Wikipedia talk:External links

Archives

Sorted by subject /workshop: Discussion of 2006 rewrite YouTube discussions Sorted by date Archive 1: 2003 - 2005 Archive 2: January 2005 - March 2006 Archive 3: April - July 2006 Archive 4: July - August 2006 Archive 5: August 2006 Archive 6: August - September 2006 Archive 7: September - October 2006 Archive 8: October 2006 Archive 9: October - November 2006 Archive 10: November 2006 Archive 11: November - December 2006 Archive 12: December 2006 Archive 13: December 2006 II Archive 14: January 2007 Archive 15: January-March 2007 Archive 16: March-May 2007 Archive 17: May 2007 - June 2007 Archive 18: June 2007 - August 2007 Archive 19: August 2007 - September 2007

Links to map services

Wikipedia currently has around 300,000 articles with geographical coordinates, which all link to a Wikipedia edited list of map services available for those locations. In addition, Wikipedia articles have 18,000 external links to specific map services. These numbers are big enough to have a guideline on what kind of external links location related topics should have.

I assume most of the external map service links are trying to give readers a helpful way to see where the location is. Such links don't however give any information of the location on their own, they're printable only to tell that the online link may have useful information, and they make Wikipedia articles dependent on external services. This is not the case with coordinates, which are general information usable anywhere. Discussions at Wikipedia talk:WikiProject Geographical coordinates may provide further information.

The Wikipedia page the coordinates link to, Template:GeoTemplate, links to over 100 different map services. In addition, next to all coordinates is a link to an entirely free Wikimedia map service. Should Wikipedia articles then have additional links to any external map services? This is also related to the recent guideline change from "Links should be kept to a minimum" to "Links should be restricted to the most relevant and helpful". In light of WP:USEFUL, the change might be a problem. --Para (talk) 18:45, 28 December 2007 (UTC)

Good point. I haven't thought about this much and I'm no expert in mapping services. But for the sake of consistency and fairness, where we have essentially the same need on 300,000 articles we shouldn't be linking to external web services on an ad-hoc basis. Best to have a standardized, predictable way of doing it so that all the users know what to expect and we're not in the position of favoring one advertising-supported commercial service over another. I like the system where the coordinates bring up the template, and the template gives users a choice of which mapping service to use for a display. We should then discourage people from including their own maps (though a special link, or free map on the page might be appropriate if there is some special reason why the geo template is inadequate and that particular map has to be used). Ideally they could set something up in their browser or wikipedia cookie to indicate a preference of one service or another, which could be launched directly from the template as it displays on the page, but that's getting fancy. (all just my opinion, of course) Wikidemo (talk) 20:54, 28 December 2007 (UTC)

Agreed on all counts. The only reason I can think of to include a direct external map link is when the service has unique data that is known to remain static, ie. when the link points to data from a certain date. Another reason could be unique satellite imagery that shows some very specific feature not visible in other services, but the problem with such links is that the contents of the external map services can change at any time without notice, often within months, leading to link rot that can't be detected without human review. There's been discussion on finding and converting the articles that link directly to an editor chosen map service, so if we can agree on a guideline for linking to external map services and the existing links are converted to coordinates, it would be possible to have a bot convert all new map service links to coordinates, with possibly a special template for the links that should not be converted. The idea to have a user set map service preference feature in the tool that shows the link page (currently called GeoHack) is a good one and I think entirely possible. --Para (talk) 23:18, 28 December 2007 (UTC)

There also is discussion about multiple links to map services at Wikipedia talk:WikiProject Geographical coordinates#Geolinks-cityscale. -- SEWilco (talk) 21:50, 28 December 2007 (UTC)

Sorry for having split the discussion here, but I felt it was necessary to have some general opinion from people unrelated to WP:GEO, since some have expressed concerns that a single Wikiproject is dictating policy. It would also be good to be able to consider this issue outside the "it's useful" aspect, and ignore the fact that they're map links. --Para (talk) 23:18, 28 December 2007 (UTC)

I've been working on converting individual map links in articles to coordinates and/or removing them when the article has coordinates already. It's a bit of a big project on my own though, so everyone please take care of a few. More details at Wikipedia talk:WikiProject Geographical coordinates#Map link conversion to coordinates. --Para (talk) 13:37, 12 January 2008 (UTC)

Well it would have been a good idea to get wider agreement before making wholesale changes. As someone who has created hundreds - and maintains thousands - of geo-articles I think that when a reader clicks on a map location they want to see a map, not a confusing selection of hundreds of options on an ugly page. If the Wiki map is good then standardise to that. First priority is ease of navigation for the casual reader; there is no other "great principle" at stake. (Sarah777 (talk) 23:13, 14 January 2008 (UTC))

Removing external map links from articles has been discussed at length on Wikipedia talk:WikiProject Geographical coordinates; over half of the page is about getting rid of all the geolinks. It was announced on everything related to coordinates and the village pump, at least. If you have general comments about external links to services that have dozens of alternatives that can be linked to using the same identifier (coordinates, here), then comment here, but otherwise please read up on the previous WP:GEO discussion and participate there. Wikipedia's first priority is the dissemination of information, and limiting readers to a single or even a few services that may not even work for them is not helpful. Editors should not be making the decision on which advertising supported commercial map service to make available for people reading Wikipedia. They can't all be included in articles, so yes, that one extra mouse click is needed. Anyway, more there. --Para (talk) 00:20, 15 January 2008 (UTC)

If we direct to Wikimapia it isn't advertising. And one extra click to a confusing page is not reader friendly. Never been to the village pump but I reckon I'll decide for myself what to read before commenting here. Ciao. Sarah777 (talk) 20:33, 15 January 2008 (UTC)

You couldn't have chosen a better example of just how wrong it is to make Wikipedia support general use external services by linking articles directly to them. WikiMapia is a site that lives on pay per click advertisements using Google's AdSense. Every time a user clicks on one of the fascinating rectangles overlayed on their map, an advertisement is served where it can best be seen: right in the middle of the screen. All they have to do is to somehow have people come to their site and click. Enter Wikipedia, a global top 10 site with topics matching theirs and thereby making click-through more likely. Some people believe that WikiMapia shares Wikipedia's ideology by serving community contributed material, but their users' contributions are not available for download and reuse in bulk, making it actually very different from Wikipedia. Still, under this misconception, many articles have a prominent link to WikiMapia, sometimes being the only external link at the end of the article. I can't think of any better way for such a site to profit. --Para (talk) 04:04, 16 January 2008 (UTC)

One extra mouse click? If only! :( The useability of those systems is very poor, and this proposal puts the cart before the horse.

I just tested this on Ballyporeen, and when I tried following the links from the co-ordinates, I was presented there with no less than thirteen screenfuls of links. As a reader, that's simply a pain-in-the-neck, and it's depressingly similar to what happens when I follow an ISBN link such as this one.

The effect of all this is that as a reader, I simply don't bother with these links. It's quicker and easier to just copy-and-paste the ISBN number or co-ordinates into one of my favourite mapping services or book catalogues.

I fully support the principle of channeling the geographical links through a centralised system, but unless and until the useability of that system is improved (e.g. by allowing readers to set a preferred map service as a continuing preference), then forcing readers onto that list of hundreds of links is not a satisfactory replacement for a direct link to a map.

The GeoHack system is a great idea, but it's still a bit raw. Hopefully in future it will evolve into something more user-friendly, but isn't there yet. In the meantime it is grossly premature to force the removal of direct links to relevant maps unless and until the generic system is improved to avoid directing the reader to what Sarah777 rightly describes above as a "confusing selection of hundreds of options on an ugly page". --BrownHairedGirl (talk) • (contribs) 22:00, 15 January 2008 (UTC)

You happened to choose a place in Ireland, for which there is no specialized entry, so you didn't get to notice that the info for many countries pops up to the top of the GeoHack page. But, yes, the GeoHack solution is similar to that for ISBNs. Maybe User Preferences should have geo and ISBN preferences. But that's outside the ability of the present tools. -- SEWilco (talk) 01:25, 16 January 2008 (UTC)

Currently Wikipedia users don't have the option for external map service preferences in articles or other lists, so moving from an arbitrary editor chosen service link or a list of links to a list of all available links shouldn't be a problem. If you believe there is a usability issue, please report what it is and let's fix it. People seem to accept that it wouldn't be right for Wikipedia to have links in all book articles to Amazon for example, but can't relate this to other situations where the same services are available from a number of sources. Why is this? --Para (talk) 04:09, 16 January 2008 (UTC)

Para, please read what I wrote above, about thr useability nightmare of a 13-screen list if links when I just want one map.
The comparison with amazon is a red herring: Amazon is trying to sell me the book, but google or yahoo is not trying to sell me the map. And as above, the book finding system stinks too. --BrownHairedGirl (talk) • (contribs) 00:04, 18 January 2008 (UTC)

If you have a specific issue with usability in mind, please say what it is and we'll see what we can do. Are there any Irish map services you would like to see added, for example? Have you ever actually tried accessing a map service through the list, and which service were you looking for? The top global ones are already at the top of the list without any scrolling needed at all. If there are any applicable local services, they are shown before. Hardly anyone needs to scroll down more than maybe one page to find what they're after, so this change really adds just a single additional mouse click. The reason why the other services are still kept on the list is that a service listed for one region may work for another too, especially near the borders, and while we have no way to map the coverage of all services, we have to give the users the chance to go to the service they feel is best. All of Google, Yahoo, Amazon, or some other advertisement supported book database profit from incoming links, from advertisers or by selling products. Wikipedia can't support them selectively. --Para (talk) 00:36, 18 January 2008 (UTC)

Personally I think the Yahoo! maps are the clearest - do THEY (shock, horror) make money when someone clicks a link? If we had a template/format that allows the reader to choose a map or a maze - surely that would do? It would keep the socialists and cartographers happy without punishing the pundit. (Sarah777 (talk) 20:49, 16 January 2008 (UTC))

Yes, they display ads as well. Even if they didn't, there are many editors who prefer using some other service, and even more readers with varying preferences. If all those links or even just the top ones on some undefined metric were included in articles directly, many articles would use less screen space for content than external links. So surely we have to let the readers choose themselves by showing them a list of all available services elsewhere, after the click. GeoHack is a tool that fills the map service list Template:GeoTemplate with the given coordinates, creating map links to all available services. Anyone can edit the template to improve it, and more complicated improvements can be discussed on the talk page. The only problem with this is for editors who are so used to seeing the direct links that the resistance to change can be overwhelming. But that'll pass. --Para (talk) 01:55, 17 January 2008 (UTC)

What on earth makes you think that users will start to enjoy the nightmare of thirteen screenfuls of links rather than one direct link? This is not about resistsance to change: it is about resistance to change to a ssytem which is useability disaster. --BrownHairedGirl (talk) • (contribs) 00:09, 18 January 2008 (UTC)

See my comment above at 00:36: hardly anyone needs to scroll more than one page, and there is an easy table of contents for those who do. If the many pages really are the only issue, then this still seems to me like resistance to change coming from someone who has never tried using the system. --Para (talk) 00:36, 18 January 2008 (UTC)

Para, you really ought to assume good faith. I have tried the new sytem, many times, and I hve described above why it is a complete pain-in-the-neck to use, and it is exceptionally rude of you to simply dismissd my objections on the basis of your entirely false assumption that I have not tried it.
Yes, there are often links to apropriate mapping services the first screenful, but there are dozens of them, and yes, there is atable of contents. My objection is to the extra hasle of having to select from all those options rather than a direct link.

If you don't see thr advantage of a direct link, let me put it this way: why link to any article when there is a search box at the side of the page you are reading? Because it's a lot easier for the reader to click on one link and get the page they want rather than have to choose from a long list of alternatives. The same applies to maps: being forced to make a selection from a 13-screen list is a giant leap bcakwards in useability, no matter how carefully that list is arranged. If you don't understand that this a useability problem, then please have the good manners to accept that other users genuinely do find it a problem. --BrownHairedGirl (talk) • (contribs) 01:40, 18 January 2008 (UTC)

If you think the problem is in my understanding, please try to explain the problem more clearly, because I still don't see where there would be any hassle and can't believe that you would have tried the system. Nobody needs to scroll 13 screens. It would be possible to make it just a couple of pages or even a single one if it's in columns, if we ignore the possibility of cross-region services being necessary, but such a change wouldn't significantly improve the usability. Please write in complete detail where all you have to click and which part of doing that makes it hard to find the service you need. When I click on your GeoHack link above, I get links to 12 different map services on the first screen without scrolling anywhere, and all of them seem to have more information on the location. What is the hassle in just clicking the first one? It might be easier for a reader not to have to do that single additional click, and have a direct link to some random service in the article, but what makes it impossible is that people have different preferences and everyone won't be happy using the service some editor happened to choose. That's forcing your own preferences to everyone reading Wikipedia, and doesn't follow the neutrality and free content principles we should base our work on here. If some people don't care which service they end up in, they can use the entirely Wikimedia run WikiMiniAtlas service, available from the globe icon next to all coordinates. Otherwise, we need to let people make their own choice. The analogy to Wikipedia is flawed, because an encyclopedia article is only related to the topics discussed in it, while with a location and map services the location is related to all the available global and local services. Furthermore, hyperlinks allow Wikipedia articles to be linked to each other without having to use any additional screen space, whereas all the map links need something additional to link from, and such interface elements wouldn't belong in articles. --Para (talk) 02:32, 18 January 2008 (UTC)

The big list of map services is needed for the same reason the big list is needed in ISBN 0-8070-3253-0. Or should an ISBN just link to a specific bookseller or library? The list gives readers several options. -- SEWilco (talk) 02:50, 18 January 2008 (UTC)

The list does not "give readers several options"; it gives them hundreds of options, when all they want is a map. The ISBN interface is broken for exactly the same reason: it offers the reafer too many options to be useful. The principle is great, but the implementation is awful. --BrownHairedGirl (talk) • (contribs) 00:54, 30 January 2008 (UTC)

In reply to para, the first time I tried this for an Irish location, I got sent to this page with zillions of links, and spent more than 30 seconds studying it to figure out what on earth it was all about. Then I went to the list for Europe, but there was no links to the sub-sections, so I scrolled down looking for Ireland, but didn't find an an entry, so I scrolled up and down again a few times in case it was wrongly sorted. Nothing, so I gave up and threw to location name into google maps.

After all that nonsense, I simply didn't bother with the system again for months, until I tried it again. Same palahver, nothing useful, so I scrolled my way back up to the top of the page to see what all this mess was about. After another chunk of reading, I saw a list of map services, tried a few which gave me nothing useful, and went back to manually using Google maps.

I tried it again, after a further break, and this time went straight to the google maps at the top of the list, which was fine.

