Wikivoyage talk:External links/What to link to

Beginnings

I'm not sure that I'm very happy with the language in here about avoiding external links. I understand that the goal of wikivoyage is to be a travel guide, not a link farm, but I think that on a page with plenty of it's own information a link to a local nightlife guide is probably a pretty useful thing to give to the traveler. That stuff can be kindof hard to google for, and it's not always going to be linked from any kind of official city page. -- (WT-en) Mark 01:45, 18 Jul 2004 (EDT)

Well, while you are right in principle and I consider event calendar links to be the least annoying extlink policy violations, the policy is as the policy is and if we allow one type of exception, we'll go down a slippery slope and have to deal and argue with all kinds of exceptions. Hotel lists are also hard to google for. And few web pages will limit themselves to just events. Then there are several such websites for many popular destinations - which would we endorse? In the end it boils down to - we are a travel guide, not a link directory. We should incorporate information on repeated major events on the appropriate article page; list one time major event in the calendar pages we have here on wikivoyage, and just leave the rest as an excercise for the traveller to find. That may run contrary to a service attitude, but it improves the quality of our database. -- (WT-en) Nils 04:29, 18 Jul 2004 (EDT)

Disagree. Nightlife guides are precisely the sort of thing that Wikivoyage cannot hope to replace, because they change so very quickly. (Another example might be [[Talk:Siem Reap|road conditions i

How exactly do you think the external link policy got there in the first place? Did we have a big dicussion about it, and come to some kind of group conclusion, or did somebody (Evan) draft a policy and stick it there? How is that different from my drafting a clarification and putting it up?

Of course that's beside the point from the main thing which is that my clarification is correct, and is the true policy, so can you please go enforce the correct policy now instead of complaining about it? Thank you, dismissed. -- (WT-en) Mark 15:41, 18 Jul 2004 (EDT)

Mark and I are for the change, you are against. Other opinions? (WT-en) Jpatokal 11:12, 18 Jul 2004 (EDT)

Okay, then I guess we won't put it off. Here are my arguments pro the status quo: If you allow event sites, you also have n Cambodia]].) And as for "leave as an exercise for the traveller to find", isn't The traveller comes first the slogan of this site? Useful information should be pointed out, even if it's on another site. (WT-en) Jpatokal 07:16, 18 Jul 2004 (EDT)

The traveller comes first, but we still have to stick to useful guidelines. And I question whether linking to such sites is a good idea. It's sunday and I really do not feel like getting into an argument today, let's put this off for a few days please. -- (WT-en) Nils 11:09, 18 Jul 2004 (EDT)

I'm going to change the policy to reflect that fact that Wikivoyage is not an events calendar, and therefore events calendars are perfectly acceptable to link to. -- (WT-en) Mark 07:30, 18 Jul 2004 (EDT)

I object to your changing of the policy without giving any time for discussion on a subject that you know there is disagreement on. You are not the dictator of Wikivoyage. Hence I am reverting the policy page. -- (WT-en) Nils 11:09, 18 Jul 2004 (EDT)to allow the flight schedule sites we had a few months back, hotel listings, classifieds, newspapers, magazines, and so on. I mean, they are all useful for travellers, and they all deal with data that changes faster than we can ever hope to cope with. Then add maybe photo album sites, and then, hey, why not link to Other Guides as they are also "useful for the traveller".

You are going to convert Wikivoyage into a web directory - a stated non-goal, and rightfully so. You will have the problem on deciding which links of each category to include and which not to. I am of the firm opinion that we really do not need to provide the traveller with the schedules of every small cinema. Similar arguments hold true for non-official websites, like your tales of asia or the thebes mapping project. Allow one in, you open the door really wide for pretty much all links to be added. We'll waste a huge amount of effort on deciding what is "good enough" and what isn't. The current rule is simple, it's easy, and it doesn't cost much time to judge a site's validity. -- (WT-en) Nils 11:26, 18 Jul 2004 (EDT)

I don't think having a few useful links for the traveller risks converting the site into a web directory. Trying to list every possible useful link would of course. So let's not do that, let's just cherry-pick a very few choice things which travellers might find useful from the WWW. -- (WT-en) Mark 15:41, 18 Jul 2004 (EDT)

So, guidelines do two things for us: they help new users get a feel for the site, and they help us resolve conflicts. For me, I think our extlinks policy should at least meet these goals:

• Avoid replacing content with links. We don't want "Go look at this hotel directory extlink", we want a list of hotels. We shouldn't shirk our job in favor of other stuff on the Web; after all, we're not just a Web site.
• Objective way to reject excessive links. One link to someone's personal site about Slobovia is not so bad; 50 links to 50 personal sites is less than useless. But how do we say which to keep and which to remove?
• Avoid becoming a target for Wiki spam.

Our current policy (first sources only) is aimed to meet these goals. but I'm worried we're throwing out some baby with the bathwater. I think that our current policy is making us cut out some useful links, which is probably not all that good. However, I have a hard time coming up with a good way of stating policy that meets all of our goals. --(WT-en) Evan 23:13, 18 Jul 2004 (EDT)

It seems to me that the contributors who know a destination well are in a good position to figure out which one (out of say) five calendar sites is worth linking to. This is part of why having people who feel a strong ownership for a given destination article is valuable.

Perhaps simply limiting the number of external links a good way to go, assuming of course that such a limit is really necessary. -- (WT-en) Mark 03:35, 19 Jul 2004 (EDT)

Numerical limit is simply unworkable, and it solves none of the issues that have been brought up. What might work is to propose external links on the Talk page, then have a vote. Say you need X supporters to add an extlink that is not an official page of some sort. And we need a clear list of what is not allowed - the hotel lists, classifieds sites, and so on. Only, we do not want to maintain an explicit list. There are two problems with this approach: Huge overhead, and the undemocratic attitude and unwillingness to follow established procedure displayed by some people so far would mean that any such process is doomed to failure. The best way, therefore, is to simply keep the current policy of disallowing non-first source links. Without Marks modifications, that he keeps to sneak in. People can always post helpful links on an article's talkpage if they want.

By the way, Mark, why did you slip those modifications back in? What gives you the right to modify policies as you please? I do expect an answer to this question. -- (WT-en) Nils 04:21, 19 Jul 2004 (EDT)

I put the modifications back because it was the right thing to do, and it wasn't very sneaky since I clearly said what I was doing and why, in both the text and the summary. Did you read the bit about how the policy got there in the first place? Have you bothered to think at all?

As for the established procedure: with one really annoyingly nit-picky exception the procedure has been to mostly trust contributors to make good decisions about this stuff, using the rule only when absolutely necessary to get rid of totally obvious spam. -- (WT-en) Mark 04:30, 19 Jul 2004 (EDT)

External link arbitration policy

Here's a novel idea. Stick to the current policy, but draw up guidelines (not a cast-in-iron policy) for exceptions, and approve sites as exceptions if vetted by either an admin (one thumbs-up enough) or a discussion (majority in favor). Once approved, the site will have an invisible but signed HTML comment next to it saying so. Can or not? (WT-en) Jpatokal 02:42, 23 Jul 2004 (EDT)

Theoretical examples:

• I link the Raffles Hotel entry under Singapore to www.raffleshotel.com. Primary source.
• I ask that Tales of Cambodia's road conditions page be added to the Cambodia page, because it offers timely info that Wikivoyage can't replicate and that primary source sites don't provide. Mark approves, signs off on link.
• John Q. Spammer links his personal whoring diary to Bangkok under Patpong without saying why. Link is deleted immediately.

Admins should not ever have "dictator" right on content. It may make sense to elect a sort of ombudsman, but that's a different topic. (Until then, Evan's vote as the maintainer/initiator ought to break any tie.) We should also avoid a situation where votes etc become the norm. Concentrate on writing instead. So let me take a stab at a policy.

I think we should drop the "primary" / "secondary" nomenclature, as it seems to be very ripe to interpretation ("hey that site is MY primary source on this subject, include it already!").

