of views. Also, much more people search for "Jerusalem" than for "Andorra La Vella".

Conclusion: I suggest dividing the number of article visitors by the number of tourists. For instance, divide 316 by the estimated number of tourists who visited Jerusalem in March. The correlation will become more meaningful. Cheers! Nicolas1981 (talk) 06:38, 3 April 2015 (UTC)Reply

Nicolas1981, what you are saying sounds about right (should have given better examples) although I must also add that I did also get the Hebrew article of the United States to the first result page when one searches the name of the location in Hebrew at startpage.com, nevertheless, It doesn't get the extremely high amount of page views per day it should deserve. I have a gut feeling that for some reason, all Wikivoyage websites are not getting the web traffic they deserve because somewhere in Google's servers WV is defined as a "fork" website, and therefore less people are sent to all Wikivoyage websites.

Another thought I have had was that due to the relatively low page views the star articles of Hebvoy are getting due to various unknown reasons (articles which I have put ALOT of work in their expansion in gradual improvement), in order to maximize the number of people getting to Hebvoy from Google (or to any of the other WV editions) it would probably be smartest at this point to have thousands of relatively small and useful articles, than only a handful of significantly expanded star articles whom barely get noticed on Google. ויקיג'אנקי (talk) 07:46, 3 April 2015 (UTC)Reply

Suggest you go to the search expedition first. Otherwise this discussion will just be a repeat of that. Andrewssi2 (talk) 07:58, 3 April 2015 (UTC)Reply

I was told the "Page Views" given on our site stopped showing real numbers months ago. Where are you getting the page views from and are they accurate? ChubbyWimbus (talk) 14:59, 4 April 2015 (UTC)Reply

Working on smaller destinations sounds like a good strategy indeed. Nicolas1981 (talk) 04:10, 6 April 2015 (UTC)Reply

Patience pays off.

Despite no updates to our Letchworth State Park article since last June, we've moved ahead -- well ahead -- of WT's version for nearly every plausible search term. I suspect the fact that their article hasn't been updated since 2009 might have something to do with it.

-- Powers ^(talk) 18:46, 4 April 2015 (UTC)Reply

That's very nice news, congratulations and thanks for sharing! As a general conclusion, I believe that it pays off to give some love to less popular destinations, where competition for top search results is lower, and good content might make more of a difference. I would focus on destinations that are not totally off the beaten track, as they have some chance of actually being searched by a reasonable number of users, and thus we might get some traffic and get those visiting the guide interested in exploring WV further.

Moreover, some of our articles on less popular travel destinations are not only verbatim carryovers from WT, they are often woefully outdated and generally in poor shape, so they could use a good brush-up. OTOH, since the destinations I mean are rarely rife with POIs, updating is not as time-consuming as for a large city. So I hope more people will feel encourage to have a look at one of the oddball destinations and give the guide, or at least a section, a nice makeover. PrinceGloria (talk) 19:03, 4 April 2015 (UTC)Reply

Have seen some improvements with cleaning up outline articles to usable which includes removing lots of old bad links. Is slow work though. --Traveler100 (talk) 20:27, 4 April 2015 (UTC)Reply

Are you ready for Google's 'Mobilegeddon' on Tuesday? Pashley (talk) 16:23, 21 April 2015 (UTC)Reply

I used their site checker tool a few days ago and Wikimedia mobile sites are A-ok. The change should have no negative effect on our search rankings. Powers ^(talk) 18:57, 21 April 2015 (UTC)Reply

Airport articles are a relatively new feature on Wikivoyage and we're missing internal wikilinks for a LOT of the new airport articles that have been created, which has a definite disadvantage for SEO purposes. For anyone interested in helping to rectify the situation:

1. Pick out an airport article (Category:Airport articles) and find its official link (example: "www.dfwairport.com" for Dallas/Fort Worth International Airport).
2. Go to Special:LinkSearch and search for articles that include an external link to airport's official URL.
3. Go to each article and replace the external link with an internal wikilink ([http://www.dfwairport.com Dallas/Fort Worth International Airport] → [[Dallas/Fort Worth International Airport]]).
4. Search engine rankings for the airport article improve since we now have more incoming links to it, and readers (and future editors) can more easily find the article since more articles link directly to it.