This system fails the basic test of useability: it should be obvious and simple to use, but it isn't, and most readers simply don't hang around long enough to try to figure out a complex way of achieving a simple task. The defences offered here all the classic defences offered by people who create an interface; they know how the sytem shoukd work, and to them it's logical. The problem is that the refer approaches a system such a GeoHack without the benefit of all that prior know;edge of what the page is trying to do and why it is designed as it is, and in most cases without the aptiejnces tp spend time figuring it out. --BrownHairedGirl (talk) • (contribs) 00:54, 30 January 2008 (UTC)

According to Nielsen, users don't scroll, or only 10% of users bother to scroll. That may have changed since then, and with more sources for just about anything nowadays, the percentage is probably even less now. Why do you scroll? It seems that you are thinking of this from the perspective of someone who just wants one map link, any map link, so why not click on the first "Map" link on the first screen? When it's under a section called "Global", is it not obvious that the service is expected to work globally? If you somehow found a global service that didn't give you anything useful, then such incomplete services should be moved to local sections, but the usefulness depends on the context and you have not said what information you were after. Maybe your perspective wasn't to get any map link after all, but you were expecting to see localised map services after having noticed the table of contents? Would an empty local section for all major regions be helpful for users like you? You should note that the mapsources page has had much more work put into it than the book sources page in terms of functionality and interface design, and so dismissing one based on the usability of the other is counterproductive. --Para (talk) 04:32, 30 January 2008 (UTC)

Again, your response is a classic interface-designer's response: that it's had lots of work put into it, so it must be good. That's an entirely false assumption: what matters is outcomes, not inputs, and both systems retain the same fundamental flaw of offering far too many choices with too little explanation of how they should be used.

You also have not read what I wrote; please re-read it. I did not just start scrolling: I started reading, and followed the most specific link available from the section heading at the top, choosing the one labelled "Europe A-M". I then started scrolling because Europe A-M offered no link to Ireland.

You may think that it should be obvious that "global" will cover everything, but I took it to mean services looking at the globe as a whole rather than detailed maps. if the intention is that "global" should be the default starting point, then the page needs to say so ... but it doesn't.

I was not expecting that following the link to Europe would give me a "localised map"; I was expecting that it would give me a map. Instead it gave me nothing, not even an indication that I should return to the "global section".

You now seem to be asking me to come with a quick solution to all the problems, which I absolutely refuse to try to do now. These things need to be done more systematically: start with proper testing by observing a range of users, then try to redesign the interface to cope with how users respond to it, and test again the various alternatives. One user's solution is always going to be inadequate for the generality ... and most importantly, this discussion is about whether this system should be forced on all users by replacing all other map links. It's not ready for that, and this page should not be used as a device for cobbling together out a few quick-and-dirty fixes. The system is not ready, so set up proper procedures to start fixing it rather than just insisting that it must be used to replace all alternatives.--BrownHairedGirl (talk) • (contribs) 14:09, 30 January 2008 (UTC)

Classic resistance to change; any proposed solution to alleged problems is going to be unacceptable. Despite your opinionated and counterproductive feedback and unwillingness to answer direct questions, I will act on your feedback and add empty sections for regions that have many coordinates on Wikipedia but no local services. This will solve the reported usability issues and tell users that there are no local services for the region, and they won't go wandering to sections for other regions unless they know that there is a useful service that happens to work for the requested location. --Para (talk) 17:01, 30 January 2008 (UTC)

Should an empty region say something like "No map services for this region, use global services"? Or is that instruction awkward or rudely obvious? -- SEWilco (talk) 17:38, 30 January 2008 (UTC)

The second part is painstakingly obvious when the global services are right next to the local section and it doesn't need to be looked up somewhere else. I think "No local map services for this region" is adequate. But is the term "Global systems" (services, sources) really ambiguous? Does it require knowledge on what type of services the first ones on the list are? If so, does it need a heading at all? --Para (talk) 17:57, 30 January 2008 (UTC)

To summarise the above thread, there is support for centralising the links to external map services on a single page, but some editors accustomed to seeing direct links in articles may feel overwhelmed by the amount of alternatives, and will have to use a couple more seconds to choose one of the first links on the list page. Perhaps a Javascript tool can be created for them to get over the worst resistance to change, and rewrite locally the external links that are being deleted. --Para (talk) 12:55, 20 January 2008 (UTC)

No, that's not an accurate summary, and after all my repeated explanations to explain the problems, it is quite dishonest of you to continue to characterise the objections as "resistance to change".

It's not a matter of a "couple of seconds": the first time I tried it, I gave up after over a minute, and if you look at user observation exercises such as those conducted by Nielsen show that readers give up after a few seconds if a page isn't helpful.

The objections, are not about resistance to change, as you repeatedly describe it; the objections are to a system which is great in principle but whose useability stinks, and which will continue to stink until there is some system by which a reader can set preferences to avoid encountering that huge index page unless they choose to go there.

The problem, para, is that you continue to describe the system from the perspective of someone who is used to it, and that's the wrong starting point. The real issue is how it works for someone who has not learnt their way around it, and for those users, it's useless.

Don't get me wrong, this system is a great idea; it's just that it still has such poor useability that it is grossly premature to force its use to the exclusion of all other methods. Javascript tools and other such widgets may indeed help, as you suggest, but don't go enforcing this mess on every article until those tools are in place and have had some decent useability testing (i.e testing by users unfamiliar with the system). --BrownHairedGirl (talk) • (contribs) 00:54, 30 January 2008 (UTC)

I agree with BrownHairedGirl's comment about usability. If it is considered important to give users a choice of map, why not have a default map that should work for most people, and then a monobook feature to replace it with another map service if and when it is desired. (If this arrangement were offered, my guess is that 95% of users would just accept the default). EdJohnston (talk) 03:03, 30 January 2008 (UTC)

There is already a default map link next to all coordinates, try clicking on the globe icon. It gives most people an idea of the location of the article's topic, and from that perspective serves everyone whose browser supports the map view. Anyone not content with it can access the full list of external links, so most of your suggested arrangement is already in place. On the monobook feature: Wikipedia has less editors than readers, and less monobook editors than article editors. Any Javascript solutions that require registration, changing preferences and/or editing javascript, are only for a very specific audience, which in most cases is only a small part of who Wikipedia is for. --Para (talk) 04:41, 30 January 2008 (UTC)

BrownHairedGirl's trials seem to represent a very minor viewpoint, when they contain such pearls as having to scroll through 13 pages to find any map link, which of course is utter nonsense as the most appropriate links are on the first screen. She's going to have a hard time finding usability testing where more attention is given to scanning the entire length of the page, instead of just the first screen.

The map sources list and its software is by no means a new system: it has been in use for about three years already, and it's the number one manually accessed service on the Wikimedia Toolserver. There have been surprisingly few complaints of its usability, despite having had the usual discussion link at the top of the page. That may have been a result of users having been forced to use single editor chosen services, I can't say. Usability is an important aspect to think about when making such a highly used page, and nobody who has followed its development could say that no thought has been put into its usability or that the usability "stinks". Resistance to change.

I have been considering this external links issue from the perspective of all users; those who don't care which service they use, those who always choose the most popular global service, those who always want some other specific service, those who prefer using a service local to a region, those who just want to see the general location of the point of interest, those who want to see detailed imagery, those who are after geographical information different from a simple street map, and many others who aren't served by a single or a couple of subjectively chosen map links in an article directly. --Para (talk) 04:58, 30 January 2008 (UTC)

Once again Para, you are describing the system from the perspective of some who knows how the designers intended it to to be used, rather from the perspective of the user who encounters it. And you continue to make the arrogant assumption that the interface is right, but the user is wrong. If that's the approach followed by those who created the GeoHack system, no wonder its usability is so poor that the only feedback you are getting in this discussion is negative.

Rather than rudely dismissing how I found the usability of the system as "utter nonsense", please have the courtesy to stop and actually read what I wrote about how I used it. (You asked me to describe it, so I did, so read it and comment on what I actually wrote rather than on your first impression of it). I did not just start reading the whole page; I started by following what appeared to be the most specific relevant link, which turned out to be a dead-end. There is not even any link at the top of the page which would take me to a succinct explantion of the designers intend the page to be used :(

Again, you continue to focus on the page's development rather than the outcome: the current interface may be the result of thousands of hours conscientious work, but huge inputs do not mean that doesn't mean that an interface is good — remember Microsoft Bob? — what matters is the user's subjective experience, and in this case you are simply refusing to listen to the user feedback you are getting. It's not just from me: you have on this page three users (me, Sarah777 and EdJohnston) all telling you that the geohack system currently has poor usability, but you dismiss them all even though there is nobody supporting your view.

I'll give you one further example of the mistaken assumptions you make. You mention above that "there is already a default map link next to all coordinates, try clicking on the globe icon" ... and sure enough, there is, except that I never knew that. The overwhelming majority of images on wikipedia articles are a link only to the image, not to any content, so I assumed that this was the same: text links bring you to content, but clicking on an image brings you only to a larger version of that image. I had no reason to think that the globe beside the co-ordinates was anything other than decoration like the flagicon beside country names, so I never even bothered to mouseover to see the tooltip explanation. The default map which comes up when clicking on the globe for Ballyporeen is not at all bad, and in many ways the result of that is exactly what a system like this should be doing: producing a quick and easy one-click link to a map, except that it's fatally undermined by the absence of any visible indication that it has a function rather than being mere decoration. Once again, a good system with the fatal flaw that the designers continue to assume that users will know how it is intended to be used ... which would be a perfectly reasonable understandable mistake if it was not for Para's repeated dismissal here of all accounts of the difficulties experienced by users as "utter nonsense".

I have been quite forceful here, because I am appalled by the arrogance with which all critical user feedback is dismissed is "utter nonsense" ... but I do want to repeat that in essence I think that the idea behind GeoHack (of centralising and automating links to a huge range of geographical services) is a really good one, and that it already contains a lot of excellent functionality. My objection is simply to the attempt to enforce it as the only mapping system before usability problems have been resolved. --BrownHairedGirl (talk) • (contribs) 14:44, 30 January 2008 (UTC)

I am relieved to see this discussion. Like BrownHairedGirl, I perceive numerous useability issues with the geo-links, but I was not prepared to undertake a comprehensive critique of the problem. For starters, I don't think the average user is even aware that clicking on a set of latitude-longitude coordinates (much less on the little globe icon next to those latitude-longitude coordinates) is supposed to bring up a map link. The only reason I am aware of this feature is that I have contributed to articles in which it has been implemented; if I were not a contributor I probably would not even notice the tiny geographic links at the top and/or bottom of many articles. For this and other reasons (many of which have been nicely articulated by BrownHairedGirl), this is a sophisticated technical feature with a lot of wonderful potential that is probably lost on 99% of Wikipedia users. --160.91.24.33 (talk) 16:39, 30 January 2008 (UTC)

What are the "numerous useability issues"? --Para (talk) 17:17, 30 January 2008 (UTC)

Of for for goodness sake, Patra, this is getting ridiculous. After screenfuls of explanations of the usability problems, asking that question is silly. --BrownHairedGirl (talk) • (contribs) 21:28, 30 January 2008 (UTC)

BrownHairedGirl is incorrect in insinuating that I would be the developer or designer of the proposed system, and saying that I don't care of the outcome couldn't be further from the truth. Most of the recent developments are indeed a result of my suggestions after trying to use the system, listening to people's opinions about them, and the community's acceptance of the changes, but I do not maintain it and have had very little to do with the software. I have no idea how it was intended to be used, but half a year ago saw that improvements could be made and since then I have successfully managed to have people implement the requested changes.

Wikipedia doesn't often use icons, but they're not inexistant. Icons are used on many computer applications and websites. When Wikipedia uses images, they are most often in frames, while icons are not. See for example Wikipedia:Reference desk where they work fine. This shows again how a habituated Wikipedia editor such as BrownHairedGirl cannot think from a general perspective, but speaks from her own minor viewpoint only. --Para (talk) 17:17, 30 January 2008 (UTC)

Para, I didn't insinuate that you are the developer; I stated that you are approaching the GeoHack system from the perspective of a developer, by starting from the position of someone who knows how the system is intended to work and is frustrated that ordinary users don't see the problem the same way.

And now you are at it again, dismissing the point about icons with your customary refusal to listen. I can think of no other case where there is an icon on articles in mainspace; there may well be some, but the fact remains that they are a rare exception. Yes, of course icons are widely used elsewhere in computing, but one of the first things a reader learns when using wikipedia is that images are always linked to a larger version of the image, and that clicking on them is pointless unless you want to see a larger version of the image.

Someone who was actually interested in improving the geohack system would now be looking at ways of resolving this problem of the icon's inconsistency with the fest of the wikipedia interface, but all that you seem to be interested in doing is telling the reader that the reader is wrong. Para, try listening for a change and you might start to understand why you are not getting any support for your impose-geohack-now plan. --BrownHairedGirl (talk) • (contribs) 21:28, 30 January 2008 (UTC)

Again to repeat, I am approaching the GeoHack system from the perspective of all the users, by thinking of all the possible uses for geographical information, and making sure the system is usable in most of those cases.

If you feel there is a problem with how the WikiMiniAtlas is displayed in articles, I suggest that you contact its developers who have added the icons to Wikipedia. Again, I have had nothing to do with it, but the fact remains that all Wikipedia articles already have a default map. It could be made more visible, but that's not a concern in light of all the other alternatives. Coordinates are easy to recognise because they are always in the same place at the top right of articles, there's a descriptive icon of what the numbers are related to, there's wikilinked text next to them saying that they're coordinates, and a tooltip that tells the user what's behind the link. If people feel it's necessary, a note could be added to Wikipedia's Help / Getting started / Basic navigation as kind of a key for the page layout, often used in encyclopedias.

Otherwise, people's personal preferences and opinions are of course listened when developing a system, but when they don't block its usage, they should not block the improvement. --Para (talk) 21:42, 30 January 2008 (UTC)

Para, you are listening to only one preference here, namely yours. Nobody is talking of "blocking an improvement"; you and others are quite free to improve geohack and roll it out further, but this discussion is about the something very different, namely your desire to remove alternatives to your preferred system. And the clear consensus here is that at this stage of geohack's development, it would be wrong to do that. --BrownHairedGirl (talk) • (contribs) 21:52, 30 January 2008 (UTC)

The improvement here is the consolidation and standardisation of Wikipedia's linking to external map services. At the moment it's a mess, but we have been working on improving the situation and that's what you're involved in now. The impartial system has been in use for three years, and has been usable throughout that time, though recently more so. When usability concerns are not major and possible resolution won't require major changes, they are irrelevant for the change and can be worked on later. The book sources list is in use now, has been for a long time, and such commercial links from articles are not allowed. Whatever a small minority of editors think of the list, it's only their opinion, but participation in the Wikipedia project depends on all users respecting the basic guidelines of the project. The minority not fond of the change seems to be ignoring those over personal preferences. --Para (talk) 22:21, 30 January 2008 (UTC)

Para, your arrogance and refusal to listen is astonishing, as your latest falsehood of referring to a "minority not fond of the change". In this discussion, there is a large majority opposed to it.

As has been repeatedly said to you by every other contributor to this page, the objection is not about "personal preference" — it is about the usability disaster of the GeoHack system. Ypu didn't agree with that, and asked me to describe in detail why it is a disaster, and I did that ... but you simply dismissed it all as "nonsense".

As to the basic guidelines of the project, you are simply wrong. There is no ban on an article including a limited number of relevant links to external sites, even if they carry advertising. For example, there are squillions of links to newspaper websites, nearly all of which carry advertising, and the test is relevancy.