---

• External links should never replace content on Wikivoyage if at all possible. They are used to elaborate only.
• Official sources about the topic of an article are always okay. Examples: Offical homepage of country, city, or a local tourism bureau.
  • For itinaries, a link to the company providing the service should be included. It may make sense to link to several companies for very popular services. Do duplicate the essentials onto the Wikivoyage article (similar to our current restaurant, hotel entries) so that wikivoyagers can just print out one website to take on the road. Examples: Ferry service, overland bus service, "tour of region X" articles and companies.
• Other online travelguides are a big no-no. "We should have this information in Wikivoyage".
• Each article should link to its sibling on WikiPedia.
• Websites that have a limited, regional audience should not be linked to. [This takes care of such things as travel agencies, which are only useful for people who live in a specific area.]
• No "yellow pages" or "classifieds" style sites. It is not Wikivoyage's goal to provide a comprehensive list of (for example) all restaurants or hotels in any given location, so there is also no reason to link to such information. Provide a good selection in the article itself instead - complete with links to their official homepages if such exist.
• Do not add a link to a page that is a dupe of an already existing link, or information already covered in the article [this takes care of the problem of too many links on one subject - first come, first serve; replacement of links possible if the new external site is much better than the previous one - slippery slope of conflict between 2 extenral sites though, may result in editwar]
• There is an exclusive "whitelist" of cases where links to non-primary sources are okay. Any official page should have preference over these. [The reason is simply that we would have the slippery slope of deciding what is the best, or we would end up with several links, etc etc. Especially a problem with commercial sites where a competitor may feel disadvantaged by our inclusion of another site.]
  • Sites that serve event schedules about a location.
  • Sites that serve de-facto as the "definite" homepage for a particular location as no official page exists.
• Dispute policy:
  • Only delete links explicitely disallowed under the policy, discuss all others first. You can still delete if there's no objection in a reasonable time.
  • If someone removed your link, and you disagree, don't simply re-add it. Discuss.
  • Ask for third party opinion. Links are difficult to evaluate.
  • Unsolveable disputes over non-official site links should be solved by a simple vote (registered users only), in case of serious deadlock the site shouldn't be linked to.

---

So the example of the travel asia page (was it about cambodia?) or the thebes mapping project would be a conflict between "no other guides" and "replacement of tourism office page" and okay as long as no such official page exists on the subject. After that, the "no other guides" rule takes over and the link is removed / replaced by the official link. Remember: We want to have information IN wikivoyage, and we are not a web directory (one with only 1 link on a topic is still a web directory).

There is always the possibility for users to stick subjectively helpful links into the talk: pages; the only editing that should be done there is to remove links to legally objectionable material (nazi hate sites or whatever), and blatant spam.

-- (WT-en) Nils 07:53, 23 Jul 2004 (EDT)

Post Scriptum: Note that through the WikiPedia link, we already provide a "gateway" to more comprehensive "technical" / non-travel information on many subjects. -- (WT-en) Nils 07:55, 23 Jul 2004 (EDT)

Shock horror! I find myself agreeing pretty much 100% with everything Nils just wrote above, and I think this would make a good policy. My only quibble is the "...as no such official page exists" bit; I think that, on a case by case basis, good 3rd party sites should be allowed over "official" sites if the official site is completely information-free. (WT-en) Jpatokal 15:00, 23 Jul 2004 (EDT)

I think this is a pretty good set of suggestions. The only thing I feel a bit uncomfortable about is the emphasis on 'official sites'. If we are talking about links in attraction, restaurant, bar, hotel, etc, entries then I 100% agree; the only good thing to link there is a web site 'owned' by the owner of the attraction, etc. But when we start defining official in terms of governments (national or local), tourist information centers, etc then I think we are making a mistake. Such sites can be (a) dull, (b) little use, (c) downright and deliberately misleading (would you trust the Myanmar Government website as your principal informant on minority rights in that country?). I think that however painful it might be to establish the consensus, we need to take our own decisions on what to link to, and not hide behind a plea that it is 'the official site'. -- (WT-en) Chris j wood 19:27, 23 Jul 2004 (EDT)

I dissagree very strongly with one of Nils' points:

• • • If someone removed your link, and you disagree, don't simply re-add it. Discuss.

This makes no sense whatever in that it gives the person doing the deleting dictatorial power, while it requires the person actually trying to add useful information to "discuss". It's quite possible that an attempt to discuss a stricken link will attract little or no attention, and not create an actual discussion. What is the person with the useful link to do then. Remember we need to assume good faith, so if the link is not obvious spam it should go back up.

It makes a lot more sense for to discuss borderline links before deleting them.

A corrallary is that if you are one of the wikivoyageers who go around deleting stuff and somebody puts one of the many things you just deleted back, you should give that person the benifit of the doubt, as they didn't just put back the hundred or so links you deleted, only one. -- (WT-en) Mark 05:07, 24 Jul 2004 (EDT)

Guidelines plus consensus makes agile process

So, I appreciate the spirit behind the extlink policy suggested above. However, I'd like to think that we can make decisions about links without recourse to arbitration committees and numerical voting.

The process that's worked well for Wikivoyage so far has been a combination of guidelines and consensus. The guidelines provide a loose framework for making decisions (some arbitrary, like Project:Spelling, others based on our goals, like Project:first person pronouns). We reach consensus about how to apply the guidelines to individual cases.

What's nice about this is that it doesn't really depend on lots of people being involved with any particular problem. It makes it easy for people to settle arguments ("the guidelines say X"), but it also allows contributors to apply judgement to individual cases and exceptions.

I really don't think that extlinks are sufficiently different from other content on Wikivoyage that we should create an entire process infrastructure around them. So, I'd like to redirect this discussion to how to make guidelines for what to link to in external links, rather than creating new processes for deciding whether an extlink is kosher or not. --(WT-en) Evan 14:30, 24 Jul 2004 (EDT)

Alternative proposal: Allow external links but unpack ASAP

Posted by: (WT-en) Beland 05:45, 25 Jul 2004 (EDT)

I rather disagree with the idea that useful but less-than-ideal external links should be disallowed. Sure, in the long run, Wikivoyage shouldn't link to secondary sources, because it's trying to *be* a secondary source. But it's got to get from here to there, and it should be as useful as possible to readers in the meantime.

I think that any external link should be allowed that is: 1.) relevant to the subject of the article, 2.) useful to travelers or potential travelers, and 3.) not redundant.

However, I think Wikivoyage should retain the long-term goal of "unpacking" most links - extracting the useful information therein and making it part of Wikivoyage. Or even better, writing original content from scratch that makes the links redundant.

Often the first step in expanding Wikivoyage is to notice that it needs expanding, or to find a good external source for information to add to it. Sometimes someone notices that some information is missing, and adds a link to a site that has that information. If we delete that link right away because the information should be part of Wikivoyage itself, then a future editor needs to duplicate the effort of noticing that there's missing information and of finding the source. Also, any readers that come along in the meantime won't have the benefit of getting that information at all, whether it's internal or external. That makes Wikivoyage less useful for them, and they are less likely to come back in the future and thus less likely to contribute any content.

Not everyone has time to "unpack" a link at the time of its discovery, and it's a bit off-putting to have an actually-useful contribution (locating a piece of relevant information) immediately deleted. At the very least, if someone wants to delete a useful link, I think that they themselves should unpack it.

I don't see very much harm to leaving in place external links to content that should be internal, if it's understood that someone should eventually come by and make it internal. It would actually be nice to have a mechanism or convention to indicate this kind of "temporary" external link, to encourage people to make this kind of conversion. (Or it could just be left to the general ebb and flow of things, if people think that would be unsightly.)

When to leave an external link permanently?

We seem to be closer to agreement on what "permanent" external links should be made. But to try to be specific, let me propose leaving up external links to sites that:

• Are "primary" sources or "official" sites of some kind.
• Contain timely information that is updated more frequently than Wikivoyage (non-recurring events, commentary on dynamic nightlife, etc.)
• Do e-commerce.
• Do advocacy. (Like a transit rider's union or something - useful to travelers if they want to file a complaint, or for some other reason.)
• Otherwise can't be integrated into Wikivoyage but is relevant, useful, and non-redundant.

It's useful to keep around "official" links so future editors and readers can keep tabs on both long-term changes and any special events not yet recorded by Wikivoyage. Plus, I think it's expected and it adds a little bit of credibility to the site.

If someone links to their personal page about that provides semi-redundant content or content that should be on Wikivoyage itself, they should be encouraged to integrate it themselves. If they refuse, then someone else should eventually come along and unpack the link and then remove it.

Proposal: Allow and encourage "travel diary" links

As for "travel diaries" and the like, I would actually find Wikivoyage more useful if it did link to random people's travel diaries. It's not necessarily appropriate to integrate this content into Wikivoyage if it's in blog form. But it is nice to see personal pictures and get first-hand descriptions of the people and of particular places. It's also easy to imagine a dozen people visiting the same place and having a dozen different reactions. I think it would be neither easy nor appropriate to try to integrate those reactions into a coherent article. It is appropriate to extract "practical" information from personal diaries, and to construct more consensus-based "best of" lists for articles.

It people want to segregate "travel diary" links, they could easily be collected on sub-pages rather than on article pages directly. (In fact, I think I would prefer that.) It might be nice to connect a list of travel diaries to "ask a fellow traveler" page, where people who had visited a certain area could trade advice and locals could chime in (it's be easy to monitor pages about your home area). Content could bubble up from such fora to the regular articles like FAQs bubble up from other discussion sites.

Linking to lots of travel diaries could help bring more potential contributors if diary writers were encourage to post links going toward Wikivoyage. Doing that would also help more people find practical, non-diary style information (our core content), help readers and diary writers connect with fellow travelers, and incidentally improve this site's Google rank.

For Locals

Someone else wrote: "Websites that have a limited, regional audience should not be linked to." I disagree; what better content to put in a travel guide than what's useful to people going to (or coming from) a particular place?