If you want to do a bit of extra work, many articles mention airports without including internal or external links, so you can also perform a text search for the airport name or IATA code and then ensure that all articles that mention the airport include at least one wikilink to the airport article. -- Ryan • (talk) • 22:52, 18 May 2015 (UTC)Reply

Swept in from the pub

This data looks promising [35] with a more than 40% increase for all languages over the last year. Travel Doc James (talk · contribs · email) 17:15, 2 June 2015 (UTC)Reply

German Wikivoyage more than doubled their pageviews in March and has kept them at that level for three months. What did they do? Powers ^(talk) 20:22, 2 June 2015 (UTC)Reply

There was a banner proposal around that time over there. But it only affects a small number of pages as of now. Hobbitschuster (talk) 21:48, 2 June 2015 (UTC)Reply

It may be because Google now gives equal PageRank to Wikivoyage and Wikitravel. Kaldari (talk) 23:33, 2 June 2015 (UTC)Reply

The Germans left WT in 2006, so it's unlikely that there'd be much content duplicated between WT/WV in that language now. K7L (talk) 23:59, 2 June 2015 (UTC)Reply

I guess we too should have left earlier too :-) The good results of the German WV is a proof that openness and community wins in the long term. Syced (talk) 07:59, 3 June 2015 (UTC)Reply

Well both pages had longer time to diverge in terms of style as well as content. Plus if I am not mistaken German WT is even more visibly dead than English WT (where a quick look at some pages reveals that Spam and vandalism just crops up everywhere apropos of nothing) Hobbitschuster (talk) 12:25, 3 June 2015 (UTC)Reply

Swept in from the pub

Posting this here rather than at Wikivoyage talk:Search Expedition for greater visibility.

In connection with some travelling I'm planning to do next month, I just ran a Google search for the term "champaign urbana wikivoyage". The first result that came up was WT's Champaign-Urbana travel guide. It was one of those deals where Google decided to "includ[e] results for 'champaign urbana wikitravel' with the clickable option below of "search[ing] only for 'champaign urbana wikivoyage'".

Is this something the SEO folks are aware of? There could very well be many people searching specifically for WV articles yet directed to WT.

-- AndreCarrotflower (talk) 17:49, 8 June 2015 (UTC)Reply

The algorithm at google seems to be severely flawed. Is there anything we can do about issues like that? Hobbitschuster (talk) 18:20, 8 June 2015 (UTC)Reply

Google is filtering the Champaign-Urbana article as a duplicate of the WT article, so what seems to be happening is that the best result for your search query is an article that was filtered out of the Google index as a dup, so Google automatically suggests the article that it thinks is the original. Searching with the same pattern for other articles that aren't filtered as dups (example "Yosemite wikivoyage") seems to work as expected. Someone else may know better, but I don't think there is any quick fix to this problem - it is one that will simply fade away slowly as our articles develop and Google begins to view them as unique content. For those wanting to expedite that process:

• See this thread for a discussion of how much rewriting was necessary for Google to no longer view the Yosemite article as a dup.
• Make sure that there are incoming wiki links (not plain text or external links) from other Wikivoyage articles that mention the destination or travel topic - internal links are a signal to Google that an article is important, just don't abuse that fact by making dozens of same-subject wikilinks in a single article.
• Make sure the corresponding Wikipedia article has a link to the Wikivoyage article - links from Wikipedia are a valuable signal to Google that our article is important.