I can see many advantages in a generalised map link system such as geohack, but only if it has some reasonable degree of usability, and if usability problems are taken seriously. However the persistent refusal of its leading advocate to accept genuine reports of difficulties makes me wonder what on earth this plan is all about. --BrownHairedGirl (talk) • (contribs) 00:26, 31 January 2008 (UTC)

Your stubbornness is astonishing. How can the concepts of information, neutrality, free content and accessibility be so hard to understand and relate to Wikipedia content? Map links will get the same treatment as book source links did, mostly for the same reasons and regardless of any possible further link page developments, whether you like it or not.

If relevance was the only criteria for map link inclusion in articles, we would have to answer everyone's personal needs and preferences, and include all the global and appropriate local links in all the location related articles, because they are all relevant. As you have repeatedly said already, such a list would be unacceptable, especially in articles. On the other end we have the current situation, where single editors are allowed to insert their preferred map service link to be offered to all readers of the article. Luckily the advocates for any single map service aren't aware of this preference, because otherwise we would see edit warring on whose commercial service this top10 website links to. The current situation cannot continue, and we must work on making all map related links on Wikipedia follow the same pattern.

If you had paid any attention to what you are commenting, you would have noticed that the table of contents you were so attracted to has been moved below the fold, following your explanation of the use experience. No other issues have been reported, except your inability to understand the meaning of the word global. Is common sense too much to ask? If the same report comes from multiple users, related to something specific as opposed to a general feeling of having to use something different than before, then it can be given attention and ideas requested for alternative options. Others are dismissed as resistance to change.

The analogy to newspaper websites is flawed, as the sources the encyclopedia gets the information from are irrelevant as long as they are reliable. Map services however have quite distinct user interfaces which people get accustomed to, the data and additional features vary, and with your logic you'd have no basis to stop someone from adding yet another map link to an article. If the distinction is lost on you and you are content with whichever service is provided (first link that says "Map", hello?), I'd recommend for you to abstain from commenting on this any further. --Para (talk) 01:37, 31 January 2008 (UTC)

Para, first of all you say that the system is fine, yet now you say that it is being tweaked it as we talk, and that it's all my fault for not checking every few minutes to see how many improvements have been made. Make up your mind.

You came here to seek consensus for a mass deletion of references, and the clear consensus is that your system is not ready to serve as a replacement for everything. If you were to ask people to volunteer to help in improving the sytem, I am sure that you would find plenty of people ready to offer their time ... but that is a different issue to the question of whether the system is ready now, and it isn't ready. Go do your usability improvements, and come back to seek support, but don't try blaming the objectors for not doing the work of improving the which system you alone want to impose against consensus.

You evidently don't spend much time reading newspapers: the same story can be covered very differently in difft newspapers, and which source is used can be important. Same goes with maps, there are reasons to prefer one over the other in difft situations ... but I would much prefer quick access to any map than an obscure and misleading 13-screenful page of links. It doesn't have to be the ideal map: just give the user a clearly-labelled link to one map. I don't need or want a choice: I just want a map that will let me see roughly where the place is. That's all: just a map, any map, not a choice of hundreds of them. (By all means, also offer me a choice to view a clearly-labelled "list of available map services", but don't confront me with a 13-screenful menu by default when all I want is a map).

(I went to put petrol in my car yesterday. I pulled up at the pump labelled petrol, grabbed the nozzle, put the stuff in, paid and left. I know that there are many many difft ways of formulating petrol, with various permutations of octane ratings and additives such as detergents, but I don't care: I just want petrol that will make my car go ... so it suits me just fine that there is only one petrol pump. If they started offering me a huge menu of difft types of petrol, I'd go somewhere else — heck, I even stopped using a garage when they introduced one new type of petrol and couldn't explain what the new stuff did differently.)

You also refer to my "inability to understand the meaning of the word global". Once again, your rudeness is exceeded only by your failure to read: if you read what I wrote (which are astonishingly reluctant to do), you will see that the problem is that there is more than one possible way of interpreting what the label "global" means in that context. This goes to the core of the problem you have in understanding the objections to your beloved system: every time a reader reports that the interface does not have the meaning you intend, you blame the reader for not guessing the intended meaning. That's a textbook case of how bad interfaces are designed, by blaming the reader rather than trying to develop the interface to clarify the functionality. --BrownHairedGirl (talk) • (contribs) 03:52, 31 January 2008 (UTC)

Links to map services - summary

In the discussion so far, the GeoHack system has been vigorously defended by Para, but and the principle of centralising links has been supported by Wikidemo. However the the only comment from other users has been critical of the usability pf the GeoHack system at this point in its development, so there is no consensus for Para's initial proposal that all other links to map services should be removed. --BrownHairedGirl (talk) • (contribs) 14:50, 30 January 2008 (UTC)

You missed my support of centralizing links, but the stumbling phrasing "but and the principle" suggests an editing problem so you might have misplaced part of your paragraph. Check your draft or notes for whatever else got overlooked. -- SEWilco (talk) 17:45, 30 January 2008 (UTC)

Leaving 300,000 Wikipedia articles free for ad-hoc adding of links to external advertising supported services is out of question; the change must be done despite the minor discordant notes. We need to be consistent on our linking to map services, as such a high number of location related articles makes map links almost part of the Wikipedia interface. The book sources list with ISBNs and external links was taken into use for principled and practical reasons, so coordinates and map service links should be handled the same way. There is no other possible alternative to the centralised list of links, and single external map service links cannot be used for the following reasons:

• Wikipedia is an encyclopedia for the dissemination of information.
  • An external map service link is only information that a single external resource may at some point have had information about the topic, whereas coordinates are information by themselves, usable in any online or offline service.
  • There are very few location related articles that require a direct link to a specific map service, instead of the general coordinate information about the location.
  • External map links are a result of lazy editing: Instead of doing what an editor should do and processing the link into encyclopedic information, it's just inserted to the article as such without bothering to generalise.
• Wikipedia has a neutral point of view and is free content.
  • The selection of map services is a reader choice and editors cannot objectively make that choice for people.
  • Direct links in articles supports the preferences of very few people only.
  • Wikipedia does not sponsor certain advertisement supported commercial map services over all the others.
  • Wikipedia does not depend on external services, which consistent external map links in articles would cause.
• Wikipedia should be accessible to the widest possible audience.
  • Links chosen by single editors reflect their personal preferences, and they may not include accessible external services.
  • Direct links break geographical information into unusable snippets that can only be used online with a modern web browser, unlike coordinates and a full list of services.
  • A single or a couple of external map service links ignore the users' needs, maybe without even the user knowing that other services with data for the location exist. We can't know why people go to geographical information services and which ones they use.

--Para (talk) 16:56, 30 January 2008 (UTC)

Para, you're still doing the same thing as you have done throughout this thread. You are restating the case you make for your objective, and dismissing the consensus that the system you advocate is not yet sufficiently mature to justify the mass removal of other links.

That's all. You have made your case, and there is no consenus for it. --BrownHairedGirl (talk) • (contribs) 21:14, 30 January 2008 (UTC)

That is incorrect. There is no consensus one way or the other on this page, so we can just make a decision based on fundamental Wikipedia policies and guidelines, which I mentioned above. If BrownHairedGirl's little revert warring friend asks to support her position when she is incapable of doing it herself, it is not a separate independent opinion. EdJohnston also made no comment of the usability, it's BrownHairedGirl together with the ip user complaining about changing the status quo, here on an page inappropriate for that purpose. But all that is irrelevant: editors' ignorance of Wikipedia policies is not a reason to stop improvements because of minor usability concerns. This is a matter of principle, and external links to map services other than in the references section must be removed, simple as that. --Para (talk) 21:26, 30 January 2008 (UTC)

Para, you are the only advocate of imposing geohack, and Para's view != consensus. I strongly resent the suggestion that I am not taking an independent view: I frequently disagree with Sarah777. But take a look at this page, and you will see that there are four editors opposing your proposal, which amounts to a consensus against.

As to wikipedia policies, there is no policy which requires the removal of a limited number of relevant external links, and there is no policy which requires the reader to be deprived of useful links to information in favour of a well-intentioned but system with atrocioius usability, whose advocates dismiss all objections as "nonsense". You consistently refuse to listen to any of the criticism of the system, and wilfully misrepresent objections — most recently with ip contributor who wrote of "numerous useability issues with the geo-links", but who you falsely claim was merely "complaining about changing the status quo".

That's all: you floated a proposal, and the consensus is against you. Of course, if and when the GeoHack system is improved, the consensus may change — and in principle I'd be delighted to support something like this if it was not such a usability nightmare — but that's in the future. Right now, with GeoHack as it is, the consensus is "no".

Given the consensus in this discussion against your proposal, I will revert as vandalism any attempt to enforce the removal of all map links except the GeoHack system. --BrownHairedGirl (talk) • (contribs) 21:45, 30 January 2008 (UTC)

I haven't noticed any proposal of a dual system yet, that allows two clickable links on each article: (1) a link that goes to a standard map, possibly one hosted by a commercial provider, like Google Maps, (2) a link that goes to GeoHack. This might be tried out on a small number of articles. It would certainly give free choice to our readers. I agree with BHG that there is no consensus for a global switch to GeoHack as the only permitted option. EdJohnston (talk) 21:57, 30 January 2008 (UTC)

That's mentioned in the above summary of points related to Wikipedia policies, mostly related to neutrality. Looking at how many popular services are on the map sources list, and how many different users have inserted them, the variety of preferred map services is too much for them to be linked in articles directly. It was tried for some time with the geolinks templates, but dismissed as too partial and impossible to maintain. In proportion, such a list would take too much screen space from articles, and would make Wikipedia more dependant on the editor chosen services. We just can't make that choice for the reader. --Para (talk) 22:16, 30 January 2008 (UTC)

It's a common problem in the wiki discussion format that people are reluctant to participate when the activity seems high and participation would require reading the entire discussion. Anyone who bothers to do that and wants to act as part of the community will naturally agree that external map links are against Wikipedia policies. You can find more views from the discussions on WP:GEO from both sides, though the people opposing the change were in a minority, and had no basis other than being used to the "useful" status quo. All reports of usability have been listened to and for the most part acted upon.

The Wikipedia policies I mentioned above came from Wikipedia:Five pillars. If you revert removal of external map links, you will be reverted in violation of fundamental Wikipedia policies. --Para (talk) 21:59, 30 January 2008 (UTC)

The objection to specific map services sounds ideological to me. I'd prefer to consider a small experiment, where we would actually collect data on how satisfied our readers were with the provided options. For this to work, somebody would have to propose a small set of articles where we would try out some alternatives. They could even be featured articles, such as New York City. EdJohnston (talk) 22:22, 30 January 2008 (UTC)

Of course it's ideological! Wikipedia itself is a very ideological project, and we have to hold on to those ideas and possibilities to keep the project on track. A user test would unfortunately be impossible to arrange on our own without severely biased results, because we can't require everyone to answer. The geolinks experiment ran for quite a long time, so many Wikipedia regulars got used to seeing those links at the bottom of articles, without considering how they fit to Wikipedia as a whole. Since people used to editing Wikipedia would then be more likely to answer any questions, we would mostly get answers from those people. The activity on this page is just resistance to change from bad practices, and it'll pass. --Para (talk) 22:35, 30 January 2008 (UTC)

I sense this is covering the same ground as the ISBN debate, from more than a year ago, where we heard about the wickedness of linking to any site that used advertising, such as Amazon. A particularly useful ISBN-validity-checking service known as http://www.isbn-check.com even got put on the blacklist, due to to the evil fact that one of its screens included links to Amazon. (The site was actually used by a Wiki project that was trying to fix invalid ISBNs). Finally that decision was reversed. A well-functioning discussion thread should elicit better and better ideas from both sides, but I'm not sensing any improvement from your side of the discussion. Whether you think an experiment is a good idea, of course, is up to you. Maybe you might consider a different experiment if you don't like what I proposed. EdJohnston (talk) 22:50, 30 January 2008 (UTC)

That sounds about right. When there's a general identifier such as an ISBN or geographical coordinates that can be used in a number of similar services, and the number is significant enough for a list to be separated from external links sections, then all such links should be removed from articles to eliminate redundancy and leave the choice for the users, especially when the links are commercial in nature. It's no doubt slightly easier for a user to click on a single direct link than click twice, but when only one identifiable link is offered, most will probably click on that and then Wikipedia is seen as supporting a single external advertising supported commercial service over all the others. I'm not sure what happened with that particular ISBN checking service, but it seems to have been abolished from Wikipedia entirely, since there are no usable links to it, not even on the book sources page. Anyhow, during the course of this and other discussions about the use of map links, the map sources page has seen numerous improvements, and its discussion page is where any ideas and reports should be written. An experiment would be great, but unfortunately impossible with the financial resources we have to test Wikipedia's usability. --Para (talk) 23:42, 30 January 2008 (UTC)

Para, what's all this about only offering one link on an article? The only person proposing that is you, with your determination to remove all other map links. Nobody here is objecting to having a geohack link on an article — the objection is solely to your obsession with removing any alternatives.

As to redundancy, you nearly made me die of laughter. I have seen few things with more redundancy than your beloved GeoHack page of 13 screenfuls of links, most of which have absolutely no relevance to the place being mapped.

I don't consider the list to the map sources page a map link itself. Therefore if a page has coordinates inserted using one of the community approved templates, and in addition someone goes to add a direct external map link following their personal preferences, that direct link is then redundant and must be removed. Good for you to keep repeating the ridiculous 13 page argument, it reduces the weight of anything else you have to say. --Para (talk) 01:44, 31 January 2008 (UTC)

Also, you continue to repeat your lie about "two clicks". (And I call it a lie, because while I tried to assume good faith, your continued refusal to listen to other experiences makes that assumption impossible to sustain). Yes, it is possible to use GeoHack to access some map services with only two clicks, but only if they happen to be the ones listed in the first screenfull and the user knows that they should ignore the useless links to "Europe A-M" etc. Unless the user has somehow managed to figure the completely opaque intentions of the page's designers, then the experience is more likely to be twenty clicks before giving up in frustration. --BrownHairedGirl (talk) • (contribs) 00:13, 31 January 2008 (UTC)

Despite your unwillingness to cooperate, you should note that all your comments get due attention, and as such changes are possible at any time. Please check the page before commenting. These "useless links" are no longer as visible as they were when you first mentioned them, so people's attention will be drawn to what the heading on the page says, the content of the page, and not on navigating the page further. But this is all irrelevant and off topic here, as the map sources page is not only as usable as the book sources page, which alone is enough to continue with the change, but it is actually better, since most of the popular and appropriate links fit on a single screen. Please place any further comments about usability on Template talk:GeoTemplate. --Para (talk) 01:51, 31 January 2008 (UTC)

Arbitrary section break

Maybe everyone should take a little chill. This issue isn't new and it's not getting resolved soon, so no emergency worth getting personal over. I don't think it's a good idea to link directly to map services, and I would probably remove such a link if it appeared for no good reason in an article I was working on. At the same time we have 300,000 and it's premature to delete them en masse from articles without having a really good system in place. The best, I think, is for anybody who feels strongly about this and/or wants to do a lot of good to develop our technology a little further, and maybe the usability of the way geocodes are displayed in articles. Once it's so good that everybody loves it, we can set about converting external map links to geocodes using bots. We could even give people some notice and warning time so they can put a "do not convert" flag somewhere for the bots, if they strongly feel that a particular external link to a mapping service should remain. Of course they could always just revert the bot too...Wikidemo (talk) 01:45, 31 January 2008 (UTC)

The 300,000 mentioned in the beginning of this topic was actually the number of articles with recognisable geographical coordinates, not the number of external map service links. At the time there were 18,000 external links to the services listed on the map sources page, but it has since come down. Unfortunately I can't give current numbers as the toolserver database for the English Wikipedia is down this week. Wikipedia has had rampant external map links in articles for such a long time that that's what many people willing to comment are used to, and seeing the disregard of Wikipedia policy on this page already, I doubt waiting for the love is the way to go. What we need is mass review and possible conversion and removal of the remaining external links, or tagging as unique to be kept and guidelines for which types deserve that treatment. It shouldn't be done with a bot to make sure that the automatically generated links give the same location and similar results as the editor inserted external link, and that it's indeed the location of the article's topic. The majority will not be reverted, and the ones that are will just end up for review again for someone else. Geographical coordinates and map service links are such a minor detail in articles that most people just aren't going to pay any attention if they are notified of an upcoming change, as was seen in the geolinks change on WP:GEO. The best solution is to just go for it and then see about the comments. Meanwhile, to keep the convenience seeking editors happy, people can work on Javascript solutions to have direct links to services. I wrote commons:User:Para/Google Maps Love.js once for this very purpose, and it works with coordinates here on the English Wikipedia too. --Para (talk) 02:35, 31 January 2008 (UTC)

As above, Para: there is a clear consensus against proceeding with your proposal at this stage.