When I went to LA recently, I took a "Not for Tourists" guide with me, and it was quite useful. Why not make "For Locals" sections in Wikivoyage where appropriate? Some visitors stay in a given area for several weeks or months; why not provide useful info for them? A lot of "welcome to the neighborhood" type information is useful for medium-term visitors, new residents, and travelers alike.

When I was at MIT, I had my e-mail and IM info posted on the web, and I was often contacted by people coming to visit Boston and looking for a cheap place to stay, or having questions about MIT. This is why I think an "ask a local/fellow traveler" page would be incredibly useful. It would also distinguish Wikivoyage from most or all of its competition.

Many localities, especially in the US, already have community bulletin boards. We should certainly link to them, and we need not necessarily duplicate anything which they are doing, like allowing visitors to ask questions of locals. They probably don't have "so I visited there and this is what I saw" collections, which is where Wikivoyage could provide a useful service.

Exception: Other travel guides

Consider the following:

• Potentially every article in Wikivoyage could link to a competitor's article, though linking to multiple competitors would be redundant, and it'd be messy to try to have a redundancy-reduction policy about that.
• Linking to competing sites looks kinda lame.
• When writing an article from scratch, it's probably a bad first step to read what has been written in other travel guides. The best thing to do is probably to start with an empty Wikivoyage template for the type of article you are writing (if one exists), and then fill it with everything you personally know. If you look at another travel guide first, it's tempting to copy the organization and style and maybe even some of the phrases from the other guide. It's easy to do this subconsciously and accidentally introduce copyrighted material into the Wikivoyage, which is a big no-no. If you already have a mostly-written Wikivoyage article, you can go to other travel guides and see if they mention any facts the native article doesn't. With some text already in front of you, it's easier to integrate facts without integrating the original phrasing, and thus avoid copyright problems. I also think that from-scratch articles are a bit fresher and more creative.

For these reasons, I suggest creating a "Wikivoyage:Competition" page, and listing known "general" travel guides which should probably not be linked to on a page-by-page basis, unless there's a good and specific reason to do so. Or, there could be a general policy against linking to "general" travel guides which could remain nameless.

I feel there's a fundamental difference between linking from a Wikivoyage page on London to a Lonely Planet page on London, and linking from a Wikivoyage page on air travel advice to an air traveler's FAQ (that is the best source on the web for this information and which should be integrated into Wikivoyage but which hasn't yet). Just to pick an example totally at random.

It's also a little weird to link from a general travel site to a another general travel site, but it's more intuitive to link to "the best place I could find with this useful information". So even in the absence of a "don't link to general policy guides" policy, I think I "don't delete a relevant, useful, non-redundant link unless you unpack it" rule would be quite workable.

Alternative to the alternative

If the eventual consensus is not to leave "temporary" external links in place, I would ask that instead of deleting them, people move them to the appropriate talk page instead. Basically for all the same reasons, but mostly for the benefit of editors (since most readers won't notice them there). I guess they would basically form a sort of "todo" list on the talk page.

Thanks for reading. -- (WT-en) Beland 05:45, 25 Jul 2004 (EDT)

---

I like Nils suggestion on policy and procedure. It looks to me to be quite workable. I think you have to be really careful about allowing/encouraging external links to anything other than official sources, even on the talk pages. (Especially commercial sources of any type.) You run the risk of spam sites getting their way onto wikivoyage.

One consideration that I think is important that nobody else mentions is ease of finding these external sites other ways. Wikivoyage does not exist in isolation. Most external websites would be trivial to find using google. No traveller is going to be entirely dependent on Wikivoyage to find information. We should not assume that just because we don't link to it travellers wont be able to find it. -- (WT-en) Webgeer 13:29, Jul 26, 2004 (EDT)

I don't really understand the aversion some people have to linking to commercial sites. Writing about commercial entities: bars, airlines, restaurants, etc. is a big part of what a travel guide does. Why not commercial websites if they are useful to the traveler?

It's really easy to tell which such links are spam, but just like email spam, it is hard to explain the difference in a policy. That's why we have guidelines instead, so that we don't have to have black-and-white rules about things like "commercial sites" -- (WT-en) Mark 14:18, 26 Jul 2004 (EDT)

I wasn't suggesting black and white rules, I just want to caution against loosening the current guidelines. Particularly when dealing with 'secondary' commercial websites. I don't have any problem with the current allowance of 'primary' commercial websites. I also wouldn't object to allowing some secondary noncommercial websites (somebody mentioned advocacy sites or even travel stories). However, I am strongly against allowing any 'secondary' commercial websites.

I'm not sure it is always easy to agree on which sites are 'spam sites'.

Would a site that lists night clubs in an area be a spam site? What if it charges money to be in the list, what if it has a large advertising component? What if they add their listing to every city that they have listings for? What if there are 7 or 8 different websites, do we add them all?? Once you allow one of these in, others will rightfully start adding themselves and before you know it Wikivoyage is a link farm. As someone who runs small webpages I can tell you that getting linked to from a page like wikivoyage has a huge benefit (Google Page Rank). (WT-en) Webgeer 15:59, Jul 26, 2004 (EDT)

Absolutely one would not add them all. There probably are 7 or 8 different websites with nightlife information about San Francisco for instance, but anybody who knows anything about nighlife in SF knows that only two of them are worth knowing about, and that those two are almost indespensable if you want to know what's going on. It's for preserving our ability to give travelers that sort of information that it's so important that the policy continue to be to trust those contributors who know a lot about a given place to choose the correct links. This is what we've done so far, with the guidelines as guidelines and nothing more.

Now how do you decide who knows a lot about a given destination? Perhaps an indication is if that contributor contributes actual info to Wikivoyage as opposed to just the same exact link on 5 different pages, which is what spammers do. Is the difference so hard to figure out? What I want to know is why does somebody who hasn't gone through the hassle of spending 6 months figuring out which nighlife guide to say, Lausanne is worth linking to and not just a for-pay rip-off should get to decide that travellers who don't have 6 months in which to do research should be deprived of the info. Some things are not so easy to google, and if we should happen to make the useful commercial sites easier to google then by all means we should, just as we should point guests to those hotels which have given us good service at a decent price. And if a site, commercial or not, stops living up to the standard which got it listed in the first place, like if they sell out (in which case the quality of the listings will drop), then somebody who cares about the destination in question should de-list it.

The main thing is that there is no way that any of us can hope to provide a full events calendar for any one destination, likewise a guide to weather or road conditions.

One of the main reasons for my wanting to contribute to Wikivoyage is that I like to help travelers figure out how to have a good time. Imagine yourself as a traveler who is going to San Francisco and you happen to run into me at a bar in Geneva or Paris. OK, so I can tell you that The Top is a cool place to dance, or at least was two years ago, and that at the time it was mostly a straight crowd, (but gay-friendly) with excellent DJs from Chicago and Detroit. That's exactly the sort of information which we should have here within Wikivoyage. Great, but what does that do to help somebody who's only going to be there on a tuesday and wednesday night during a particular week? Nada. Zip. Nothing.

OK, I can tell them to grab the Weekly and the Guardian, and to try to split the difference, but hey, that means that I've just consigned the traveller to either hope that there happens to be something going on at the place I've mentioned, or that they can find the paper I mentioned and figure out how to parse the listings withought pouring over it for 5 hours in the stupid hotel room.

Isn't it so much simpler just to say "check out S.F. Station, they've never steered me wrong." That way they can also find out about the cool brazilian band that just happens to be in town the same night they are.

I've spent 3 years trying to figure out if there's a similar site for Paris, and haven't found it. All the ones I've found are crap. Meanwhile I've found one for the town I live in now, and I'm thrilled to be able to share it with travellers, just as I am thrilled to share hotel and restaurant info, as well as info about parks, attactions, mountains; just as I am thrilled to be able to share info about toy stores with fellow parent travellers, etc.

So I have no intention whatever to sit back and allow some people who haven't experienced this particular problem of traveling limit this Website. In fact in making those limitations I feel that in some ways such limits could destroy part of the value of Wikivoyage. I won't let that happen if I can help it. -- (WT-en) Mark 18:46, 26 Jul 2004 (EDT)

Spammers frequently add themselves to just a single webpage (eg.contributions of User talk:80.58.35.107); also, tracking down who added a particular link (to see if the link is from a well-known person who we know wouldn't add spam) is hard. It's unclear to me how you would like these "trusted" additions to be distinguished from "untrusted". Maybe add a mandatory section to the talk page providing info regarding who added it, and why it is the Right One to add?