-- Ryan • (talk) • 19:01, 8 June 2015 (UTC)Reply

Strangely, "champaign urbana voyage" puts us first. Google weights URLs heavily, so whatever synonym algorithm Google uses bumps us for "voyage", bumps WT for "travel", but doesn't seem to use "wikivoyage" in its dictionary analysis of search query terms. -- Ryan • (talk) • 19:09, 8 June 2015 (UTC)Reply

Are we absolutely certain that links from Wikipedia matter? Because all URLs at Wikipedia are set to 'nofollow' (to discourage spam and SEO manipulation at Wikipedia), and Google respects that. WhatamIdoing (talk) 20:45, 12 June 2015 (UTC)Reply

Internal links are not nofollow. Check the link attributes on this link for an example: w:Main Page. -- Ryan • (talk) • 20:56, 12 June 2015 (UTC)Reply

And are we sure that Google follows links in this format ok (without the "http") – href="//en.wikivoyage.org/wiki/Whangarei". Nurg (talk) 02:45, 13 June 2015 (UTC)Reply

MediaWiki inserts the http:// or https:// itself (to match the URL type of the currently-viewed page) before Google sees the page. Interwiki links do not get "nofollow", something which has come up as an issue before with the use of these links to promote for-profit Wikia Inc. Links from Wikipedia do matter; it's a top-ten website and search engines tend to favour its results rather heavily. We also should be doing more to convince Wikipedians to edit here, if we are to grow. K7L (talk) 18:46, 17 June 2015 (UTC)Reply

Anyone know if there is a way to tell how many articles at "usable" status or higher do not have a link from Wikipedia? Nurg (talk) 08:31, 18 June 2015 (UTC)Reply

K7L, are you sure about that? I looked at the page source for a destination article and it was exactly as Nurg quoted: no http: or https:, just the double slash. Powers ^(talk) 14:47, 18 June 2015 (UTC)Reply

Since Wikipedia is on forced-HTTPS now, those templates could be changed to use HTTPS directly. WhatamIdoing (talk) 16:16, 24 June 2015 (UTC)Reply

Special:Diff/1850508/2851079 makes a minor tweak to MediaWiki:Lastmodifiedatby to change "this page was..." to "this travel guide was..." for footer attribution (mainspace only). The word "travel" previously appeared only once in our page frame ("Travel destinations" in the left nav), so I suspect an additional mention might give us a slight bump in searches that include the word "travel". If anyone objects please revert, although hopefully this change is so minor that no one will care.

If there are any further suggestions for making some of our shared content more travel-focused they would be welcome. Such changes definitely shouldn't hurt us in the eyes of search engines, and would also be a usability/branding win in making the focus of the site clearer to new users. -- Ryan • (talk) • 23:19, 4 September 2015 (UTC)Reply

I'm all for including the word in the footer, but strictly speaking our travel guides encompass more than a single page. Powers ^(talk) 23:36, 4 September 2015 (UTC)Reply

True. I've changed "travel guide" to "travel guide page". Nurg (talk) 04:00, 5 September 2015 (UTC)Reply

The pagebanner extension seems to be introducing an issue that has serious implications for SEO. I was interested to see if the <h1> fixes with the new pagebanner improved our ranking, but instead I saw us fall from fourth to seventh in a search for "Culver City travel guide". The culprit seems to be the fact that "travel guide" no longer appears in our <title> tags for pages with pagebanners (per MediaWiki:Pagetitle). I've opened a ticket to hopefully get this addressed as soon as possible: phabricator:T111593. -- Ryan • (talk) • 01:32, 5 September 2015 (UTC)Reply

User:Sumit.iitp has a patch queued already - really, really impressive response time - so this issue should be fixed soon. -- Ryan • (talk) • 06:25, 5 September 2015 (UTC)Reply

This issue now appears to be fixed. -- Ryan • (talk) • 01:54, 10 September 2015 (UTC)Reply

Swept in from the pub

Now I know there is no official policy on this (and there shouldn't be), but we should be aware that there are wikis out there that are not in any way related to the Wikimedia Foundation. If and where authors on this here wiki are active on one or several of those projects, you might wish to check out:

1. whether there are links to that other site
2. whether a link to Wikivoyage would make sense (don't try to shoehorn something in where it shouldn't be)
3. whether the community knows about Wikivoyage and whether a brief information on our existence would be of use.