To implement it against consensus would amount to tendentious and disruptive editing at best, and could be construed as vandalism, and should earn you an instant block. --BrownHairedGirl (talk) • (contribs) 03:06, 31 January 2008 (UTC)

I have been converting and removing external map links for months now and the only one to revert me has been Sarah77, who then directed BrownHairedGirl here. This 3-3 discussion where the opposing viewpoint completely disregards policy won't stop anyone from improving Wikipedia and removing the editor chosen external map links from articles. Any attempt to interfere with it by empty threats, abusing admin tools, tendentious editing ("editing which is partisan, biased, skewed—in other words, it does not conform to the neutral point of view", such as direct external map links) or disruptive editing ("people who seek to exploit the site as a mouthpiece for viewpoints that constitute original research", such as enforcing personal map service preferences on others) is a violation of all the mentioned Wikipedia policies. There is absolutely nothing to back you up. --Para (talk) 03:40, 31 January 2008 (UTC)

Para, there is no consensus for your changes, and the relevant guideline (note guideline, not policy) does not prohibit direct links to map services. You are clearly in live with the GeoHack system, but that does notmean that it is wikipedia policy to prohibit appropriate direct links to map services, and you have not offered any clear evidence of a policy to support mass removals. --BrownHairedGirl (talk) • (contribs) 03:59, 31 January 2008 (UTC)

Does the 3-3 include ScottHatton, JRG, myself and the others who have commented the same way on WT:GEO, or only the ones who have contributed here? Orderinchaos 11:17, 2 February 2008 (UTC)

No, BrownHairedGirl and yourself have repeatedly belittled the role of WP:GEO as a place to discuss this topic, along with the consensus achieved there, so the count does not include people such as Dschwen, Docu, Geonick and others who have supported removal but have not joined the discussion here. --Para (talk) 13:30, 2 February 2008 (UTC)

Wow. Massive threadage. I'm just here saying that I _like_ clicking co-ordinates and being given a selection of mapping services to chose from. It's obviously contentious, so I'm not going to start editing them in. I guess existing links to mapping services shouldn't be removed yet, but that leads to the weird situation of links to maps and links to "map disambigs". About the usability / interface design: what would help? Different wording, bigger fontage for the first few links, horizontal rules? Dan Beale-Cocks 20:14, 31 January 2008 (UTC)

I strongly support adding the GeoHack links to articles, because there clearly are users such as Dan who have figured them out and like using them ... and don't see that adding those GeoHack links is contentious. The contentious point is the removal of direct links to maps, and I think it is important to separate those two issues.

I have several idea about how GeoHack could be improved, but I'm not going to waste time discussing them with Para, who has persistently dismissed all concerns as "nonsense". If there is anyone else involved with GeoHack who is prepared to actually discuss usability problems and to seek solutions (rather than Para's zealot-like approach of telling users that the syystem is fine, it's the users who are wrong), then I'd be happy to help. --BrownHairedGirl (talk) • (contribs) 21:42, 31 January 2008 (UTC)

Wikipedia needs to have a single consistent standard for linking to external map services, and the coordinate links are the only existing neutral way to do that. If direct links to Google Maps for example were added in addition to the coordinate links, then those are what people would most likely click on because the brand name is more recognisable than anything Wikipedia can offer. Wikipedia should not promote any advertising-supported commercial service over another. Anyone who has concerns on usability should report them on Template talk:GeoTemplate and/or Wikipedia talk:WikiProject Geographical coordinates. --Para (talk) 03:22, 1 February 2008 (UTC)

I am in much the same situation as BrownHairedGirl. I've been having this discussion for over a month at WT:GEO but to no avail - it seems that those who wish to make the changes will do so whether we like it or not, and have no respect for consensus as a model for getting changes through. I personally think ultimately this is going to end up at dispute resolution, which is from a scale-of-problem point of view so unnecessary given we're only talking about the behaviour of two users who could act in a more appropriate manner and put this matter to rest by simply talking to and dealing with their opponents in good faith instead of antagonising them unnecessarily. Orderinchaos 11:12, 2 February 2008 (UTC)

(drama break) BrownHairedGirl (talk · contribs) has started reverting removal of map service links. See Wikipedia:Administrators' noticeboard/Incidents#BrownHairedGirl involved in tendentious and disruptive editing. --Para (talk) 03:22, 1 February 2008 (UTC)

Perhaps the people who are committed to promoting Geohacks as the single one-solution-for-all way to display geographic links can provide a tutorial for those of us who are unable to figure out how to make it generate maps that meet the needs of particular articles. Here's an example of an article where I find this tool frustrating: Clinch River Breeder Reactor Project, which has a land area of 1,364-acre (6 km²), making it comparable to an airport. On Google Maps, the ideal map scale for this entity would be the one generated by z=12. I cannot get the Geohacks tool to generate this map scale. When the 1:10000 scale is specified in the coord template, the resulting Google Map is the one generated by z=15 (much too large a scale). However, when the "airport" scale (1:30000) is specified, the resulting map is the one generated by z=10 (too small of a scale to illustrate what I want the article to illustrate). Perhaps I have not yet spent enough hours reading documentation of Geohacks, but I have pretty well concluded that the tool is not capable of providing the map I want, so it's not worth my time to fool with it.

At least the scale of those Google Maps is more useful than some of the links that Geohacks serves up. When the largest (1:10000) scale is specified, the Shaded Relief map that is generated is about 200 km across; the NASA/MSFC GOES satellite image that is generated is one that covers a region about 1000 km across; and Geohacks serves up the Degree Confluence Project page for a location that is not even the nearest degree confluence point. (The actual nearest degree confluence is this one -- but how many people who follow the map link in an article are actually interesting in photos at a degree confluence?)

--Orlady (talk) 04:30, 1 February 2008 (UTC)

This would have been best asked on WP:GEO, but I'll answer here just to demonstrate how versatile and reusable coordinates and the related information about them is. Reading the documentation of coordinates, the type parameter categorises the object into a class that can later be used for visualisation of the information as icons for example. It additionally sets a default scale for the object, and most objects of that type will be visible with the default. If however they are not, the scale parameter can be used to set the correct scale. Assigning the coordinates of a proposed power plant an airport type would be wrong; it's a landmark at best. It's hard to say which scale value would be suitable, and the easiest way to find one is trial and error, by changing the scale parameter in the GeoHack url or in the coordinate template until you get a map at a scale that contains the entire object. type:landmark_scale:50000 in 35.897°N 84.378°W / 35.897; -84.378 (Clinch River Breeder Reactor) might be appropriate. You should then be able to change between map services and get approximately the same view with all of them. Shaded Relief and NASA/MSFC GOES don't seem to have imagery at the requested scale, so they give the best they have. You might want to try ACME Mapper, which has similar data at a larger scale. --Para (talk) 05:27, 1 February 2008 (UTC)

Issues with inclusion or exclusion of map service links

Reading the above discussion is not necessary for participation in this discussion: this section is an attempt at a summary.^{{sofixit}}

The issues related to the inclusion or removal of external map service links in articles directly needs to be discussed, without letting personal preferences affect what will eventually be a guideline on what all Wikipedia editors should be doing, be it to keep on adding them or remove them. The proposed alternative to direct external links is a link to a map service list page, linked in a way similar to what is now used in the Alexandria article. Below I have presented relevant issues based on Wikipedia:Five pillars. They were already mentioned above, but have not been discussed, so that's what we should do to be able to close this. Please add any other issues as new sections, add counterarguments between the existing ones, and try to keep them short enough for most people to read. Note that this is not a vote, but a discussion with the objective of getting concise arguments that will help with later decision making. --Para (talk) 17:02, 2 February 2008 (UTC)

Wikipedia is an encyclopedia for the dissemination of information

• An external map service link is only information that a single external resource may at some point have had information about the topic, whereas coordinates are information by themselves, usable in any online or offline service. --Para (talk) 17:02, 2 February 2008 (UTC)

• Most location related articles do not require a direct link to a specific map service, as similar maps and other data is available from a number of services accessible with the general coordinate information about the location. --Para (talk) 17:02, 2 February 2008 (UTC)

• External map links are a result of lazy editing: Instead of doing what an editor should do and processing the link into encyclopedic information, it's just inserted to the article as such without bothering to generalise. --Para (talk) 17:02, 2 February 2008 (UTC)

• The phrase "a picture tells a thousand words" comes to mind. If the above contentions were indeed true, GeoHack should cease to exist. Orderinchaos 18:54, 2 February 2008 (UTC)
  • Maps are indeed information, but the links to them are not, and as is mentioned at the first point in this section, they serve no purpose other than to tell that some have existed of the location. Imagine printing such a link. A statement saying that maps exist of the location does the same, without linking to any specific map, and that's what GeoHack does. Since the data is dynamic especially with satellite imagery, there is no guarantee that a link to a specific service will always give the same results. See Wikipedia:Avoid statements that will date quickly. --Para (talk) 20:17, 2 February 2008 (UTC)

This is a situation with copyright over the maps (very, very few maps are available under GFDL/CC). Were we able to display maps from such providers, we probably would. However, the next best thing is providing the links. Orderinchaos 22:06, 2 February 2008 (UTC)

OpenStreetMap is one such resource. For example, the map of Perth, Western Australia looks quite nice to me, and could very well use a few more Wikipedia links. Comte0 (talk) 10:23, 3 February 2008 (UTC)

Wikipedia has a neutral point of view and is free content

• The selection of map services is a reader choice and editors cannot objectively make that choice for all readers from a neutral point of view, because the needs and preferences of most readers are not known and an editor choice reflects the editor's personal preferences. --Para (talk) 17:02, 2 February 2008 (UTC)

• Direct links in articles supports the preferences of very few people only. --Para (talk) 17:02, 2 February 2008 (UTC)

• Wikipedia does not sponsor certain advertisement supported commercial map services over all the others. --Para (talk) 17:02, 2 February 2008 (UTC)

• Wikipedia does not depend on external services, which consistent external map links in articles would cause. --Para (talk) 17:02, 2 February 2008 (UTC)

• I don't see how any of these issues relate to the current debate. Streetmap.co.uk and Street-directory.com.au, to cite two examples, give users exactly what they would get if they bought a published street directory off the shelf. Google is pretty uncontroversial, especially since they updated their maps about 2 years ago. A satellite picture is a satellite picture no matter on whose site it appears. This doesn't appear to me to be an NPOV issue. Orderinchaos 18:53, 2 February 2008 (UTC)
  • By linking to those sites only, you take the position that readers only want to see street maps as they are when bought off the shelf, or satellite images as Google provides them. What if they want to see topographic maps or oblique imagery, or other information that is easily available with the coordinates? Most services for geographical information give similar material, but it is not the same. Satellite imagery from Google, Microsoft, Yahoo, Ask.com, TerraServer, MapQuest, MultiMap, etc is mostly from different sources, and they are often from different dates. This can easily be compared with the Flash Earth service for example. If an editor chooses the type and source of data to offer to readers, that is a strong POV. --Para (talk) 19:27, 2 February 2008 (UTC)

I don't think anyone is saying abolish the link to GeoHack - I certainly never have. Some people find a great use for it. I simply hold the view that people shouldn't be *forced* to do so. That in itself is a strong POV (and also one which is quite insulting to our users). Orderinchaos 22:01, 2 February 2008 (UTC)

If articles have prominent direct links to Google Maps for example and none of the other alternatives, then those are what people looking for maps would most likely click on because the brand name is more recognisable than anything else Wikipedia can offer. That's a clear sign of Wikipedia promoting that particular service. This should probably be noted as its own point as it's likely to come up elsewhere as well. Why would it be insulting to make people choose? --Para (talk) 03:25, 3 February 2008 (UTC)

• This is a primary argument against ad-hoc linking, although there's a terminology issue. It's not NPOV as in article neutrality, but neutrality as among web services so as not to be arbitrary. If there are multiple services that all achieve the same thing we shouldn't favor one or the other. And it's not free as in "libre" but free as in noncommercial. We should not be favoring one commercial service in particular, although individual users should be able to choose a service if they wish. Wikidemo (talk) 22:05, 2 February 2008 (UTC)

Think of a service that shows outdated maps without recognising some new country that has existed for years, or a service that is known to provide imagery censored from sensitive locations. If all location related articles link to such a service as an only resource for more geographical information, that's a POV that takes some neutrality away from articles. --Para (talk) 13:01, 3 February 2008 (UTC)

Wikipedia should be accessible to the widest possible audience

• Links chosen by single editors reflect their personal preferences, and they may not include accessible external services. --Para (talk) 17:02, 2 February 2008 (UTC)

• Direct links break geographical information into unusable snippets that can only be used online with a modern web browser, unlike coordinates and a full list of services. --Para (talk) 17:02, 2 February 2008 (UTC)

• A single or a couple of external map service links ignore the users' needs, maybe without even the user knowing that other services with data for the location exist. We can't know why people go to geographical information services and which ones they use. --Para (talk) 17:02, 2 February 2008 (UTC)

• The current arrangement is unusable for most editors with a modern web browser, because it makes assumptions about their understanding of the technology which are in many cases false. Users presented with a sea of links are simply going to abandon ship, rather than get the information they want. Look at this site for a counterexample of how another big organisation does it - it lists a few key stories and then has "Click here for more" - this site does exactly the same, as does this one. These approaches, which is basically what we were doing until late last year, would address the entire third concern. This is in my view why editors should develop consensus as to which sites are appropriate for a region, and then employ across the affected range of articles. Sites which would fail EL more generally can be excluded, as we've done in the past. Orderinchaos 18:49, 2 February 2008 (UTC)
  • What assumptions does this arrangement make? Why wouldn't users click on the first or any of the links that say "Map"? The difference between Wikipedia and several news organisations is that Wikipedia attempts to be "the sum of all human knowledge", with information and links to all relevant reliable resources. Most news organisations however only link to other content in their own organisation. We have no criteria on "best" services, see #Wikipedia articles can link to relevant resources. --Para (talk) 20:17, 2 February 2008 (UTC)