Also, just a comment on some of these extlinks. It is Helpful for links to be embedded into the main article whenever possible. For example, the recent Road Condition extlink made no sense to me because it was just a "useful link" in the extlink section. If it had been in Stay Safe or Get Out with a sentence describing that road conditions change frequently and one should check the Road Conditions site before attempting passage, then it would have been easier to comprehend. With the nightlife guides you propose, if it could be in the See/Do sections with a sentence describing the guide as the place to go for the latest bands/events/whatever that would be very helpful both for the user using the material and for the person taking a second look (it would be obvious that the extlink was pointer to time-sensitive info) -- (WT-en) Colin 19:13, 26 Jul 2004 (EDT)

I absolutely agree that it's better for external links to be embedded in relevant content, and for there to be considerable explaination as to why they are relavant. That's exactly what I always try to do, and that's exactly the sort of link that people have been removing, sometimes along with the paragraph of content in which they were embedded. -- (WT-en) Mark 08:36, 27 Jul 2004 (EDT)

• Spam, spam spam

I think the anti-spam norms we have for talking about commercial entities (restaurants, hotels, airlines, etc) in original content and in making primary commercial links are strong and proper, and that secondary commercial links should be judged (for spamminess) by the same standards.

I don't know if those standards have elucidated in an explicit manner, but with regard to both content and links, I look for usefulness, relevance, and non-redundancy. Spam is a sort of commercial content that tends to be non-useful, non-relevant, and/or non-neutral - at least the latter is an explicit community norm.

Yes, the fact that secondary commercial links are allowed means there's a possibility that a spammy secondary commercial link will be created. But there's also the possibility that someone could create spammy original content, and that innocent travelers might be exposed to it before it could be removed or neutralized. That's just how wikis work; it doesn't matter whether we're talking about content vs. links or primary links vs. secondary links.

-- (WT-en) Beland 03:19, 27 Jul 2004 (EDT)

• "Most external websites would be trivial to find using google."

A good point. Of course, it's also convenient to have public "bookmarks" on a given topic, especially for people that don't know a given site exists, don't know that a given site is actually more useful than its peers, or can't quite remember the name at the moment.

Official sites are a good example. Should the article on Maui link to the official Maui government homepage? I think so. Personally, I could find it on Google in about 30 seconds. My mom, who might someday want to take a vacation there and who is surfing the site on dialup, might appreciate the direct link.

Perhaps less controversial is say, the list of discount airlines. It would be "trivial" in a similar sense to find any given airline's home page once you know it's name, but it's much friendlier to make the name of the airline into an external link that takes you directly to the appropriate site, without fiddling around however many clicks and page views it would take to go through a search engine.

Some people might also not stop to think that a given entity has a website, and instead grumble about having to find the phone number on their own, etc.

-- (WT-en) Beland 03:19, 27 Jul 2004 (EDT)

• Inline external links

I concur that inline external links make sense. On the Wikipedia, external links are rendered distinctively from internal links. Would it be possible to twiddle or upgrade Wikivoyage so that's true here, too?

• The unpacking question

Do we have general agreement that non-commercial content that should be on Wikivoyage, but which is not in a competing "general" travel guide, can be pointed to with an external link until someone has the time to integrate the information properly?

For example, I think my link to Air Traveler's Handbook from Cheap airline travel in North America falls into that category. I also linked to a Google Directory page - after reading the discussion so far, I think the best thing to do would be to keep this until someone has the time to go through and see if there are any useful and relevant sites in that category, list them on their own, and then delete the link.

-- (WT-en) Beland 03:19, 27 Jul 2004 (EDT)

Oh, another good test case just occurred to me. I made some inline links to Orbitz, Priceline, Hotwire, etc. on Cheap airline travel in North America. If I had a single piece of useful advice to give someone about finding cheap airline tickets in the US, it would be "Check orbitz.com." Don't call the airlines, don't try to book 20 different flights on 20 different carriers, and don't sit down with a calendar and try to guess which calendar days are cheapest. Doing all that won't work nearly as well as just using Orbitz, and you'll be grumpy and waste a lot of time. I'm not an Orbitz employee, I'm just a fellow traveller who uses them and is happier and less poor as a result.

I would have to say that an editorial policy that prohibits mentioning Orbitz in this context is too restrictive. An editorial policy that says you can mention good web sites but not link to them, I would have to say is just too arbitrary and a little silly. At least some people seem to think that the current editorial policy does one or both. What guidance would you provide? -- (WT-en) Beland 03:31, 27 Jul 2004 (EDT)

If a link would make for good research information that could be absorbed into Wikivoyage, there's no reason not to put it on the accompanying Talk: page. That's what talk pages are for, after all: helping us develop the guide. --(WT-en) Evan 12:01, 27 Jul 2004 (EDT)

I reckon we are getting far too hung up on anti-spam measures in this discussion. Real spam is easy to spot, and I've seen very little of it on Wikivoyage; from what I see on the recent changes page it tends to get spotted and removed pronto. None of the links that initiated this discussion could reasonably be described as spam. They were all sites that at least one genuine, contributing, wikivoyager thought was useful information for readers.

I'd like to propose the following:

Wikivoyage should be happy to carry any external link, provided:

• it is Useful (to our target audience)
• it is Reliable (the content is not obviously incorrect)
• it is Not Competative (we don't link to other competing travel guides)
• it is information Not Appropriate For Direct Inclusion in Wikivoyage
• it is the Best Information Available (on the subject, to the best of our knowledge)
• the nature of the link is Well Described (a minimum of one sentence of descriptive, gramatical english)

The last criteria sets up a barrier which should ensure that links don't get added without some thought on the part of the contributor.

Thoughts. -- (WT-en) Chris j wood 12:50, 27 Jul 2004 (EDT)

Open Directory twin page links

So, one thing that's worked for us so far is having TwinPages links to similar Wikipedia pages. This has helped us keep focused on travel information rather than extraneous encyclopedic information that might or might not be of practical use to a traveler.

Therefore, I'd like to suggest the following idea: rather than expanding our domain of acceptable external links, we add similar twin pages to Open Directory topics covering the same subject.

For example, the page for Barcelona would have a twin pages link to [[Dmoz:Regional/Europe/Spain/Autonomous_Communities/Catalonia/Barcelona/Barcelona_City/]] , and the page for Minot would have a twin pages link to [[Dmoz:Regional/North_America/United_States/North_Dakota/Localities/M/Minot/]].

As with Wikipedia links, these would end up on a separate part of the page.

In this way, we'd be providing functionality to Web users interested in finding out more about some destination, without adding tons of marginally appropriate extlinks to Wikivoyage guides.

Comments and suggestions welcome. --(WT-en) Evan 12:36, 27 Jul 2004 (EDT)

I don't see this answers the issue. We are not talking about tons of marginally appropriate links; we are talking about directly relevant links that don't meet our current criteria because they are not primary sources.

Incidentally my experience is that the Open Directory isn't (yet) very inclusive. Nor does it supply the kind of on the ground knowledge that should be the strength of Wikivoyage (ie. which of the San Francisco nightlife guides are useful, and which are me-too-marketing dross).

Incidentally my last contribution clashed with yours (Evan) and it has ended up above. -- (WT-en) Chris j wood 12:58, 27 Jul 2004 (EDT)

Wikivoyage is not a Web site rating service. We're writing travel guides. Making and providing Web directories is someone else's job, and we need to defer to them on it.

I'm not 100% crazy about Open Directory, and it'd be great if there were a wiki Web directory we could link to instead (suggestions welcome). Right now, though, Open Directory is the biggest and most complete one. --(WT-en) Evan 20:09, 27 Jul 2004 (EDT)

I like the idea of twink linking to DMOZ. It doesn't solve all the problems being raised by a longshot, but at least it provides a ptr to somewhere to try for those articles which have nothing helpful in them. For example, places which do not yet have Sleep sections. It also helps firm up this slippery slope. -- (WT-en) Colin 20:19, 27 Jul 2004 (EDT)

I like the Open Directory idea-- it would really help avoid long discussions about every link as well as move the responsibility of choosing and maintaining links to a site who has that as their primary goal. I just can't see how else to make simple clear-cut guidelines. And, hey, anyone can also contribute to Open Directory to make their set of links better, right? This seem like a good area to work with others... (WT-en) Majnoona 12:20, 28 Jul 2004 (EDT)

OK, this is implemented for all three languages. See Project:links to Open Directory for how to do it. --(WT-en) Evan 16:34, 7 Aug 2004 (EDT)

Proposed offsite link guidelines

Chris J Wood has offered some preposed guidlines which strike me as being very much like the way I've always thought about these things anyhow:

Wikivoyage should be happy to carry any external link, provided:

• it is Useful (to the traveller)
• it is Reliable (the content is not obviously incorrect)
• it is Not Competitive (we don't link to other competing travel guides)
• it is information Not Appropriate For Direct Inclusion in Wikivoyage
• it is the Best Information Available (on the subject, to the best of our knowledge)
• the nature of the link is Well Described (a minimum of one sentence of descriptive, grammatical English)

These guidlines seem very well thought out, and in fact seem describe what I've thought are the criteria all along. Kudos to Chris for putting this into words (I made one edit, as we know who our core audience is). -- (WT-en) Mark 15:24, 27 Jul 2004 (EDT)

I made some spelling corrections.