I don't think this would do all that much to our SEO woes, but it might create a few pageviews or better yet give us new editors who already know (and like) wikis. Hobbitschuster (talk) 21:56, 19 December 2015 (UTC)Reply

Any effort is worthwhile effort. -- AndreCarrotflower (talk) 00:00, 20 December 2015 (UTC)Reply

Indeed. That's why I wanted to raise awareness of this. If any of you has a blog, this might of course also be an avenue worth exploring, but I don't want to prescribe or forbid any specific behavior... Hobbitschuster (talk) 00:09, 20 December 2015 (UTC)Reply

This single search term is currently responsible for nearly 7% of incoming search engine traffic on Wikitravel. That could amount to nearly 20% over here since they get three times the traffic we do overall. So it's a lot. Take a look: WT and WV, just scroll down a bit

I had created that page as a redirect to Nha Trang along with a couple other similar search terms but they got removed. Starting discussion here as suggested. See my talk for previous discussion. Acer (talk) 13:13, 26 July 2016 (UTC)Reply

Based on our discussion at User talk:Acer, I would support reinstating the redirects I deleted to see whether they help with SEO, as a pilot project for Nha Trang only. If they work really well, then we should discuss the policy implications of possibly having similar redirects for lots of destination articles. Ikan Kekek (talk) 13:18, 26 July 2016 (UTC)Reply

I am in favor of having this trial with a predetermined period after which we analyze the results and see where we go from there. Hobbitschuster (talk) 21:35, 26 July 2016 (UTC)Reply

6 months, perhaps? Would anyone else like to make a comment? Ikan Kekek (talk) 02:32, 27 July 2016 (UTC)Reply

Just for the record, I'm not saying this is some silver bullet, but I think there's ṕotential here. In particular this might counter somewhat the penalty we get for not having "travel" in the domain name. Acer (talk)

No-one else seems to have had enough interest to find this thread and comment, so I'm going to go ahead and reinstate the redirects in question. Ikan Kekek (talk) 13:24, 29 July 2016 (UTC)Reply

The following redirects have been restored: Nha Trang travel, Nha Trang travel guide, Nhatrang travel guide and Nhatrang travel. The reason for restoring in each case was "Restored as an experiment, per discussion at Wikivoyage talk:Search Expedition#Nhatrang travel". I would like for the statistics to be checked, with the report put here in 1 month, 3 months and 6 months. If there is a clear and disproportionate improvement by that time, we should discuss whether to extend this experiment to other articles. If the experiment is a failure, we should re-delete these redirects. Ikan Kekek (talk) 13:30, 29 July 2016 (UTC)Reply

That's fair, thanks. PS: I just noticed that the other travel site has disabled access to their stats page, it was still accessible four days ago, now it's not. Maybe they keep an eye here :) Acer (talk) 14:36, 29 July 2016 (UTC)Reply

The experiment might get skewed a bit by other factors such as improvements to the article. Nurg (talk) 23:14, 29 July 2016 (UTC)Reply

Very true, but I don't suppose we want to have a "control group" destination article that's edit-protected. So how would you suggest we proceed? Ikan Kekek (talk) 01:09, 30 July 2016 (UTC)Reply

I'm not sure this will be a problem in practice. If we don't see any improvement then it's clear it hasn't worked. If traffic numbers improve we can always compare against other articles that have undergone improvements in the same period. If it stands out in the crowd we might be on to something. Not exactly foolproof but I think it'll suffice for out purposes here. Acer (talk) 08:57, 30 July 2016 (UTC)Reply

[unindent] OK, this is more than a month later. Acer, would you like to give us a report on results so far? Ikan Kekek (talk) 10:29, 12 September 2016 (UTC)Reply

@Ikan Kekek. Short version, it didn't work. Feel free to remove the redirects. I tracked our google rank for that article and we gained only one position and even that wasn't due to the redirects. Also, and this is a brand new functionality, I actually only found out about it today, the pageview analysys tool now has the ability to automatically check the number of hits in all redirects to a given page. See here. So that closes the case.