We're not saying any service is better or worse simply by providing a collection of standard links. Doesn't GeoHack exhibit the same problems with ordering? For example, if something is near the top of the list, are we saying it is better than something near the bottom? In history articles we link to or cite particular books or sources. Are we saying that we are favouring or promoting those sources over alternative sources which may exist and someone somewhere may prefer? This could go on all year. Orderinchaos 22:03, 2 February 2008 (UTC)

Citation sourcing isn't analogous to map linking. A citation establishes verifiability, so as long as it is good enough it does not matter which one we choose. There have been some longstanding attempts to specify which citation to use (or when multiple citations are appropriate) when they all do the same thing: the one that's most reliable, most accessible, most authoritative, most informative, and so on. But those efforts all stalled out. With source citations, each link is different and depends on context. With maps, we're in a situation where nearly every single external link uses all the same mapping services in the same way. Hence the choice among them is pretty much arbitrary, and the practice is much easier to make uniform. Perhaps a better analogy would be, say, the way we link to logos, stock price quotations, financial reports, and so on, for companies. That one is only partly uniform, possibly because nobody has made the effort yet. Another analogy would be case citations for lawsuits. If we had a legal research tool as good as geohack I think it would be the same issue as here that we would like to that rather than random pages where a judicial opinion is reprinted. Wikidemo (talk) 22:18, 2 February 2008 (UTC)

What would the criteria for providing in-article links be then, if it's not the editors' subjective idea of "best" links? Most map services are dynamic and have a user interface you can do various things with and not just have a static image (or text) file dumped on the screen like with citations usually. I think you can draw some conclusions from the order of the listed services both in geolinks and geotemplate, though I've never noticed people zealously moving them around. I tried looking once at Google Pageranks and Alexa data for a solid source for ordering them, but couldn't get anything conclusive for the map pages only. Being restricted to a two dimensional interface and a culture where information is mostly handled left to right and top to bottom, there isn't much we can do to avoid the ordering issue. Commons Picture of the Year competition for example had Javascript shuffling the candidates to a random order every time the page was loaded, but that would be silly for an interface that people may get used to and things are expected to stay static. I proposed the idea once[1] that the service list should allow each user to move the services to their preferred order with Javascript, and keep the changes for later. That shouldn't be too hard to implement. --Para (talk) 03:38, 3 February 2008 (UTC)

Wikipedia articles can link to relevant resources

• All map services are relevant, useful, helpful, informative and factual. All of them placed in articles would take too much space. There are no "most relevant" map services, because of readers' varying needs and preferences. --Para (talk) 17:02, 2 February 2008 (UTC)

Not true. Elsewhere I have used the example of Clinch River Breeder Reactor Project. I contend that, for readers of that article, several of the map service links found on the first screen of the Geohacks page, such as the US Census map and regional weather satellite images, and the link to the degree confluence project are totally irrelevant (not to mention the fact that the Degree Confluence Project returns a link to the wrong degree confluence because it rounds fractional coordinates down to the nearest whole degree). Furthermore, the small-scale maps served up on that first page, such as the one from the Shaded Relief Map of the World are far less likely to be helpful to readers than maps that can be used to show closer-in details. After the first screen, almost all of the Geohacks information is totally irrelevant, since a person who wants to see a map of this particular site is not looking for general information on online maps or a tutorial on how to download Google Earth, and definitely does not want to look at maps of Austria, Bulgaria, Australia, China, or the other world locations found farther down the page. On that page and many others, one or two very specific map references (for that article, I would probably choose Google Maps and Topozone) would be far more relevant than the smorgasbord of geographic information supplied by the Geohacks page. --Orlady (talk) 21:15, 2 February 2008 (UTC)

Ok, so some services are less useful than others. That should've then been "Most map services are relevant, ...". The services for other regions may be helpful near region borders, but I'm sure they can be hidden or even removed from the default view if enough people feel that they're mostly useless. Otherwise the usefulness of services that really are for the region is not really for any of us to say, you don't know why someone is interested of the reactor project article for example, and it seems some people do like the census maps enough to insert them in articles and link to them. Everyone isn't looking for close-up detail. --Para (talk) 03:44, 3 February 2008 (UTC)

The problem is that the vast majority of the links and other information provided on Geohacks are irrelevant to any given article. Which small fraction is relevant depends on the article. Census maps can be very useful, but they are not relevant for the particular example I provided because the site of interest happens to be 1,364-acre (6 km²) of US-government-owned land where there are no buildings and no people. If the Geohacks content could truly be tailored so it would provide only what's relevant to a particular article, I'd be happier about pointing to it. As it is, the reader of that particular article is served a 94-KB menu of mostly irrelevant material, without guidance on how to select a map link to click on. In effect, you are handing readers a giant haystack and telling them to find a needle in it, and you are insisting that every reader must be given that same haystack because you can't assume that they actually want a needle. --Orlady (talk) 04:54, 3 February 2008 (UTC)

Census maps still show landforms, borders and roads, and they're public domain so they're quite relevant for Wikipedia goals, and yes even for this article of yours. What kind of guidance do you think needs to be given for the services? It needs to be something that can be said for all the services for all the regions, like the service type is now, and it needs to be manageable information that won't be out of date the next day. I have removed dozens of map links that only give a gray screen; they may have worked with the requested scale at some point, but nobody cares to keep them working because it's such an immense amount of work to constantly keep checking all available services without any possibility for automatic verification. More organised efforts for large regions have failed as well, see the inconsistency in the coverage of geolinks templates (open the pulldown table), which tried to offer services relevant for the article editors, but couldn't even among themselves agree on what would be relevant to the region. --Para (talk) 13:01, 3 February 2008 (UTC)

Wikipedia articles containing geographical information should be referenced

• There was a discussion at Wikipedia talk:Cite sources/archive20#References for geographical features, where the majority agreed that Wikipedia articles can cite specific dynamic map services as references. The discussion here concerns external links other than references, but the other discussion ought to be mentioned for completeness sake. --Para (talk) 17:02, 2 February 2008 (UTC)

Wikipedia should have related information available as easily as possible

• Ease of use is important, but it is not Wikipedia's ultimate goal at expense of the above points. How has usability of different features in Wikipedia been evaluated in the past, and how has it affected their introduction? --Para (talk) 17:02, 2 February 2008 (UTC)
  • Wikipedia:ISBN and Wikipedia:Book sources was taken in use as an acceptable replacement for direct informative links for books. Template:GeoTemplate is similar with map services, though more usable. --Para (talk) 17:02, 2 February 2008 (UTC)
• This is in fact my key ground of objection to the changes being rammed through - First point, information about a topic should be on that topic's page. This is a principle held to in article text, I don't see why it should be any different for templates. On the second point, we have people who do not speak English, people who are disabled or otherwise impaired, people with learning difficulties etc using our site and I have no wish to exclude any of them from obtaining the full value possible from Wikipedia. The changes to GeoHack are an improvement, but the ugly-looking one line template that Template:Mapit-AUS-suburbscale and others like it have been reduced to gives no indication what to expect when one clicks on the link, and I still believe that there should be some kind of links on the page. Most people who have contacted me complaining believe the links have been entirely removed. Until January of this year, GeoHack was completely unusable, and while significantly improved now (I can actually use it now), it is still a bewildering array of links and quite confusing to the novice user. Orderinchaos 18:43, 2 February 2008 (UTC)
  • 1) Map links are not information, see #Wikipedia is an encyclopedia for the dissemination of information. Wikipedia articles on most topics do not contain all the information mentioned in it, but link to other articles or resources. See Wikipedia:Summary style for more information.
    2) The English Wikipedia is not for people who do not understand English.
    3) Totally agreed that we shouldn't exclude anyone from the full value possible from Wikipedia, which is why all the services need to be linked to let everyone choose for themselves. This is covered in #Wikipedia should be accessible to the widest possible audience. --Para (talk) 19:04, 2 February 2008 (UTC)

Try asking someone who is vision impaired what they think about having to wade through software-formatted tables to find a simple link. The more barriers we put up, the further we get away from our goal. I think the saying is "cutting off the hand to spite the face". Orderinchaos 22:04, 2 February 2008 (UTC)

From my reading of the HTML standard the new geohack page is far more accessible then what we had. It should read correctly as far as I know, if you have evidence to the contrary please tell. — Dispenser 02:53, 3 February 2008 (UTC)

What's the difference between software formatted and human formatted, when both are designed and controlled by people? I've understood that many people with visual impairments can use browser features to help make pages easier to read, such as font size and colour changes, alternating row colours, more spacing between letters and rows, magnifiers, and for really difficult cases choosing the sequential order of aural or braille reading between the visual screen order and the order within each element. The tables used in GeoTemplate are only for layout purposes without any semantic ordering, and all column names are repeated throughout the cells, so these interfaces should handle the page well. Is there a checklist somewhere that we could use to test with? --Para (talk) 03:21, 3 February 2008 (UTC)

• Now this might be crystal, but using the {{coord}} templates instead of external links might actually help vision impaired people as more browsers become able to independently process the informations contained in microformats. They are also impervious to tool server down time. And the plugins can be user-configured to list the map services of their liking. --Qyd (talk) 23:39, 8 February 2008 (UTC)

Wikipedia content should be consistent, not arbitrary

• When the same thing is done again and again we should choose a single method of dealing with it so as to be consistent and predictable. This is why we have style guidelines. It makes our content more encyclopedic, and in the case of external links, prevents accumulation of unnecessary excess links. There is no reason to have multiple mapping links so we should settle on a single method for all articles (with exceptions as needed of course). Geohack is the best method available so we should choose that one.Wikidemo (talk) 22:09, 2 February 2008 (UTC)

This is one of the main goals of WP:GEO as well, and the same principle goes for any Wikipedia content. --Para (talk) 03:27, 3 February 2008 (UTC)

• Consistent linking to any single external service only is a clear sign of Wikipedia promoting that particular service. --Para (talk) 03:27, 3 February 2008 (UTC)

Responses

Pardon for seeming somewhat confused, but reading through the discussion didn't help me to understand too much except that there are heated feelings flying here. My opinion is that the use of the GeoHack page is not a violation of Wikipedia standards, and the listing of multiple external mapping links is not a violation of Wikipedia standards — I find it much more of an issue of convenience. My primary focus in Wikipedia is US geography, so I'm always encountering these links. I always replace external mapping links with {{Geolinks-US-cityscale}} when I find them: not because I find it more acceptable under official policy, but because the use of this template, as formatted by the WikiProject Geographical coordinates, is required by the guidelines for US places. You can see that a previous version of the Geolinks, instead of using Geohack, contained multiple external mapping links; it has been changed by conensus of the geographical coordinates wikiproject. As I said, I really don't care strongly; it's quicker to find maps with the multiple-links, but the geohack is much more comprehensive, and thus I'm somewhat more pleased with it. The big issue for me is that things should be standardised: since the geolinks template is so useful (regardless of its format), I believe that it should be used in place of specific external links, and the question of geohack vs. various ELs is properly applied to the question of the content of the geolinks template. Nyttend (talk) 17:48, 2 February 2008 (UTC)

The discussion all the way from #Links to map services did get a bit heated, so I thought listing the issues in a neutral manner would resolve the situation, and people could start reading from #Issues with inclusion or exclusion of map service links. I'd like to see that entire section edited and counterarguments added. For example, I think #Wikipedia has a neutral point of view and is free content is fundamental Wikipedia policy (WP:NPOV) and is strongly related to Wikipedia promoting certain external advertisement supported commercial services, but everyone may not agree on the following arguments, and I would like to establish a consensus on which way people are leaning. A simple threaded discussion did not seem to be the right way to do it, so please let's try not to continue it here in Responses. --Para (talk) 17:58, 2 February 2008 (UTC)

I had initially raised objections similar to some of the objections being raised now on a different talk page. At the time, I opposed removing the more specific geolinks because the usability of the tools page was a mess. However, I have changed my opinion completely. The tools page is much better organized now and makes it quite easy to find appropriate maps. I think Para and SEWilco should be commended for persistence in pushing these improvements through. older ≠ wiser 18:25, 2 February 2008 (UTC)

Okay, this looks like material for the #Wikipedia should have related information available as easily as possible section on the current usability of the proposed link page. Can you add it there, or another new section, so that it doesn't look like my monologue? --Para (talk) 18:34, 2 February 2008 (UTC)

Conflict of interest

Map links seem like a good idea, but they pose a huge potential COI: Which one do we go with? Google Maps, Mapquest, or maps.live.com? Linking to all three would be redundant and ugly. Linking to just one would be contentious and a COI, since it's not really fair for us to give Google ad revenue over Microsoft or Mapquest. This doesn't really seem critical. It seems rather silly to push so passionately for it, as if it were some kind of critical feature desperately needed for encyclopedic article content. Since it is not, the argument ultimate relies on the arguments to avoid "its useful" or "its interesting." ☯  Zenwhat (talk) 23:06, 2 February 2008 (UTC)

Exceptions

Moving on; there doesn't seem to be any solid argument for keeping external map service links generally, but what about exceptional cases?

• Reference use, as mentioned in #Wikipedia articles containing geographical information should be referenced, and recognisable without manual review using the ref tags. Can be kept.
• Links that provide more information of a feature discussed in an article, the data is known to remain static, and it's not available in other services, ie. they're links to data from a specific source and date. Can be kept.
• Links that use a feature other than just pinpointing a location, such as plotting a route or a border through many given points. Those can be kept, and need to be tagged somehow to help manage all external map links. Ideas? The tagging information could be on the toolserver or somewhere similar external location, but for the benefit of all reuse it'd be best on wiki.
• What about links to services with dynamic data, for example a satellite image of an object temporarily at the location? Should a direct link be provided, or should it be edited like "as of date, service shows object at coordinates" and users would then have to click on the coordinates and go to the mentioned service, and possibly look on others if the editor hasn't?
• External map links in article bodies for addresses and other minor locations without a Wikipedia article. Such links should either be converted to coordinates or removed. Generally, if coordinates are irrelevant information to put in an article inline or hidden somewhere in the end, then so are map links.