I have some problems with these guidelines. First, we aren't "happy to carry any external link". External links are mostly clutter for someone reading a printed guide, so we need to be really selective about them. Even the focus of that phrase is about providing a service to Web sites -- something that's not one of our goals. We're not a Web directory.

OK, so let's scrap that sentence. -- (WT-en) Mark 01:38, 28 Jul 2004 (EDT)

Actually it is quite common for printed (only) travel guides to print URLs of websites, where they think them useful. Such links only become clutter if they fail to meet the other suggested criteria. And before somebody says 'or there are too many of them', I'd point out that in itself would violate the criteria (they cannot all be the Best Information Available. -- (WT-en) Chris j wood 13:24, 28 Jul 2004 (EDT)

Second, I find them really subjective. Subjectivity means conflict -- we'll have to argue about whether a link is useful or reliable, and definitely whether it's the "best information available".

Maybe reducing the thing to a question will help. If the website being linked to were a print publication would it be OK to talk about it? Should Wikivoyage Chicago note that there are up-to-date events listings in the Chicago Reader which circulates on thursday nights at most bars, restaurants, and cornershops? Should Wikivoyage mention that, oh, by-they-way, the Reader website has all the same content as the paper, and if you happen to have webaccess, maybe you should google for the URL, which we won't tell you because we are not a web directory. -- (WT-en) Mark 01:38, 28 Jul 2004 (EDT)

After thinking about your objection to subjectivity for a bit I find myself wondering how a travel guide can possibly avoid being subjective. We don't seem to be avoiding subjectivity in regards to restaurant, bar, or hotel listings. We try to write from a neutral point of view, yes, but how does one avoid subjectivity in deciding which restaurant to list? Is subjectivity good for some things in a travel guide and bad for other things? -- (WT-en) Mark 04:11, 28 Jul 2004 (EDT)

I agree with Mark. By its very nature Wikivoyage is subjective; the only way a travel web site can be non-subjective is by becoming a unselective, hard-fact-only directory type service (yellow pages for travel) and that is very definately not one of our goals. Yes subjectivity leads to arguments, but WikiWiki has mechanisms for resolving those arguments and we shouldn't be afraid of them. -- (WT-en) Chris j wood 13:24, 28 Jul 2004 (EDT)

Finally, I don't think you should need to put in a full sentence to say that the link to the Golden Gate Bridge is http://www.goldengatebridge.org/ .

Actually I put that in because one of WikiWiki's defense mechanisms is raising soft barriers. If someone cannot be bothered to think up a description for whatever is being linked to, then they probably aren't that commited to the link and we probably don't want it. Of course there are a few attractions or whatever that are so universally known that this might be overkill for them, and the Golden Gate Bridge is one such. But such things are relatively few and far between compared to the bulk of what we are writing about. -- (WT-en) Chris j wood 13:24, 28 Jul 2004 (EDT)

That all said, I think this might be a good start to defining exceptions to the primary-sources-only rule. --(WT-en) Evan 19:56, 27 Jul 2004 (EDT)

I think the proposed criteria are too broad. It could easily allow links to a large number of external links on every page. I could see in a single article having links to event calendars, shopping information sites, bed and breakfast listings, restaurant reviews and many others. All of those would be allowed and easily justifiable under those criteria. Once you start allowing these types of links, the precedent will be set and others will in good faith add links to areas they are editing. Many of these types of links are very spammy (Primarily advertising). Right now I don't think we are going to easily come to a consensus on broad wording that can be interperted in many ways. Can I suggest we try to come up with very specific exceptions to add to the current external links policy. I suggest:

Exceptions to the external links policy are:

• Official Government Sponsored tourist websites,
• Official sources of information on the attraction, restaurant, hotel etc... being discussed.
• If a nightlife calendar for the destination is widely accepted it may be included in the drink section,
• Travel topics articles may link to external sites that provided the link
  • is Useful (to the traveller)
  • is Reliable (the content is not obviously incorrect)
  • is Not Competitive (we don't link to other competing travel guides)
  • is information Not Appropriate For Direct Inclusion in Wikivoyage
  • is the Best Information Available (on the subject, to the best of our knowledge)

Where there is disagreement about borderline external links, they should be moved to the talk page until a consensus can be reached.

I used the previously proposed policy for the travel topics articles as I think it is quite appropriate for those articles, and probably easier for wikivoyagers to come to a consesus on the external links on a case by case basis. I suggest that if there are other exceptions that people suggest them for inclusion. I know that a broad policy is prefferred to a list of exceptions, but I think that it is unlikely that everyone will be able to agree on that now. I hope I'm not out of line in the suggestion, but I think we have to try and come to a consensus so we can move on. -- (WT-en) Webgeer 19:12, Jul 28, 2004 (EDT)

>> If a link would make for good research information that could be absorbed into Wikivoyage, there's no reason not to put it on the accompanying Talk: page.<<

True, though there's also an affirmative reason to put it on the main page - so that readers can see it. Given that's likely the only way they'll get this information if it doesn't exist on Wikivoyage itself. -- (WT-en) Beland 21:56, 27 Jul 2004 (EDT)

Linking to DMOZ seems vaguely useful, though I found the existing Wikipedia links to be somewhat buried and easy to miss. -- (WT-en) Beland 21:56, 27 Jul 2004 (EDT)

(WT-en) Beland 23:42, 29 Jul 2004 (EDT) writes:

Well, I could compromise on a relatively strict policy if yet-to-be-unpacked links are stored on the talk page. I don't think travel topic pages should be governed by a less stringent policy.

I think the "exceptions" proposed actually align well with the proposed general criteria. Re-phrasing and tweaking the proposal a bit, it might go something more like this:

---

What to link to

External links posted to article pages should:

• Be Useful (to the traveler)
• Be Relevant (to the topic)
• Be Reliable (the content is not obviously incorrect)
• Be Not Competitive (we don't link to other competing travel guides)
• Contain content Not Appropriate For Direct Inclusion in Wikivoyage or provide services we can't provide.
• Contain the Best Information Available (on the subject, to the best of our knowledge)

Links to content that is appropriate for direct inclusion should be posted to an article's talk page, not the article itself.

Disputed links should also be moved to the talk page.

Additional guidance:

• Links to official sites (government of a city, owner of a hotel, etc.) are useful and appreciated.
• Many "secondary" sources contain content appropriate for direct inclusion, and if so should be linked to only from talk pages, if at all.
• Depending on volunteer commitments, it may not be practical to keep Wikipedia updated with fast-changing information, such as local events or the scoop on a dynamic nightlife scene. External sites that track content like this might not be "appropriate for direct inclusion", and they could be considered for linking by the other criteria.

Include content directly

("Unpack" links.)

A link is not a substitute for actual information. Our goals include creating pages useful as printed guides. So, we need to include information that's at the other end of a link, even if it may seem redundant for on-line use.

For example, in a restaurant listing, get the address, phone number, hours, and prices for the restaurant, even if it's right there on an external Web page. Someone using a printed guide won't have access to whatever's on that page.

You might run across an external site that has useful information that Wikivoyage should have internally, but you don't have time to extract it yourself. (It's probably something more complicated than just a phone number.) In this case, don't link to the site from the article itself, but put a link and a short note on the article's talk page, directing the attention of future editors (possibly including yourself) to it.

That sounds good to me -- (WT-en) Chris j wood 12:53, 30 Jul 2004 (EDT)

Sounds good to me too -- (WT-en) Mark 12:55, 30 Jul 2004 (EDT)

Way too vague. There aren't many extlinks which would fail the be useful test. So unless we really want to be a link farm, these rules won't work. Example: a link to a motel directory is useful because it will include more entries than we want to have here in wikivoyage. That's useful. And directories of various sorts are the #1 spammers we get here. -- (WT-en) Colin 13:17, 30 Jul 2004 (EDT)

I agree with Colin. "Useful" is too vague. Everyone will argue that their website is "useful" and they'll be right-- for someone at some time, but not for Wikivoyage. I think we need to lean towards as few links as possible. I mean Yahoo! started out just trying to list "useful" stuff ;-) And we need to focus on travel reference, not just "Stuff about a place." I mean blogs, movies, webcams, etc all contribute to knowing about a place, but it's not the same thing as a travel guide (Wikitculture?).

But adding reference links to Talk pages for other people to mine seems like a fine idea, we already kinda do that. (WT-en) Majnoona 14:10, 30 Jul 2004 (EDT)

How about "indispensable" rather than "useful". That's actually closer to where I'm coming from with this anyhow. -- (WT-en) Mark 14:17, 30 Jul 2004 (EDT)

I think I see where y'all are coming from. I notice that there's a similar problem with restaurants and hotels and recurring events and shopping places and attractions. With a casual search, I couldn't really find much guidance on how large these lists should get, other than to avoid negative reviews with certain exceptions. The rule "only put in good things" is even more vague than "useful" plus the five other criteria listed for external links. I would think that any policy for determining how comprehensive a list of links can be should have some analog to that for internal content. Lest Wikivoyage become a dreaded "list farm" as well as a link farm. Or is one desirable and the other not?