Besides the redirects I tested a couple other things as well. I added Wikivoyage templates linking to that article from other Wikipedia languages. That's what caused the modest improvement in Google rank. Too little gain to be worth it compared to the effort involved though. Since all language editions are part of the same project domain they are in effect the same website. Google probably applies some sort of law of diminishing returns to additional links from the same website so the effect in the end is marginal.

The third thing I tried did produce noticeable results, not in google page ranks but in direct pageviews. I added a note at the top of Wikipedia:Bora Bora linking to the article here, It had averages of just four pageviews a day in June, increased to 21 hits/day in the last 30 days after the note went up. So both the interwiki links on the sidebar and the Wikivoyage template at the bottom are not very good at attracting clicks due to their location. Better positioning makes a significant difference. But any change in these would require consensus at WP. Acer (talk) 16:57, 13 September 2016 (UTC)Reply

Thanks for your report. There's no need to rush to delete the redirects, though. It's totally up to you whether you feel it might possibly prove useful later to keep them. We could revisit this again in 2 months if you like. Ikan Kekek (talk) 19:26, 13 September 2016 (UTC)Reply

Alright, I'll keep watching to see if anything changes, though I'm not optimistic. Acer (talk) 12:17, 14 September 2016 (UTC)Reply

I'd be interested in another update on these redirects. Ikan Kekek (talk) 21:47, 6 February 2017 (UTC)Reply

Swept in from the pub

As mentioned in #Update on WP links to WV/WT situation above, there has been an ongoing effort to replace WT links with the corresponding WV links in various-language Wikipedias. The unused WT templates are then nominated for deletion, and are usually speedily deleted.

For some reason, the WT template on Portuguese Wikipédia has been through the full VfD process once or twice, with seemingly no consensus. It's been nominated for deletion again yesterday: w:pt:Wikipédia:Páginas para eliminar/Predefinição:Wikitravel

I'd comment, but I am a bit linguistically challenged (não falo português?) K7L (talk) 04:08, 7 September 2016 (UTC)Reply

I haven't seen Texugo in a while, but Ibaman or Nricardo perhaps would like to help. ϒpsilon (talk) 04:42, 7 September 2016 (UTC)Reply

Yep, I tried renominating it again. It's not being used anymore and it's the only WT template left on any WP so I think they might axe it this time around Acer (talk) 14:57, 7 September 2016 (UTC)Reply

• Yes, I'm around and would love to help. But I'm not a WV:PT administrator (only WV:EN); Maybe I'm of no help here. Texugo is very missed on both languages. Ibaman (talk) 12:06, 8 September 2016 (UTC)Reply
• No need. The VfD has completed and the WT template on pt.Wikipédia has been deleted this time. K7L (talk) 12:39, 14 September 2016 (UTC)Reply

At this point the "expedition" for improving Wikivoyage search results seems to be largely dormant, so I've split the informational bits out into a separate Wikivoyage:Search engine optimization page, leaving the list of expedition members, tasks, and page rank list in the expedition page. Hopefully that's not a particularly controversial action, but if there are disagreements I'm adding a thread here to explain the split and to provide a place for discussion. -- Ryan • (talk) • 23:56, 22 September 2016 (UTC)Reply

Swept in from the pub

I am asking because the search engine rank according to startpage of Nicaragua is quite bad while that for Managua is a lot better, despite what I would perceive more edits to the former than the latter. I would like to see where there is still copied content in those articles to possibly edit it out as best practicable. Hobbitschuster (talk) 20:35, 22 September 2016 (UTC)Reply

"Copyscape", linked from Wikivoyage:Search Expedition#Tools, is somewhat useful although not superb. -- Ryan • (talk) • 20:52, 22 September 2016 (UTC)Reply

Swept in from the pub

Hello

While I am no SEO expert and don't know whether this is an effective use of our time, I think the following is at least worth a shot:

Search for our page (preferably high profile pages like countries or continents) and the page on that other site and then put their URLs into this tool. Then try to bring the percentage figure down by editing the sections that it says are copied. We can see after a few weeks (using anonymous search via startpage) whether our efforts were worth it. Hobbitschuster (talk) 22:21, 25 September 2016 (UTC)Reply