Any others? --Para (talk) 22:04, 8 February 2008 (UTC)

Some of these may have been mentioned in other parts of this Talk page, but here are other External Map Link examples to consider for inclusion in the Exception group:

• Older map with different names or boundaries.
• Map depicting movement (i.e. animal migration, battles)
• Political map (i.e. red states, blue states)
• Public place map (i.e. university, zoo, state park)
• Weather map
• Map related to a news story (i.e. areas burned in a fire)
• Map depicting geological features (i.e. Churchill craton)
• Special region map (i.e. wine regions of California, amusement parks in Florida, school district boundaries)
• Nautical and aeronautical charts Rosiestephenson (talk) 08:48, 10 February 2008 (UTC)

External map links and external map service links are separate things, and this section has until now only covered the services. There are indeed many articles with links to static maps that provide more information than the general services do, and with popular articles such links have been replaced with free images created by Wikipedians based on those maps and other sources. This process will however take a considerable amount of time, and I doubt any guidelines can be formed there.
One of the issues with both static maps and dynamic map services is that so much data on so many topics is related to locations, that the amount of possible visualisations and services is endless. Inclusion of links to specialised services may need to be looked at on their relevance to the article, but is it possible to draw a line somewhere, or can we rely on common sense? For example, a link to a weather service would not be relevant in an article about a location, but in an article about the weather in the location it might, and perhaps also in an article related to both the location and weather, like with a natural disaster in the location. But there too we end up with the same problem as with general map services: there are so many good and relevant weather services. How many links is too many? --Para (talk) 09:21, 11 February 2008 (UTC)

Haven't really read all of what was being discussed, but isn't there a cord template made for this reason? You use the cord template and it generates links from the software for map services. There really isn't a need to manually point a map link on an article for a specific service. -- Ned Scott 05:36, 10 February 2008 (UTC)

Point. I wikified coordinates. --Para (talk) 09:21, 11 February 2008 (UTC)

Our site has a new domain, hundreds (thousands?) of links to it in Wikipedia need to be updated

Is there a way to globally update links that contain our old domain to the new one? Opening up each article individually isn't practical (see linksearch results below). Our old site was located here: www.tsha.utexas.edu and the new one resides here: www.tshaonline.org . Everything is still in the same place, just a different domain ( www.tsha.utexas.edu/education is now www.tshaonline.org/education for instance)

I put the old URL in the Wiki linksearch and got a rather large list: http://en.wikipedia.org/w/index.php?title=Special:Linksearch&target=http%3A%2F%2Fwww.tsha.utexas.edu%2F&namespace=&limit=500&offset=0 Any help would be appreciated! Txnomad (talk) 17:22, 4 January 2008 (UTC)

Sorry, but there is no simple way unless it is put in a template (and putting external links in templates would make it an easy target for spammers). The best way is to use AWB or maybe Twinkle for that. Maybe you can request a bot for the changes, though. Or see for someone with editcountitis to help. -- ReyBrujo (talk) 17:49, 4 January 2008 (UTC)

From what I see there are 3035 links. Some (the www.tsha.utexas.edu/handbook/online/ ones without determined page) can be removed since they aren't giving information to the articles. The 3000 others are the ones to change. -- ReyBrujo (talk) 17:51, 4 January 2008 (UTC)

See Wikipedia:Bot requests. -- SEWilco (talk) 17:55, 4 January 2008 (UTC)

I notice that http://tshaonline.com identifies itself as run by the Texas State Historical Association. Is there some reason why the existing page at http://www.tsha.utexas.edu couldn't redirect all visitors over to tshaonline.com? Is there some legal reason why they can't anymore be included at utexas.edu? Remember that this change will not only break all the links at WP, which are fixable, but all links elsewhere on the web, which will reduce the usefulness of the information in this historical collection. State history is surely not so controversial that U of Texas can't link to it. Perhaps Txnomad could have a conversation with the webmaster of utexas.edu. EdJohnston (talk) 18:40, 4 January 2008 (UTC)

The site is at tshaonline.org, not .com. (Strangely, the .com site seems to be a different but related site that links to stuff in the utexas.edu site.) *Dan T.* (talk) 05:11, 1 February 2008 (UTC)

There is a template at {{Handbook of Texas}}. I updated the URL in that template. There are still some 2000 links to convert to the template-- User:Docu

Everything is still in the same place, just a different domain ( www.tsha.utexas.edu/education is now www.tshaonline.org/education for instance) You can do this in your server's config... by issuing a Permanent redirect from the old domain name to the new domain name. Contact your hostmaster or webmaster to do this. ≈ jossi ≈ (talk) 02:04, 17 January 2008 (UTC)

I don't know if this would apply to this situation, but situations like this can often benefit from being added to Meta:Interwiki map, if said site passes the criteria there. This would allow for such an update with minimal work. Alternatively, you can also make a template for the link. Granted this doesn't really help anyone out now, nor can we always predict when someone will change domain names, but I thought I'd throw it out there. -- Ned Scott 06:26, 17 January 2008 (UTC)

Recently Hu12 raised a spam concern about some of these links. See my reply to him here. I left a message on Txnomad's talk asking him to return to this discussion, since he's been active today. So we have two questions: (a) is mass conversion appropriate? (b) should we be keeping the generic links to the TSHA site at all, or the generic links to their handbook? The links to specific historical documents, from relevant articles, are likely to be OK in my view, but I don't know if those are the majority. EdJohnston (talk) 18:13, 17 January 2008 (UTC)

Regrettably, we have got no response from Txnomad who originally raised the question. So I have started a very slow-moving update process, where I use a Linksearch for references to the Handbook of Texas and try to replace the links I find with instances of the Handbook template. For example at White Rock Lake I used this format: {{Handbook of Texas|id=WW/row4|name=White Rock Lake}} . EdJohnston (talk) 22:34, 30 January 2008 (UTC)

The original crisis has gone away, since www.tsha.utexas.edu is now redirecting to www.tshaonline.org, which is what some of us hoped would happen. So our articles won't be stuck full of dead links. I still think the new template {{Handbook of Texas}} is superior to the old way of linking, since it doesn't create an unnecessary clickable link to the top level of the Handbook, and it lets our readers go to the Wikipedia article on the Handbook if they wish. However conversion of the 3,000 links now drops so low in priority that it won't happen soon. EdJohnston (talk) 04:38, 1 February 2008 (UTC)

The server isn't configured correctly, it should be sending out a 301 (permanent redirect) code instead of the 302 (temporary redirect). The difference is that in former (in theory anyway) bot would update the links cementing the change. — Dispenser 03:06, 3 February 2008 (UTC)

Blogs of the article subject

Is it really reasonable that a personal blog of the person the article is about be removed? I refer to Aaron Turner, where his own blog has been removed from the external links section. Surely that's a little overly-literal in the interpretation of the letter of the law? Thoughts? If I'm wrong, please direct me to the relevant guideline, so I can gape at it in disbelief. Thanks. Seegoon (talk) 11:48, 6 January 2008 (UTC)

By my reading of these guidelines, a blog authored by the article subject is okay to link to in most circumstances. My feeling is that a blog authored by the article subject falls within the "official page of the article subject" exception; so, provided the blog isn't prohibited by this guideline's Restrictions on linking, I don't see a problem with adding it to the external links section. --Muchness (talk) 12:12, 6 January 2008 (UTC)

I'd agree. As long as the blog is provided solely as an external link and not as a source, and as long as the fact it is indeed their own blog has been authenticated, I think there should be no problem. Orderinchaos 22:12, 2 February 2008 (UTC)

Lists of manufacturers and suppliers

I should like to explicitly exclude lists of external links to manufacturers and suppliers. Not only are these unencyclopaedic, but since they are rarely complete they are potentially commercially damaging to non-included companies. May I have any views, please? TerriersFan (talk) 23:58, 17 January 2008 (UTC)

I am not usually happy with long lists of manufacturers, and even less happy with list of suppliers. I could see an exception for hard-to-find items where normal search methods might not succeed. But it is entirely appropriate to link to useful material loacated on manufacturer's sites (and occasionally on supplier's sites). --Gerry Ashton (talk) 04:44, 27 January 2008 (UTC)

I now understand the distinction. I have reinserted using 'Lists of ..' and hope that finds favour. TerriersFan (talk) 23:42, 28 January 2008 (UTC)

Linking to Yahoo! News articles

I continually find links to Yahoo! News articles that are dead links. Yahoo! News is infamous for expiring their articles relatively quickly. Most of these links are to Associated Press or Reuters stories. Yahoo! News (and Yahoo! Finance as well) should not be used for this purpose.

I would like to see this become Wikipedia Policy.

If you find a story on Yahoo! News that you want to link to, then find the same story on a site that doesn't expire the articles. There are numerous (probably dozens, if not hundreds of) sites that carry AP or Reuters stories that don't later become dead links.

And speaking of external links...

Among the sites with AP and/or Reuters stories that remain permanent are:

• CNN
• MSNBC
• The Washington Post
• The New York Times
• The Los Angeles Times
• The Chicago Tribune

This is by no means an exhaustive list. I know of several smaller sites that keep the articles as well, but you get the idea.

Just say no to Yahoo! News (and same applies to Yahoo! Finance as well).

Thank you. --angrykeyboarder (a/k/a:Scott) (talk) 00:27, 27 January 2008 (UTC)

COI policy unclear for newcomers

I often remove spammy/COI links (new users or IP addresses whose edits only consist of adding links to a single website), thereby referring to WP:EL. I dare to say that the majority of disputes on external links are about whether or not something is spam. Often, all discussion on whether it somehow fits in the suitable/considerable categories of external links can be bypassed by the fact that it is (likely) a site owner who added the link.

Currently, the section WP:EL#Advertising and conflict of interest is buried somewhere halfway the page and easily overlooked by the good-faith editor who believes that his/her site falls under the category of acceptable external links.

Hence, would it be an idea to add an extra line about CoI under "Points to remember"? Also, the wording "You should avoid linking to a website that you own, maintain or represent, even if the guidelines otherwise imply that it should be linked." is rather vague: (1) "Should" suggests that it can be done, but usually preferably not, and (2) "avoid linking" also invites liberal interpretation. I would suggest a wording along the lines of "You must not link to a website that you own, maintain or represent, even if ...".

Han-Kwang (t) 13:12, 27 January 2008 (UTC)

P.S. IP users trying to add an external link get an automatic warning on the page submission form, referring to the WP:EL page. Again the good-faith editor will see first that his/her site might be suitable, and not read all the way down to the COI warning. I think moving the COI warning further up in the WP:EL page could reduce spam considerably. Han-Kwang (t) 17:09, 27 January 2008 (UTC)

Non-free and registration-required links

Looks like we have some policy warriors going on here. Currently, WP:EL states that it is not allowed to have external links to the New York Times, the Washington Post, the Los Angeles Times, and so forth. This is clearly (a) ridiculous, and (b) not consensus policy, as I have never seen anyone delete links such as these. Does anyone want to defend the practice?

Simply declaring that the current policy of forbidding links "is consensus" is probably not going to be helpful. 69.17.73.214 (talk) 16:53, 27 January 2008 (UTC)

Well, I personally did delete non-free links plenty of times, if only because newspaper articles in many cases don't belong in the external links section anyway. Occasionally, a non-free article is is under external links, but was actually used as a reference for a statement, in which case I change it into a reference. AFAIK references can be non-free. Han-Kwang (t) 17:01, 27 January 2008 (UTC)

Links to pay and registration sites are routinely deleted, and we don't have them here. That's the current norm / consensus, clearly - sorry if you don't think that's helpful but that's how it is. Incidentally the New York Times has recently opened up its entire archive so you can almost always find a free non-registration link to any NYT article. It's only a matter of time before all the others follow suit. Wikidemo (talk) 17:32, 27 January 2008 (UTC)

You are mistaken -- you are required to register to view free content on the New York Times. So are you suggesting that we delete all Times links? Let's just be clear! 128.135.197.189 (talk) 17:36, 27 January 2008 (UTC)

OK, so does anybody want to defend the idea that we should never have a "non-free" link in the external links section? 69.17.73.214 (talk) 17:03, 27 January 2008 (UTC)

there should not be Non-free links in EL sections. what some newspapers do do is have a free registration process. which is OK for the EL section. β^_command 17:13, 27 January 2008 (UTC)

OK, betacommand2 -- you seem to be saying that registration-required links are OK for EL. That's not what you kept reverting about, but then you're a bit of a strong man on these things. 128.135.197.189 (talk) 17:38, 27 January 2008 (UTC)

Sites that require registrations are fine for references if you inline them. Otherwise, we would stop using books as references because you are required to buy it in order to check it out. However, in the external links section adding a link to a site requiring registration is not really needed. Say, an article about some author, with an external link to The Wall Street Journal that requires registration. In this case, the link is not really useful because it may refer to any section of the article itself (the biography, the career, the death, etc), everything or maybe nothing at all (achievements by the author, a section that hasn't yet been created in the article). However, when you add it as an inline reference, you are pointing out that the reference is used to back some claim up. -- ReyBrujo (talk) 17:21, 27 January 2008 (UTC)

Well, we can do an informal poll... Han-Kwang (t) 17:22, 27 January 2008 (UTC)

Poll: non free links

Proposal: 'non-free' links are allowed in the external links section of articles.

• Oppose it is harder for other editors to judge whether the link belongs there if he cannot see the contents without payment. (I'm neutral about content requiring free registration) Han-Kwang (t) 17:22, 27 January 2008 (UTC)
• Oppose. It's hard to put a finger on it, but I think it has to do with convenience to readers and with our being a free encyclopedia. If we link to pay sites, we're really saying that the experience we're offering is going to be better for people who pay than people who don't. We also avoid spam, endorsements, etc., so directing people to a place where they have to pay makes us more commercial. And finally, what's the chance that anyone reading an article is already a subscriber or is going to whip out their credit card just because we sent them there? On a case-by-case basis, very low. So an external link to a pay site is only of use to a tiny fraction of the readers. We require external links to be minimal and of significant interest - so a pay site link almost always would fail that test. If a site requires registration but is free it still has most of these problems. Most people will be turned off by the process so it's not a useful link. And if there's a registration it's not really free - there's an exchange of value. I believe there's wide consensus on this point. Wikidemo (talk) 17:30, 27 January 2008 (UTC)
• Oppose per Wikidemo, word for word. Franamax (talk) 16:26, 28 January 2008 (UTC)

Reg. required links

The original discussion was about registration, but it looks like "non-free" means something else.

Does anybody want to defend removing all links requiring registration from external links sections? 128.135.197.189 (talk) 17:36, 27 January 2008 (UTC)

It helps if you use section headings that are a bit more descriptive than "sigh" (I changed it now). Han-Kwang (t) 17:40, 27 January 2008 (UTC)

Poll - Proposal: should 'free registration required' links be allowed as external links?