Or maybe the usefulness or obscurity of a given item (whether link or otherwise) is something that should be decided on a case-by-case basis, using good editorial judgement? It's a complex problem to decide how long a list can reasonably be before it's usefulness is diminished, whether sublists or subpages would be appropriate, what the size of the audience that will appreciate a given item is going to be, etc. Even useful/not useful or indispensible/not indispensible is pretty subjective. Though I guess the resulting arguments would be different, because of the different thresholds of subjective utility. Hmm.

Perhaps it would help to settling on specific language if we had some examples. Are there any examples of items that are "useful" but not "indispensible" we might consider? Personally, I like it when the hardware store people tell me what other stores sell "useful" widgets that they don't, not just the "indispensible" ones. -- (WT-en) Beland 03:17, 31 Jul 2004 (EDT)

I'd like to challenge the assumption that changing our rules on external links in the way described above will somehow turn Wikivoyage into a 'link farm' or 'directory'. This objection seems to crop up in just about every style or policy debate, yet it is never justified by any cogent arguments. Why exactly would allowing links given the above criteria lead us to become a 'link farm'. It seems to me that the criteria 'best information available' means we will only ever end up with a maximum of 1 link per attraction/subject/whatever (that was certainly the intent), and that the criteria 'useful, relevant, reliable, not suitable for direct inclusion, non-competative' will actually mean much less than 1 in practice. In fact my belief is that a rigorous application of those rules rather the current primary/official rules would actually significantly reduce the number of external links, since many of the currently linked 'official' and 'primary' attraction sites are simply marketing eye-candy, or something the attraction owner's 12 year old kid has knocked up after web design 101 at school.

I dispute that the criteria are vague. They are unquestionably subjective and open to interpretation. That is the nature of the subject, and can be leveled at any criteria we come up with, including the current ones.

In fact the illusory quest for rules that will allow us to build a non-subjective travel guide, with no human judgement involved in whether to include something or not, is much more likely to push us into being a directory service, than any of the changes proposed here. A 'source that includes everything that meets its pre-set criteria' is pretty much a definition of a directory. -- (WT-en) Chris j wood 07:36, 31 Jul 2004 (EDT)

Pondering this dilemma more carefully...perhaps what we need is not a tweak to "Useful" but some additional criteria that rule out the "slippery slope" cases that people are worried about.

Be Travel Reference not General Reference. If external content provides general background about a place (its history, what movies were filmed there, etc.), then it's appropriate to either include that content in the appropriate Wikipedia article, or link to it from the Wikipedia. Wikivoyage readers should be directed to the Wikipedia article for all information about a place beyond the scope of travel advice.

Be Travel Links not General Links. Some external links associated with the subject of an article (such as a particular geographic location) are neither "general reference" nor specifically travel-related (links to famous bloggers who inhabit a particular city, links to webcams that have views of that city, etc.). These are appropriate neither for Wikivoyage nor Wikipedia, but you may submit them to the DMOZ project. In general, articles should be linked to the most-general DMOZ category appropriate to their subject.

--(WT-en) Beland 03:08, 2 Aug 2004 (EDT)

So is this silence an indication of agreement? Should this proposal be posted anywhere else before it is adopted? I'll be out of town for a week or so; I guess I can move this to the main page if no one has objected by the time I get back? Maybe everyone will swarm to the table over the weekend. -- (WT-en) Beland 22:59, 5 Aug 2004 (EDT)

Why would we move external link policy to the main page?

Anyways, I really don't feel comfortable with these guidelines. I don't like the idea of having secondary sources linked from Wikivoyage, and I'm having a hard time seeing my way around that. I think a combination of a) links to primary sources plus b) twin page links to equivalent dmoz.org pages is a way better bet. --(WT-en) Evan 00:13, 6 Aug 2004 (EDT)

I mean the main external link policy page, as opposed to this talk page. Is there an example of a "bad" secondary site that's not covered by the amended proposal? -- (WT-en) Beland 13:51, 7 Aug 2004 (EDT)

Actually, I think the burden of proof here is on folks trying to change the rules. We get pretty much everything we want with having only primary links. How is adding a bunch of new links to Web-only resources going to make Wikivoyage a better guide? I find the ideas proposed so far -- blog entries, personal photo galleries, on-line reservation systems like Orbitz, the Thebes site, hotel directories -- distinctly unconvincing. Even Mark's super-great nightlife Web site is pretty uninspiring to me.

What are we missing that the new rules are going to help us get? What is so important that it's worth changing our very objective rule (primary source links only) to something very fuzzy and potentially conflict-causing? --(WT-en) Evan 15:08, 7 Aug 2004 (EDT)

I don't honestly thing that I'm actually arguing for a change so much as I'm trying to preserve an older interpretation of what counts as a primary source. As recently as a few months ago you seemed to agree with this interpretation, at least as indicated by your comments in Talk:Chicago.

As for primary sources, have we really defined what a primary source is? Most of the time in say, history or political science a newspaper is a primary source. Is it in travel? If I point out that a town has a newspaper or two, or a weekly or two does that not make the websites belonging to those newspapers primary sources for the newspapers in the same way that a website belonging to a hotel is a primary source if I mention the hotel?

As for DMOZ, I think it's great -- for a directory -- but we are not making a directory we are making a travel guide. Where we should have one or two links they have dozens. What good is it to turn the travel over to a completely overwhelming directory without at least saying, oh by the way, if you look at the DMOZ link you should check out such-and-such? Anybody looking at a directory is swamped with data. We're trying to prevent that for travellers, no? So why shouldn't we list the one or two primo things to look at on the web and then hand off to DMOZ from there?

Otherwise I don't see DMOZ starting to highlight the one or two links recommended by Wikivoyage for travelers. If they were to do that I suppose I would be totally happy with that arrangement. What the heck, let's try to talk them into it. If we can't, and until we can, I think we should go back to the status quo of six months ago or so, in which contributors who were not obviously spamming should have a fairly wide latitude to provide whatever info they want to the traveller.

One more point, if I may. Lonely Planet certainly wouldn't forbid their writers from mentioning the Chicago Reader, or sfstation, or the squid list, or whatever local up-to-the-minute data source, why should we handicap ourselfs like that? -- (WT-en) Mark 17:03, 7 Aug 2004 (EDT)

You don't find linking to Orbitz from a page giving advice on how to find cheap airfare to be useful? Or do you think it's actually "primary", since going to individual airline sites is not a better alternative? Or do you think it's not worth arguing whether or not it's a good link, even though it's worth arguing over each restaurant, hotel, subjective judgement about safety or character, etc.? No matter what the policy is, I'm prepared to argue that it should be included. We might as well be having an argument based on good criteria instead of ill-fitting ones that are supposedly "objective". It doesn't matter to the traveller whether a link is primary or secondary. If it's valuable, otherwise hard to find, and distinctly travel-related, it's worthwhile to have whether it's primary or secondary. I think "primariness" is a good thing to think about because it highlights the fact that many secondary sites are appropriate for direct inclusion, or that travellers would be better served by linking directly to primary links from Wikivoyage rather than having them go through an extra intermediate step. Especially if those external sites link to yet more secondary sites, which they occasionally and annoyingly do. In the end, I can't really see how the traveller is served by *not* linking to Orbitz, etc., if that's the intention. And I mean, if that *is* the intention, are we allowed to mention Orbitz but not link to it? -- 151.203.122.167 16:18, 7 Aug 2004 (EDT)

External Link Policy

Moved from Project:travellers' pub by (WT-en) Evan

Someone please explain to (WT-en) Pjamescowie what our External Link policy is. He seems unable to read it himself and keeps adding links to Luxor/Western Valley that should not be there. See talk page for our argument. If people disagree with a policy, that policy should be revised; but ignoring an established policy is not the way to do things. (I personally think the external link policy is perfect the way it is, anyway.) I'd remove the TMP link again, but he'll just re-add it so someone else please nuke it. Thanks. -- (WT-en) Nils 18:37, 17 Jul 2004 (EDT)

Obviously, there is more discontent about the extlink policy. Mark had already changed it around (without discussion or consultation - bad example you set there as an admin, Mark. I think I will sneak in a policy "Nils is always right" next time). Anyway, concerned parties should hop over to Project:External links. -- (WT-en) Nils 11:59, 18 Jul 2004 (EDT)

How exactly do you think the policy got there in the first place? Was there a lot of discussion about it before it was written, or was it drafted by somebody and put into place? This is part of what bothers me about blind obediance to policy especially while ripping out stuff which is obviously in good faith: we didn't discuss the policy in the first place, and sometimes folks like you Nils don't seem to think about how the policy fits the goals and non-goals of the site which is really the only binding thing around here.