This is one good idea. There are others at Wikivoyage:Search_Expedition and more on its talk page, Mentioning WV often on social networking sites also helps; every time a friend mentions someplace he or she is about to visit or is moving to, post a link & suggest they contribute as well. Adding simple things, like more good photos from Commons or more wikilinks, also helps. Pashley (talk) 23:27, 25 September 2016 (UTC)Reply

Automated editing I would additionally support automated or semi-automated editing that changed out many words and phrases for roughly equivalent ones (e.g. "delicious" for "tasty"). —Justin (koavf)❤T☮C☺M☯ 01:27, 26 September 2016 (UTC)Reply

Please don't change individual words using synonyms - search engines are smart enough to recognize that effort as an attempted SEO cheat and will often penalize the offending site in rankings. When rewriting please rewrite full sections. -- Ryan • (talk) • 01:34, 26 September 2016 (UTC)Reply

@Wrh2: Thanks--I didn't realize this. Makes sense. If someone made a list in userspace of the most-viewed articles that we have which also have the most similar content to Wikitravel, we could start working on massaging those texts (and updating them as well--it's important that high-traffic pages are up-to-date). —Justin (koavf)❤T☮C☺M☯ 01:41, 26 September 2016 (UTC)Reply

I'm seeing some very different results with some of the articles listed under "Article Rankings". I don't have time to update them right now but someone else might want to take a stab at a few. Powers ^(talk) 13:42, 26 July 2017 (UTC)Reply

Are you using plain vanilla google? Because that gives you the search results "you want" and not the search results "everybody gets"... Hobbitschuster (talk) 13:55, 26 July 2017 (UTC)Reply

Don't worry; I am quite familiar with how to turn off personalized results on Google. Regardless, I assure you I was using startpage.com, as specified in the introduction to the "Article Rankings" section. Powers ^(talk) 19:21, 26 July 2017 (UTC)Reply

Okay. Was there a general trend discernible in the results you found? Have we gotten better, has that other place gotten worse? Or were/are results all over the place? Hobbitschuster (talk) 19:31, 26 July 2017 (UTC)Reply

Definitely better results for us. Powers ^(talk) 18:25, 27 July 2017 (UTC)Reply

Nice! Hobbitschuster (talk) 18:31, 27 July 2017 (UTC)Reply

I see that Oregon Trail (created here by Gorilla Jones in 2015, no WT equivalent) is not only the first result on startpage.com for "Oregon Trail travel guide" but is set above the rest as a featured snippet if the same search is made directly on Google. We also do well for "RMS Titanic travel guide" (created 2012, no WT equivalent). Differentiate ourselves from that other wiki and there's no reason why Google wouldn't treat us fairly. K7L (talk) 03:27, 19 August 2017 (UTC)Reply

I was thinking about adding an FAQ section or article to a few of our key articles. You see lots of these travel questions come up on other travel sites, and I think if WV provided text in the form of a question, together with answers and links to our guides, we would get traffic from people typing these common travel questions into their search engine. Would anyone have a problem with my trying this with a few pet articles, and seeing over a few months if it has any effect? Of course, if it is successful, then how to incorporate this into our guides would be the next question, but I feel we don't need to answer that yet. Inas (talk) 22:00, 6 November 2017 (UTC)Reply

I'm not sure I understand what you mean. Do you think creating a Travel FAQ article might be a good idea? Sure. Why not? Do you think having a FAQ on Paris (or the likes) article would be a good idea? Eh. Maybe not. And adding stuff beyond the standard sections to our travel guides can make them pretty cluttery pretty soon. Hobbitschuster (talk) 22:19, 6 November 2017 (UTC)Reply

Right now, I was going to actually add a section to the bottom of about five articles, with "Freqently asked Questions". For Sydney, it could be one that comes up all the time like "Can I use an Opal card on the Manly ferry". This would then link back to the info in the relevant articles. This is purely to see how this effects the search results. I appreciate it's a little messy. If in a few months time, these questions don't rate in a search engine, we just delete them. If they're on the first page of results for these questions, we can then have a discussion of if this could be useful. Inas (talk) 22:56, 6 November 2017 (UTC)Reply