• Neutral - I will follow consensus. Han-Kwang (t) 17:40, 27 January 2008 (UTC)

*Bad poll question and stop edit warring. The consensus, and status quo, is that these links are not permitted. The question isn't whether anyone "want[s] to defend removing all links" but whether we should change the rules to permit them. I think not, for the reasons discussed above. A number of anonymous IP editors are now edit warring on the guideline page, never a good thing, on a flimsy claim that this brief conversation taking place on a Sunday morning is establishing some kind of new consensus (despite some clear opposition to boot). If you keep that up we're going to have this page semi-protected. Wikidemo (talk) 17:47, 27 January 2008 (UTC)

Just to be clear: consensus is that no registration-required links should appear on external links sections? 128.135.197.189 (talk) 17:48, 27 January 2008 (UTC)

Can we please not vote on this issue? I think it is much more helpful to discuss, as we are admonished at the top of the page! 128.135.197.189 (talk) 17:44, 27 January 2008 (UTC)

As ReyBrujo mentioned upthread, non-free sites and sites requiring registration are perfectly acceptable as references, provided they're reliable sources. However, external links sections are intended to provide useful additional material, and I agree with Wikidemo's point that registration severely limits the general usefulness of the link. I also want to make the point that WP:EL is a guideline, not a policy page, and there's always scope for talk page discussion to determine that a non-free or register-to-view site may be linked from an article's external links section on a case-by-case basis. But as a general guideline, I think this practice should be discouraged for the reasons mentioned previously. --Muchness (talk) 17:50, 27 January 2008 (UTC)

(aside) I do agree that we shouldn't be throwing up poll questions. Polls do have a few uses but usually not, and only towards the end of a discussion or when there's a deadlock, to get people to come off the fence and state what they think. If you're reading this, Han-Kwang, please keep that in mind because polling can actually make a disagreement worse instead of better.Wikidemo (talk) 18:00, 27 January 2008 (UTC)

OK, so both Muchness and Wikidemo think that all external links that require registration should be removed except through discussion on the talk page? I just want to be clear here. 128.135.197.189 (talk) 17:52, 27 January 2008 (UTC)

This guideline page, which has been stable until today, prohibits such links so yes, that is the consensus until proven otherwise. I agree with that, and admonish people to be slow and avoid contention and edit warring when considering changes to guidelines.Wikidemo (talk) 17:56, 27 January 2008 (UTC)

OK, I will go and clarify on the page that that includes various important newspapers. 128.135.197.189 (talk) 17:57, 27 January 2008 (UTC)

Just a heads up to the annons you both could be blocked for WP:3RR right now. lets give the conversation a few days before attempting to change policy. β^_command 18:06, 27 January 2008 (UTC)

Gosh, that was weird. I was in the process of issuing a 3RR warning when 128.135.197.189 was blocked. This discussion is such a mess to read and rather unconstructive to boot, and one of the main participants is now blocked, so I suggest we declare it closed and if anyone wants to take the issue up again later they may start a clean new section. Wikidemo (talk) 18:10, 27 January 2008 (UTC)

Agree, β^_command 18:12, 27 January 2008 (UTC)

I requested protection at WP:RFPP. Ideally, I did not want to have the ips blocked because they could add another point of view. I don't like having the list of accepted sites requiring registrations there, mostly because it invites people to add other sites. The anonymous came from somewhere, so it is likely more will come. The only problem was that they changed the guideline directly (and created two different threads here, making "staying in focus" harder). -- ReyBrujo (talk) 18:15, 27 January 2008 (UTC)

current discussion on reg-required links

OK, as far as I can tell from the current discussion (let's leave pay links aside for a moment):

• Betacommand2 thinks that reg-required links are fine.
• Wikidemo thinks that reg-required links should always be removed.
• Muchness thinks that reg-required links should be removed, except if talk page discussion resolves otherwise.

My personal view is that it's pretty clear that major sites such as the Times are absolutely fine as ELs as long as they satisfy the other criteria. And it's also pretty clear that that's the working rule on EL sections all over the wiki -- i.e., the page as it stands does not reflect the consensus of people who do not obsessively monitor guideline pages.

For example, a Times book review seems quite OK. Currently, the page as it stands says that all reg-required links should be removed. What do people think we should do next to clear this up? 128.135.197.189 (talk) 18:17, 27 January 2008 (UTC)

In this situation, I'd cite the Times book review in the section of the article that discusses the book's critical reception. --Muchness (talk) 18:26, 27 January 2008 (UTC)

The policy explicitly says reviews and interviews are OK for EL. So, do you want to prohibit review ELs, or registration required sites in ELs? 128.135.197.189 (talk) 18:28, 27 January 2008 (UTC)

We must differentiate the external links from the references. I favor adding links to sites that require registration when they are used as references (specifically, inline references). But I don't like the idea of having these links added to the External links section, because they can refer to the whole article, a determined section of the article, or nothing in particular. For example, a reliable site that requires registration like The Wall Street Journal or The New York Times or The Times would be used as reference and not external link. So, instead of having an external link that says "Times review for The Book (requires registration)" I would have a paragraph in the "Critical reception" section of the article about The Book to say "British newspaper The Times was critical about The Book presentation, although it considered it an interesting lecture.^{[link to The Times review]}" This not only makes these links appear in the references section (which is true because they are being used as a source for the article) but also they remove them from the External links section (which is likely to be spammed with Amazon links to buy it, reviews hosted in GeoCities-like pages, etc) and allows the user to immediately know what The Times opines about The Book, even though he cannot access the site because it requires registration. -- ReyBrujo (talk) 18:33, 27 January 2008 (UTC)

We are drifting off topic. We all agree that reg-req inline references are fine. The question is: should reg-req be prohibited in ELs? ReyBrujo, could you be explicit on what you think about this issue? Allow, discourage, or forbid? 128.135.197.189 (talk) 18:34, 27 January 2008 (UTC)

As I said, there is no reason for having registration required links in the external links section. They can either be used as references, or removed altogether. Consensus in articles can modify that, of course, but as a general rule, if you have registered to the site and know what the article is about, have the courtesy of using it as a reference for those who haven't registered. If you want to put it in one of those three options, discourage tending to forbid. -- ReyBrujo (talk) 18:41, 27 January 2008 (UTC)

As an aside, the New York Times is a bad example because its articles are available without registration.Wikidemo (talk) 18:44, 27 January 2008 (UTC)

That's not true; it may be the case if you're accessing from a .edu IP, but in general, you are asked for registration information. It is the payment demand that has gone away. 128.135.197.189 (talk) 18:47, 27 January 2008 (UTC)

There is a special link URL that allows accessing their pages without registration, which is used by all outlets when linking to them. Show me a URL they require registration, and there will likely be another that makes it accessible without it. -- ReyBrujo (talk) 18:53, 27 January 2008 (UTC)

That's interesting to know, but we're drifting off topic again -- my mistake as well. Let's stay focused. 128.135.197.189 (talk) 18:57, 27 January 2008 (UTC)

The guideline page's prohibition on external links to registration sites is no accident or anomaly. It is emphatic and explicit on the point. "Sites requiring registration" is a heading under the "Links normally to be avoided" section, where it is clearly stated, twice, that registration site links are prohibited. As enumerated item #7 in that section, sites that "require payment or registration" should normally not be linked. I don't see any support for your claim that this rule is some kind of accident or oversight. Further, I don't think your claim is right that these links are often used. External link sections tend to be magnets for inexperienced new editors, and for all kinds of spam, but other than having a high noise content and having to be cleaned out regularly, I just don't think it's a regular practice for articles to include external links to registration sites. If you think otherwise, could you point to some? Wikidemo (talk) 18:38, 27 January 2008 (UTC)

I quite understand your point of view; and I agree that the policy is currently clear and unambiguous. It is also silly and out of line with usual practice on the wiki -- experienced users or not. As for reg-req links -- gosh, everywhere, as experienced users I think know! Helen Vendler and Harold Bloom are two articles that I've looked at recently that both have reg-req links in the ELs. 128.135.197.189 (talk) 18:47, 27 January 2008 (UTC)

Nope. Neither article has any external links to registration-required content. The Harold Bloom article's external links section, though, is a complete mess in need of a clean-up and illustrates my point about newbie editors. Wikidemo (talk) 19:06, 27 January 2008 (UTC)

Oh, right, Vendler's is NYT, yes. As for Bloom: we are in a bit of a bind: any time there's an article with a reg-req it's a case of a "newbie editor" -- according to you! So perhaps you can come up with a better way for me to show you you're wrong? I should get back to work here, but perhaps (since my own attempts are being reverted) you and others can come up with a better version of the paragraph? 128.135.197.189 (talk) 19:10, 27 January 2008 (UTC)

(ec x 5) Do you agree with me that the one who added the link to the external links section should have added it as a reference backing some text to give you an idea of what the contents of the links are? And Wikidemo, the NYTimes requires registration if you don't have a login and have cleared your cache (it actually takes you to this page). I think it has to do with the referral (to prevent deep linking). -- ReyBrujo (talk) 19:13, 27 January 2008 (UTC)

I don't have a NY Times login and it's not asking me to register. My comment about newbie editors is an observation, not a rule. The Harold Bloom article doesn't prove anything except that people can make a mess out of an external link section. Among the 19 links, one is a dead link, one or two is to a site peddling a book, and four are duplicative links to the same person's self-published personal website. Yet none are to a registration site. Bad editing doesn't demonstrate anything about consensus. You haven't supported at all the claim that people are actually linking to registration sites, and frankly, from what I can see your claim to the contrary looks like a simple error of observation.Wikidemo (talk) 19:32, 27 January 2008 (UTC)

The Wall Street Journal is a registration site and is linked in the Harold Bloom article -- the reason you didn't see it is that the article that is actually linked is a copyvio reproduction of its content on the syr.edu site; I am having the same problem as ReyBrujo accessing the NYT article on Helen Vendler without a login. Would you like further examples of ELs that have reg-req links? 69.17.73.214 (talk) 20:26, 27 January 2008 (UTC)

That's zero examples so far, so sure. No doubt there are a few out there, but I seriously doubt the practice is widespread. Wikidemo (talk) 20:53, 27 January 2008 (UTC)

Sorry, that's two: NYT in HV, WSJ in HB. You are being deliberately obtuse. 69.17.73.214 (talk) 21:00, 27 January 2008 (UTC)

You've come up with no examples and no cogent argument to support your position. It's not worth discussing this further with you. Wikidemo (talk) 21:43, 27 January 2008 (UTC)

I certainly agree with your latter point. Please make good on your threats and go away? 69.17.73.214 (talk) 22:17, 27 January 2008 (UTC)

• suggestion: since the purpose of links is to benefit the readers, if the site is important enough, the trivial very indirect cost that the reader has in letting the site owner know his registration information is less than the benefit of accessing the site. As long as all readers can access,and the site is important enough, it should be permitted provided that it is clearly the best site for the purpose, and is so specified on the talk page. the requirement for specification on the talk page is intended to discourage spamming. DGG (talk) 04:04, 1 February 2008 (UTC)

Unprotect page

Can we get an assurance that if the page is unblocked, nobody is going to break 3RR again or insist on making changes to the guideline over anybody's principled objection? I've tried discussing this with User:128.135.197.189 on the user's talk page but that's not going anywhere. If I don't see a sign that this is all proceeding in good faith, then as far as I'm concerned this discussion is over and the proposal to change the guideline is simply rejected.Wikidemo (talk) 19:13, 27 January 2008 (UTC)

Leave it protected, it won't harm anyone. And as I said, there is a possibility that both users have come to this page directed by another page (it is interesting that two anonymous arrive here to make changes to the same section of the guideline at virtually the same time, and just after one was warned for 3RR). -- ReyBrujo (talk) 19:17, 27 January 2008 (UTC)

Protecting policy and guideline pages is a harm in itself - it has to be unprotected sooner or later, and as far as I can tell the only source of trouble is these two anonymous IP editors. I share your suspicion that this whole matter has not been handled by all parties in good faith. My position against changing the guideline to allow links to registration sites is clear. So unless anyone has anything new to add, I'm on to more productive things.Wikidemo (talk) 19:38, 27 January 2008 (UTC)

If you don't want to join in the discussion, you're welcome to leave -- but do please refrain from reverting and undoing other people's work on the page in that case. 69.17.73.214 (talk) 20:27, 27 January 2008 (UTC)

Your word play is not a fair request. My standing position is clear. I have gone well out of my way to patiently explain things to a very misguided editor, but my standing position is clear and I am well within our policies (and defending the integrity of the project) to revert any inappropriate changes made to the guideline. By the way, are you the same user as User:128.135.197.189, or related in some way? Wikidemo (talk) 20:50, 27 January 2008 (UTC)

You are welcome to your position, but if you leave the discussion, please do not return to revert as you have previously done. 69.17.73.214 (talk) 21:01, 27 January 2008 (UTC)

I have done nothing. Your request is improper and you have not been editing in good faith. Please drop this matter.Wikidemo (talk) 21:24, 27 January 2008 (UTC)

I think the best place to accuse me of things is on my talk page (pick one!), as you have been previously accustomed. 69.17.73.214 (talk) 21:27, 27 January 2008 (UTC)

As you requested I moved the discussion here. You crossed the line there and here of civility and personal attacks. This started when you reverted the same section of the guideline page ten times in less than a day in the face of five experienced editors who restored the guideline language and others who disagreed here on this talk page. You still don't seem to get that that's not allowed here judging from your ongoing warnings to me and others that we shouldn't revert your edits. Your contributions on the project page and in this discussion have not been productive. You are wasting everyone's time. As I said on your talk page, if you're going to edit here, please familiarize yourself with and follow Wikipedia's policies and guidelines, WP:3RR, WP:EW, WP:NPA, WP:BRD, and WP:NPA being a good place to start.Wikidemo (talk) 21:42, 27 January 2008 (UTC)

Both IPs are out of Chicago Illinois...so they are either connected or this is the biggest coincidence ever. IrishGuy ^talk 20:59, 27 January 2008 (UDT)

Hi IG -- yes, of course! 69.17.73.214 (talk) 21:01, 27 January 2008 (UTC)

pay-per-view

In contrast to the discussion above: There are now several sites out there, where you can create your own content, and when someone visits that page the site provides payment to the creator, or the site (strongly) advocates to add such advertisements. Examples include hubpages ; ehow ; mywikibiz. It is very tempting to create your own page there, and add these to wikipedia, or provide information on such sites which people like to link, resulting in others linking for you.

There is nothing in this guideline about such sites. I understand that there may be in some cases information on such pages which allows it as a reference or even as an external link, but there will often be issues with (some or all of, and maybe others as well): WP:COPYRIGHT, WP:COI, WP:SPAM &c. (these all include reasons to not include the link, but it is not directly mentioned here, and when I add the link to such a page which is not mine, none of them would strictly apply).

May I suggest that we add something about these sites to the guideline, e.g. something along the lines of: 'It is strongly discouraged to link to sites which provide payment to the maintainer of the site, even if it is not your site'?? --Dirk Beetstra ^{T C} 19:05, 28 January 2008 (UTC)

Major news media sites (such as the Washington Post) carry ads. I am not privy to the contracts between these sites and the advertisers, but I would imagine there is some connection between the number of views and the payment. So your suggested wording could be construed to strongly discourage links to these reliable sites. --Gerry Ashton (talk) 20:38, 28 January 2008 (UTC)

You are right, it should not be the maintainer .. what I meant was the person who published the page on the free hosting site. --Dirk Beetstra ^{T C} 22:29, 28 January 2008 (UTC)

And these sites are worse than Wikia.com, how? -- Konk Republik (talk) 03:59, 3 February 2008 (UTC)

Google books

Sorry for the basic question: Does Google books violate copyright? Is it OK to link to a Google books result of, for example, a page in a copyrighted textbook? Thanks much, delldot talk 07:00, 29 January 2008 (UTC)

Not in my opinion: GB, in the material it displays to the public, does not violate copyright. It as been asserted that some of the internal handling procedures of GB violate copyright (more specifically, the making of an electronic copy of the entire book for its internal storage even though it only displays a short excerpt), butt hat has not yet been brought to a legal decision. Probably it never will, since GB is actively trying to make arrangements with all possible publishers to prevent such litigation. In any event, most material on GB is free from copyright restrict: essentially everything before 2014, and the later material from the large number of cooperating publishers, who control the amount that appears. In general, refusing to link on the basis of copyvio require some reasonable confidence that it is copyvio, and that is not present here. DGG (talk) 04:10, 1 February 2008 (UTC)

Good summary DGG, I had been wondering the same.--Hu12 (talk) 04:15, 1 February 2008 (UTC)

"2014"? --DocumentN (talk) 02:43, 4 February 2008 (UTC)

"Cannot be saved"

The "Restrictions on linking" section currently includes the text: "Sites that match the spam blacklist without being whitelisted. Pages that contain such links cannot be saved." What does the "cannot be saved" part mean?