So I plunged forward and made a tiny clarification to the policy so that the policy obscesed could make better policy based decisions which more closely conform to the goals and non-goals of Wikivoyage. Plunging forward is the name of the game here. -- (WT-en) Mark 15:50, 18 Jul 2004 (EDT)

It was established before I joined Wikivoyage. Likely, Evan or someone simply put it together when Wikivoyage was started. But that still does not give you the right to change the policy without having a discussion first. Sure, someone had to start things, but now we have a community here and it is very impolite to change a policy on your own personal whims. And no, you did not clarify, you added an exception that was simply not there. Please be a good citizen . Rules are there to coordinate the work on Wikivoyage. If the rules do not work well, or not well enough, they must be changed. But you cannot simply change stuff around as you please, creating anarchy on the way. It is especially rude since you know there is disagreement on what the "right" policy should be. I urge you to sleep over it and think about your methods. They are absolutely not acceptable. -- (WT-en) Nils 04:27, 19 Jul 2004 (EDT)

Nils, I've noted before that your tone and your manner is not acceptable, and is basically making me want to try to run you off again. I think you should carefully consider why people contribute to Wikivoyage, and what it means to make a contribution in good faith. I think perhaps you should think about why you need so badly to adhere to rules.

We've had an established way of working here for some time in which we give fairly wide latitude to contributors to make these kinds of decisions about minor points of policy, because the contributors know best. Until now it has been unwritten, but since you like to spend your time adjusting everything on the site to the letter of the policy I find myself wanting to adjust the written policy to reflect what the unwritten policy has been all along. -- (WT-en) Mark 04:39, 19 Jul 2004 (EDT)

External Links - Primary / Secondary Links

Moved from Project:travellers' pub by (WT-en) Evan

Further to this discussion, I'm sorry to admit that Nils and I had a mini edit war regarding the inclusion of a link to the Theban Mapping Project (TMP) on Luxor/Western Valley. I dipped out of the discussion yesterday as it was just going around in circles, but today - in the light of further discussion regarding the relative value of external links - I thought I might invite some arbitration / straw poll on the issue.

Nils wants the TMP removed as he believes it fits the "other guide" category and should therefore be excised immediately. I'm willing to admit that it is another guide (of sorts), but one which the rules could / should accommodate..... Far from being just another travel guide, it is actually a sober, academically-oriented and commercially disinterested site that presents a vast level of detail on the Theban tombs, to the extent that we almost certainly couldn't / wouldn't want to on Wikivoyage (we can't replicate the Flash-based presentation, for example). As a simple link, however, it provides travellers with the opportunity to "dig deeper" on topics of interest if they so desire. Why deprive readers of a choice and an opportunity?

Nils also objects that the TMP site is "secondary" and should therefore be removed under the rules. As I have informed him (and being a professional Egyptologist who has worked at Luxor, I might know.....), this particular site is as "primary" as you are going to get under the circumstances. The TMP holds the official Egyptian Supreme Council of Antiquities concession for archaeological survey in the Theban necropolis - the chairman of the SCA, Zahi Hawass, is falling over the TMP director, Kent Weeks, Professor of Egyptology at the American University in Cairo, with encouragement and support. The TMP is responsible for all the interpretive tourist signing within the Valley AND they're currently excavating a tomb (KV55) within the Valley of the Kings. They are responsible for numerous academic and semi-popular publications on the Theban necoprolis whose proceeds are ploughed back to meet costs and further additional research. Need I say more? This sort of disagreement will happen time and again when writing articles about Egypt on Wikivoyage, mainly because the Egyptians are presently (and unfortunately) lacking the funds, inclination and wherewithal to put up anything similar. There is no Egyptian government website regarding the Theban necropolis..... Couldn't we adjust the policy to solve this apparent dilemma? I'd really appreciate it if interested parties could take a look at the site and give their opinion here on its inclusion on the Luxor pages - we need to sort this out before / if I'm going to continue writing on Egyptian topics. Thanks in advance. (WT-en) Pjamescowie 06:59, 19 Jul 2004 (EDT)

May I suggest a compromise. As someone who has long had an interest in Egyptology, I access the TMP site and it is excellent. Perhaps we could allow a See Also category. This should only be used infrequently with no commercial advertising allowed and would not compromise the External Links policy. Nobody seems to complain about the WikiPedia links on some pages - to me that is pretty much a See Also link already. --(WT-en) Nzpcmad 19:41, 19 Jul 2004 (EDT)

I think "See also" should be reserved for internal links, and that Wikipedia and DMOZ links should go under "External links", because that's what they are. I don't think sorting sites into commercial and non-commercial ones is particularly useful. Useful catgories are topical, and many commercial and non-profit sites cover the same topics. -- (WT-en) Beland 12:53, 31 Jul 2004 (EDT)

On the merits of this dispute...I'm going to go by the new proposal, which we seem to be reaching consensus on. The only dispute remaining is between "useful" and "indispensible", and I would say the TMP site is "incredibly useful." It is pretty clear to my conception of "competitive" that this site is not Wikivoyage's competition. Wikivoyage is not an atlas, and this site is almost exclusively that. Also, the content is not appropriate for direct inclusion, since Wikivoyage is not an atlas, and it is really not feasible to re-create the maps. -- (WT-en) Beland 12:53, 31 Jul 2004 (EDT)

You should probably wait until the actual guidelines page is changed. I dont think we've really reached an agreement on this. --Evan

Inductive method

So, I figured I'd put up some links I think are appropriate and inappropriate for a travel guide, as a way to see if proposed rules will provide a good filter. Feel free to add or remove items from these lists, rather than in replies.

Appropriate:

• Official tourism Web site in the external links section of a city guide
• Official airport site in the "Get in" section
• Restaurant's official Web site in a restaurant listing
• Museum's official Web site in an attraction listing
• Official site of an airline in an air-travel travel topic
• Official site of a hostel network in the Hostels travel topic
• Same-subject links to Open Guides as per Project:Cooperating_with_OpenGuides
• Same-subject links to Wikipedia
• Same-subject links to DMOZ
• Sites where you can buy tickets for things mentioned in Wikivoyage

Inappropriate:

• The blog of a local resident in a city guide
• A blog entry by someone who went to the city in a city guide
• A picture gallery from a personal Web site
• Restaurant review from a newspaper in a restaurant listing
• A Web forum (bulletin board) by local residents in a city guide
• A Google search with the city name as the keyword
• Hotel-finder services
• Restaurant-finder services

Any other examples? --(WT-en) Evan 15:08, 9 Aug 2004 (EDT)

What about the web site of some company that happens to be in that city ? And they have an above-average company tour?

If it's an attraction that deserves mention in Wikivoyage at all, I think it deserves an external link. I know of no "objective" policy that decides the former, except for "avoid negative reviews", which I disagree with. -- (WT-en) Beland 02:39, 12 Aug 2004 (EDT)

OpenGuides Issues

I find the OpenGuide links to be duplicative and not terribly useful for most places. If you look at the one for San Franciso, there's hardly anything there. And they don't exist at all for most places. The one for London has more listings than Wikivoyage, and Wikivoyage has a lot of info not on the Open Guide. But the Open Guide is very much a competing secondary source. Pretty much everything there is appropriate for inclusion in Wikivoyage, in my estimation. (I don't see why comprehensive listings can't be accomodated; they need not be included in a printed version, and all but the "best of" can be segregated to a sub-page - which is exactly what Open Guide does.) Open Guide has a more database-like structure, which is nice in some ways (it lets them do cool things with geolocation and RDF), but Wikivoyage seems friendlier to editors. (Though I think Wikivoyage will soon outgrow, if it already has not outgrown, its article format.) Having two seperate wikis devoted to tourism seems rather redundant to me...it would be nice if the two project merged in a way that preserved both of their best features. Barring that, I think it's a legitimate question whether Wikivoyage should link to OpenGuides, or just suck in all of its content instead. And of course they are allowed to suck in all of our content. If this were done on a consistent and regular basis, either one site would dominate the other or there'll be a lot of mindless copying back and forth that will need to be done (perhaps some part of that could be automated). Another question: if it's OK to link to OpenGuides, shouldn't it also be OK - if not preferred - to link to a site like citysearch.com, which is a much, much better source of information, especially if you're looking for comprehensive listings? -- (WT-en) Beland 01:46, 12 Aug 2004 (EDT)

Proposal: No external links allowed

Crazy, crazy idea: What if all wikivoyage articles were "no external links allowed" ? After all, you can't click on them after they're printed out.

• Sometimes someone notices that some information is missing, and adds a link to a site that has that information. If we delete that link right away because the information should be part of Wikivoyage itself, then a future editor needs to duplicate the effort -- Workaround: don't delete links. Instead, move links (temporarily) to the discussion page, with a "unpack me" comment. After the information is unpacked, move the link to the sibling article on WikiPedia.
• What other problems does this cause ? -- (Can we find workarounds for them as well ?)