This may be worthy of more eyes. Maybe this discussion should be moved to the pub and then swept here when resolved? Hobbitschuster (talk) 23:00, 6 November 2017 (UTC)Reply

I like the idea but I'm liking it more on travel topics than on destination articles. We would need to answer the question of how to determine which Qs are FA, though. Powers ^(talk) 01:19, 7 November 2017 (UTC)Reply

Perhaps these should be put on the city talk pages. This is where a traveller who genuinely found information missing from the article might ask a question. Talk pages are also scanned by search engines - I just searched for "Barangaroo is a brand new area in the Sydney CBD that has been recently opened to the public as a new parkland and exhibition centre." which on the Talk:Sydney and it came third in Google. AlasdairW (talk) 20:02, 7 November 2017 (UTC)Reply

That's a good idea too. Or a /FAQ subpage. Powers ^(talk) 20:11, 7 November 2017 (UTC)Reply

I think Google really likes question and answers, as they are moving more towards their "AI' search interface. It seeks them out on WP. So if you ask a question that's answered on WP, it will give you the answer immediately below the search, and then a link to the page where it found it. But do we really want that link to be to a Talk page? Why would we want to send searchers to our project pages? Inas (talk) 03:50, 8 November 2017 (UTC)Reply

Hi everyone! I'm fairly new to Wikivoyage and still catching up on everything, so if this has already been discussed please just point me to it, but I was wondering whether anyone here or at Wikimedia has had the thought to reach out to Google to see whether they'd be amenable to flipping off whatever switch they have that downgrades Wikivoyage as a fork. They are assumed by most experts to be boosting Wikipedia results, so they might want to act similarly for us but just haven't given it thought yet. On the flipside, I know Google is starting to move into the travel guide space, so they might not want to help out a potential competitor. But I think it'd at least be worth looking into, since Google is (at least hopefully) trying to give its users the best results, so our goals here are fairly aligned with what theirs ought to be. - Sdkb (talk) 23:11, 27 December 2018 (UTC)Reply

Yes, we have had the thought to reach out to Google, but I don't think we know how to do it effectively. If you know how to contact Google in a way that might get results, please go right ahead. If you are successful, the community will be very grateful to you - and to Google. Nurg (talk) 06:30, 29 December 2018 (UTC)Reply

@Nurg: I have some friends who work at Google, but I imagine the best way to go about this would be to ask someone who works for Wikimedia to do it, since it'll be treated more seriously if it comes directly from the foundation. Looking around, I'm not sure how to tag someone who works there; do you know how to do that? - Sdkb (talk) 08:34, 16 January 2019 (UTC)Reply

See Category:Wikimedia Foundation staff. The most active one at WV is User:Whatamidoing (WMF). Nurg (talk) 09:16, 16 January 2019 (UTC)Reply

Cool; @Whatamidoing (WMF): are you able to help with this or to pass it on to someone who could? - Sdkb (talk) 15:45, 16 January 2019 (UTC)Reply

I can try to find a contact. How certain are we that Google treats Wikivoyage as a fork? Whatamidoing (WMF) (talk)

I'm hearing that Google doesn't manually change its algorithm even to promote its own things, so it seems doubtful that they even have a switch to flip on these two sites. There never was any deliberately chosen penalty, so there wouldn't be anything to remove.

One factor that seems to matter is incoming links (from outside the WMF wikis). Maybe steady promotion on social media would pay off eventually for that?

I'm still looking for ideas and information; I'll post if I hear something useful. Whatamidoing (WMF) (talk) 18:23, 24 January 2019 (UTC)Reply

There have been no edits to the project page (except mine) since 2017. Therefore, I have added the historical template to the top of the project page. --Comment by Selfie City (talk | contributions) 22:35, 4 August 2019 (UTC)Reply