--(WT-en) DavidCary 00:22, 10 Aug 2004 (EDT)

David rants for a while

One of my pet peeves is when people casually mention that someone is somewhere on the web, but neglect to give a URI. For example, the Brussels article currently mentions "the Hotel A La Grande Cloche (they're on the web)"

I googled for "Hotel A La Grande Cloche" and discovered that "Most external websites would be trivial to find using google." is incorrect in this case.

I found hotel room reservation information

• http://www.hotel-belgie.nl/Hotel%20A%20La%20Grande-Cloche.html
• http://www.bookings.org/be/hotels/grandecloche?id=300047

Were you thinking of one of those sites, or something entirely different ?

I agree with putting actual information (rather than a link farm) on the paper version of an article ... but if the URI is useless to a traveler with only the paper version, then mentioning that something is "on the web", without the URI, is even more useless.

If you're going to say something is "on the web", please give the URI.

--(WT-en) DavidCary 00:22, 10 Aug 2004 (EDT)

Minimalist approach

Evan stated the following goals:

1. Avoid replacing content with links. We don't want "Go look at this hotel directory extlink", we want a list of hotels. We shouldn't shirk our job in favor of other stuff on the Web; after all, we're not just a Web site.
2. Objective way to reject excessive links. One link to someone's personal site about Slobovia is not so bad; 50 links to 50 personal sites is less than useless. But how do we say which to keep and which to remove?
3. Avoid becoming a target for Wiki spam.

We could construct a set of minimalist guidelines to implement these directly:

• Content Appropriate For Inclusion (CAFI) should be included, not linked to. Links to CAFI may be moved to the relevant Talk page until the CAFI is actually included.
• External links on a main article page should be the "best" links. Links that aren't worth mentioning should be deleted. Links that are worth mentioning but would clutter up the article's link section should be de-cluttered by sorting into categories or by moving them to a appropriate subpage.
• No spam. Spammy links are those that are more self-interested than of interest to the traveller.

"CAFI" is defined by reference to Wikivoyage's content policy.

"Spam" might be defined differently than my very minimalist take here, or it could be a "we know it when we see it" sort of thing.

"Links worth mentioning" could be refined a bit more, either through examples (it's the official site, it has real-time info, it's a good Wikipedia article, etc.) or through principles (useful, reliable, not redundant, best available, not competitor, not if it would be better linked from Wikipedia, not if it would be better linked from DMOZ, etc.). More rigorously defining that can make the decision-making a little more objective but less minimalist. I think it's good to be flexible - sometimes an unofficial site is much better than the official one but not include-able in Wikivoyage, for instance.

I think it might be a good idea to take a "common law" or "bottom up" approach for a little while, because we know that the current policy is not quite right, but we are having trouble constructing a good replacement. This would involve leaving the external link policy intentionally somewhat vague, and deciding on a case-by-case basis which links are good and proper, through voting if necessary. Final decisions and a summary of the reasons behind them can be listed on this page. If we consistently reject certain things and accept others, we can then cofiy those patterns. If apparently contradictory decisions have been made on different pages, then we can reconcile those on a specific and issue-by-issue basis, rather than trying to imagine all possible circumstances a priori. In the long run, we should be able to have a relatively objective, very practical set of guidelines. -- (WT-en) Beland 02:39, 12 Aug 2004 (EDT)

I wonder why this discussion (and several sections above) ended up in no expansion of allowed links policy. Noone interested anymore, or there were some significant changes I missed? --(WT-en) DenisYurkin 03:35, 23 November 2006 (EST)

links to photo communities

Why linking to external PHOTO communities are considered as AvoidLinkingTo?

I believe that communities like TrekEarth.com are both non-competing to WikiTravel and guaranteeing quality by their design. TrekEarth, for example, has very strict rules like OnePicPerDayFromUser, NoFriendsOnThePhoto and some others -- all result to a talented community and works.

What if we have a list of PHOTO communities that are considered GLOBALLY acceptable, for any region? I'd submit and advocate TrekEarth.com to the list. And I believe that for many destinations (like Santorini and Oia), having a link to a great selection of photos would a useful addition for those considering where to go. After all, WikiTravel could not compete with sites like TrekEarth in quality and selection of photos -- by goals, by design, and ultimately in understanding places visually before going them. --(WT-en) DenisYurkin 11:42, 29 Sep 2005 (EDT)

Rail Timetables as Links

Would rail timetables as external links count as official? What if the company is privately owned (more and more government railroads being privatized)?

I ask these questions because it seems the policy may not allow them and yet it would be extremely useful to have - especially for planning trips.

There is another issue here - which i'll post in a sec separately - namely number of links, being a printable guide and how printable guides are used.

Hi Howard! Rail links are primary sources when you are talking about arriving by rail under Get in and are therefore completely appropriate. It's there in the policy. It's the same as linking to the hotel site when you add a listing for a hotel. Have a look at Geneva or Lausanne for examples of how I've done it. -- (WT-en) Mark 15:50, 30 April 2006 (EDT)

What about www.seat61.com? Seat61.com is an excellent resource for rail travel around the world and is maintained by a rail fanatic, who ensures the info is accurate. How could this site be included into the guide? 70.49.6.198 17:38, 30 April 2006 (EDT). 70.49.6.198 17:39, 30 April 2006 (EDT).

Generally no. The seat61 site is not a primary source since it's not maintained by an actual railroad. I'm not sure if there's a place for links like this on Wikivoyage. If it belongs anywhere it would probably be Rail travel in Europe, but I'm not sure I'd put it even there. -- (WT-en) Mark 17:45, 30 April 2006 (EDT)

Hmm, thanks for getting back Mark. This is getting more interesting and yet more troubling. I'm going to start a new topic called "Incredibly Useful Specialist Sites - Should we link?" and continue this there. 70.49.6.198 18:32, 30 April 2006 (EDT)

I can't get my name to appear - yet I'm logged in. --70.49.6.198 17:40, 30 April 2006 (EDT)

That's a bug. The line at the top which indicates that you are logged in is sometimes misleading. Try logging out and logging back in again. -- (WT-en) Mark 17:45, 30 April 2006 (EDT)

Thanks. 70.49.6.198 18:32, 30 April 2006 (EDT)

Printable Guide & External Links

The discussion about external links is extremely interesting. - i'll come back to this in a mo...

Hi again. I'd just like to point out that it's a lot easier to follow the discussions on talk pages if everybody signs their comments. You can sign using -- ~~~~ and it will be automatically transformed into your name (or IP if not logged in) and a timestamp. -- (WT-en) Mark 15:52, 30 April 2006 (EDT)

Thanks - still new to this - Howard (WT-en) Howard 16:04, 30 April 2006 (EDT)

Incredibly Useful Specialist Sites - Should we link?

There are sites such as http://www.seat61.com (rail enthusiast has developed and maintains details on routes, costs and ways to travel world by rail - excellent for planning) and http://www.waytorussia.net (very professional info on visiting Russia and CIS) that are resources I would love to tell the world about. Wikivoyage would seem an ideal medium. Looked at from the other perspective, I would love to know of similarly useful and detailed sites for other parts of the world when I'm planning a trip there.

Yet as I understand the policy on external links and as Mark has suggested (see above mini-discussion about Seat61.com in "Rail timetables as links"), Wikivoyage does not currently have a home for these resources. Phrases such as "Throwing the baby out with the bathwater" come to mind. I genuinely appreciate the need for cleaning up Wikivoyage and clearly there needs to be some kind of line drawn. I'm not so sure that the line has been drawn in the right place.

Comments from Mark and rest of the world are very welcome... (WT-en) Howard 18:39, 30 April 2006 (EDT)

Yeah, I took the side you've presented during the original argument way back when. However the general consensus amoung pretty much the rest of Wikivoyagers supported letting somebody else (like DMOZ) do all the linking. -- (WT-en) Mark 18:45, 30 April 2006 (EDT)

I was one of the dissenters though and I think there is still room for well-justified exceptions. For example, for travel in Cambodia by train, Seat61 is the best resource available as there is no primary link -- and as long as their trains look like this, state operator CFC is highly unlikely to put up a website in the foreseeable future. (WT-en) Jpatokal 21:53, 30 April 2006 (EDT)

Language Links: when and when not?

Hi, I'm relatively new here and already have a question to which I could not find an answer in the FAQ and the Pub. Is there a rule about when to put a language link and when not? I stumbled upon a really awfully written article in another language (moreover, it contained hardly any info - clearly a stub, but not tagged as such), and thought it better not to link to it in the article I was editing, but keep an eye on it in case someone there improves it. Is that correct, or should we always link to other language versions of an article even if what they have there is rubbish? My knowledge of the language in question is about level 2 or even 3 on the Babel scale, so I recognized what I saw. -(WT-en) Dezwitser 12:06, 18 September 2007 (EDT)

The short answer is that languages are always linked to as long as they have a corresponding article. Sure, some articles are crappy today, but they might not be crappy tomorrow... (WT-en) Jpatokal 12:26, 18 September 2007 (EDT)