The text was apparently added in this edit, but I'm not sure where the discussion it came from is. --DocumentN (talk) 02:36, 4 February 2008 (UTC)

What it means is that when you try to save a page that contains a blacklisted spam link, you get a spam filter notice informing you that the spam filter blocked your page save because it detected a blacklisted hyperlink. --Muchness (talk) 02:47, 4 February 2008 (UTC)

Makes sense. I've attempted to clarify the wording with this edit. --DocumentN (talk) 02:57, 4 February 2008 (UTC)

Linking to Adult sites

Is there a Wikipedia policy about whether it is appropriate to link directly to a subsection of a bigger website, if the subsection is marked 18+ due to its adult content and the direct link bypasses the age restriction warning within the website. Meowy 13:21, 5 February 2008 (UTC)

TfD nomination of Template:FreeContentMeta

Template:FreeContentMeta, which creates external link boxes similar to those of Wikimedia's sister projects, has been nominated for deletion. You are invited to comment on the discussion at the template's entry on the Templates for Deletion page. Thank you. — Jpatokal (talk) 06:06, 6 February 2008 (UTC)

Any site that misleads the reader by use of factually inaccurate material or unverifiable research. See Reliable sources for explanations of the terms "factually inaccurate material" or "unverifiable

I would like to suggest that this guideline is vague and highly open to interpretation which leads to subjective views distorting the "truth" of a given page. I think this guideline should only be applied if there is a deluge of external links on a given article which require pruning. Otherwise subjective opinion on a given subject is perfectly valid if the information in the main article is subjective or contains unverifiable claims. I believe that it is important to give power to the reader by not deciding what constitutes as "unverifiable" for them.

This guideline merely gives power to those who can decide what is "relevant" and what is "irrelevant" thereby heavily slanting articles in favour of the administrator who commonly edits them. See: Prem_Rawat - Teachings and the lack of a criticism section or critical external references within this article as an example. Icky Media (talk) 20:43, 7 February 2008 (UTC)

Why would we want a link to a misleading site? External links talk page should not be used by editors as platforms for their personal views.--Hu12 (talk) 21:04, 7 February 2008 (UTC)

Minimum removal

In December, the page was changed from saying 'a minimum of links' to its current version of 'restricted to'. Was there a discussion somewhere? The new version is ambiguous to allow for a lot of links, as long as they're relevant, while the previous explicitly urged a small number. I prefer the old version, or a more explicit discussion that it should be a minimum, but would generally like some clarity. Anyone got a link? WLU (talk) 21:26, 8 February 2008 (UTC)

User making the changes was Leranedo, with less than 2K of edits, and is now apparently dead, and there seems to have been no discussion. Anyone object to me turning it back? WLU (talk) 21:30, 8 February 2008 (UTC)

Wikipedia:Linking to other wikis

Requesting comments for a new style proposal for wikis listed in the EL section is at Wikipedia:Linking to other wikis. Everyone is encouraged to leave feedback. -- Ned Scott 05:37, 10 February 2008 (UTC)

Style guideline for official websites

Official websites fall under What should be linked, but there is currently no style guideline on this page for linking to official websites. For example, I've seen (and have probably used at one time or another):

• Official Website
• Official website
• Wikipedia's Official Website
• English Wikipedia's official website
• Wikipedia - official website

Is there any consensus for establishing a consistent style of linking to official sites across Wikipedia? Should there be a consistent style or template added to the style guidelines? I would think that consistency would only make the encyclopedia easier to navigate. dissolve^talk 10:43, 13 February 2008 (UTC)

While a standard may be a good idea, we have such a massive amount of variation now it seems like almost masochistic to try to 1) come to agreement and 2) then get everybody to do it that way in the future, 3) then fix all the existing ones. 2005 (talk) 10:50, 13 February 2008 (UTC)

Definition

In the Links to normally be avoided section on the subject of other Wikis it states "Links to open wikis, except those with a substantial history of stability and a substantial number of editors.". What is the definition of both occurrences of "substantial" in the sentence? - X201 (talk) 18:24, 14 February 2008 (UTC)

There is no set definition - editors need to use common sense and good judgment. The main point is to ensure we only send readers to Wikis that we have good reason for thinking provide high quality, NPOV information that is likely to be accurate and remain so over time. The definition of substantial will depend to a large extent on the subject area - for instance an area that has a lot of practitioners/interest/study around it would need a larger number of supporters than one which is a very niche area. But some subjects may just not have enough critical mass to develop a suitable site as an open wiki. -- SiobhanHansa 21:35, 14 February 2008 (UTC)

New external link templates

Please take a look at {{externalimage}}, {{externalvideo}} and {{externalaudio}} and how they are being used in articles. If these uses seem incompatible with the current wording of WP:EL, please invite further discussion here. Kaldari (talk) 02:55, 16 February 2008 (UTC)

wp:links

This page Wikipedia: External links implies (in the top template) that "wp:links" is a shortcut that redirects to this page. But that shortcut actually redirects elsewhere.

Should we fix the redirect so it actually does link here? Or remove the mention of "wp:links" in the template on this page? --68.0.124.33 (talk) 01:00, 19 February 2008 (UTC)

WP:LINKS (all capitalized) forwards here, while WP:Links (not capitalized) doesn't. While I don't think this is ideal, it's in the nature of a system that finds capitalization important. Personally I believe it is more useful to educate users to pay attention to capitalization (assuming we aren't going to change the software to ignore capitalization in most cases) than to ensure every possible combination points to the same place. -- SiobhanHansa 03:10, 19 February 2008 (UTC)

In a similar vein, is there any reason why WP:XL doesn't redirect here? I've used "xl" as an abbreviation for "external link" in edit summaries for some time and I must have picked it up from somewhere, so I was rather surprised that it didn't work as a shortcut.  —SMALLJIM  10:51, 21 February 2008 (UTC)

Question about criteria 4

In number 4 of what should be linked, it says 'reviews and interviews.' Interviews are in as far as I'm concerned, inherently notable and if not reliable, then a definite candidate. However, for reviews, there's an ongoing discussion at WP:NOVEL style guidelines about the comparative reliability and notability of reviews. I am of the opinion that the reviews should be ones in reliable sources - Time, Newsweek, national newspapers and other sources considered reliable. Several other editors has made the case that any review is better than none, even if not professional. Accordingly, I wonder about a) an opinion that my standard is excessive, or b) if the community agrees, the wording adjusted to say "interviews and professional reviews". Any thoughts? WLU (talk) 22:20, 23 February 2008 (UTC)

And to clarify, we've not been talking about reviews from such sources at Amazon or Epinions, etc. We're specifically talking about reviews from FantasyLiterature.net which is my own review site. I have invited reviewers who I consider excellent -- one is a "professional" (paid by genre magazines) and others are writers (some award-winning), a student who is finishing her master's in literature and plans to be a critic, etc. I have a Ph.D. in psychology and neuroscience, have been published in scientific journals, and teach research methods and scientific writing at my university (the point is that I care about facts and excellent writing). Kahooper (talk) 23:28, 23 February 2008 (UTC)kahooper

The 'genre' professional - is the genre fantasy? Have they been published in major newspapers or magazines? How would answers to these questions line up with the links to be included/considered? I see it as excluded but others have disagreed. FL.net also advertises, and profits from, traffic on the site (though kahooper has said it's been changed to minimal). I'm also interested in a broader input here, not just this one external page. WLU (talk) 23:37, 23 February 2008 (UTC)

Yes, fantasy genre magazines: Blackgate, Mytholog, Dragons, Knights, and Angels, and also non-genres like Rosebud Magazine. Rob Rhodes is also co-author of the forthcoming "The Sword in the Mirror: A Century of Sword & Sorcery" and has a story coming out in a fantasy anthology.

I realize that you're interested in the broader issue of what could be linked, but I did want to distinguish us from Amazon, etc.

About profit: professional reviewers (who you say should be linked) do profit. WE DO NOT PROFIT. We might off-set expenses with any Amazon kickbacks. If we were in it to profit, we would give all good reviews, wouldn't we?? Kahooper (talk) 00:37, 24 February 2008 (UTC)kahooper

I think what I see as your general intent with this wording - that not all reviews are good enough is appropriate, in keeping with Wikipedia's policies and generally accepted practice. But I'm not sure changing the wording here is the way to handle this. We already talk about wanting all links to be reliable and from knowledgeable sources. And we talk about avoiding Links to blogs and personal web pages, except those written by a recognized authority.

Personally I think WikiProjects may be the best places to really come up with appropriate guidelines for what works within the area they concentrate on. A more general guideline focused on reviews could be very difficult to word such that it can be applied appropriately to different genres. At WP:WikiProject Albums, for instance, they require professional, independent reviews, though "professional" for them explicitly includes volunteer staffed productions. This may be a very different standard than is appropriate in other arts. At the very least I think it would be a mistake to develop additional wording for this guideline without soliciting comment from all the active Wikiprojects who work on subject areas that use reviews. -- SiobhanHansa 02:02, 24 February 2008 (UTC)

Edit warring again on external links to map services

We have three editors (so far) warring over whether to generally prohibit external links to mapping services.

The problem

A couple weeks ago one editor inserted a new item in the list of "links normally to be avoided": links to external map services where similar information is available on many other sites.[2] This came after a long and rather bitter discussion between him and another editor on the subject, that spilled over into an edit war/mass-reversion and AN/I report in late January (here). Yesterday the change was reverted three times in six hours by its primary opponent and by another editor with whom she often collaborates (and obviously, reinstated twice).

The proponent of the change claims it's already consensus and policy that all geographic locations should be geocoded, and external map link services generally avoided. The opponents say that no agreement has been reached on this page. Both sides argue substantive reasons why external map service links should or should not be avoided.

My opinion and analysis

Consensus

It appears that in a broad sense, there is consensus on Wikipedia as a whole that articles about geographic locations should be geocoded using a template that links to the "GeoHack" service. Nearly all of the 300,000 mapped locations on Wikipedia use this feature, so that when a user clicks on the coordinates in the upper left of the screen (for example, Moai), it pulls up Wikipedia's "GeoHack" map service.[3] The exceptions are very few, generally stub and brand new articles created by inexperienced editors, or old articles created before the GeoHack service was available and that have not yet been updated.

There appears to be a consensus nearly as broad that external map links should be avoided. Of the 300,000 articles, only 12,000 contain external map links. Of these perhaps half have a specific (ostensibly) encyclopedic reason for an external map link, for example Fruitland, New Mexico, which links to a census map and Mapquest to show a disparity of naming, or Intercourse, Pennsylvania, where a specific terraserver map link and scale is chosen to illustrate that the small village is "surrounded by farmland." The other 6,000-odd map sevice links seem to have no special purpose, e.g. Great Stirrup Cay or Maudslay State park.

It is not clear whether there is a consensus on this page for or against discouraging external map service links. The recent discussion was long, disorderly, and contentious to the point of incivility, without many participants. I don't think any meaningful conclusion can be drawn from it. Over at WP:GEO, the project that deals with such things, the consensus appears to be to delete external map service links and to insert GeoHack coordinates in their place. They have done this with more than 50,000 articles in the past several months.

Substance

I don't see the advantage of allowing specific map service links as a general matter. There are plenty of disadvantages, the main one being lack of consistency and predictability. If I happen to like google maps or topo maps, I should expect that any geographic article lets me click in the upper right hand corner, then click down somewhere on the page, to display my map (in the future, GeoHack could set this up as a matter of user preferences as well). Another reason to avoid them is to avoid becoming a link farm. We can't link to every single major service (and even doing that would be a bias against the minor ones), and the more links we add the more we clutter up the external link section, which is supposed to be kept short and manageable. I also don't like the favoritism of directing dollars to one commercial enterprise over another - when people visit a map service it gets traffic that turns into revenue from their targeted ads. The services also use cookies to track visitors so they can target ads to them.

There is little to no advantage for the article's editor to hand-pick a single map service to display. Most of the examples I give where there is some reason (e.g. showing that there are farm fields around Intercourse, PA or that the government and commercial map services use different names for a feature) or saw (showing route maps, showing that a local road is closed) are not cited properly in the article for the claims and constitute original research citations to primary map sources. In addition many of those claims would be equally possible to illustrate with any map service. In the few cases where there is a good reason to use a map service, the language change does not apply. The entire section says that there are links normally to be avoided. One can always add an external link that actually has encyclopedic value, and specify on the talk page, article page, or in a comment that there's a special reason. Even allowing that there might be 6,000 legitimate external map links out of 300,000, that's well within the language of "normally" avoiding them.

If we don't tell people that external map links are generally discouraged, the harm is that new and inexperienced editors will add them, and it becomes a maintenance problem that people must constantly work on. With a statement here on WP:EL that reflects the reality of the Wikipedia practices, those editors might notice, and if they do not it is easy to point them to the guideline rule. The only harm to adding it is that a few people might misunderstand the rule as a blanket prohibition, so people wanting to add these links will have to do some extra work to justify themselves (which they ought to do anyway).

Looking at analogous situations, we generally do not allow links to external reference sources where there are many different sources for the same information and choosing one link would favor one over the other. We tell people they can use google or their other favorite search engine. Where there is a universally recognized meta-service (e.g. for films, metacritic, rottentomatoes, or imdb) or a feature within Wikipedia (wiktionary, for word definitions) we refer people to that rather than choosing specific for-profit sites (New York Times reviews, Meriam-Webster dictionary, etc). That system works just fine, even though there are occasional reasons to include these links that we normally avoid. It will work fine here too.

A final note on consensus. I think this is already consensus. We don't need to reach a consensus agreement here on this page to recognize something that is already consensus elsewhere on Wikipedia. Guidelines are supposed to reflect consensus, only seldom to actually create consensus. It's a little much to think a few contentious edit warring editors here should shape the rules for 300,000 articles rather than the editors involved and the main wikiproject. The only issue here ought to be to figure out what consensus actually is, not to reach an agreement on the subject.

- Wikidemo (talk) 23:20, 23 February 2008 (UTC)

1) I have almost no opinion about the substance of the disagreement. 2) Para's insistence on adding this text is inappropriate since, to say the least, there has been no consensus on this page to do so. 3) The wording suggested speaks in the negative rather than the positive. Saying a link to a geo-hack map or whatever it is called is appropriate, that is the way to go, NOT to say the other links are inapproriate, since in some cases there are as Wikidemo has stated. Other types of map services should provide something of merit the generic thing does not. if they don't, they are redundant to the generic thing, and like we sometimes use a dmoz link when there are tons of similar choices, then the individual services should not be used strictly because they add nothing. Another way to look at it is we say official sites are good links. if some other site has the exact same info as another site, it would not add anything and should not be linked. So, by stating the generic thing as a positive, we limit links to specific maps that offer unique info for some location. 2005 (talk) 04:41, 24 February 2008 (UTC)

Would you mind clarifying (sorry if I'm being dense here), would you favor a positive statement about map links that says the that the GeoHack link should be included and that specific map services may be included as well if they add something that the generic version does not? That makes sense too and perhaps that could satisfy all parties. Wikidemo (talk) 05:17, 24 February 2008 (UTC)

I have no idea what the Geohack thing is, I don't want to comment more on that specifically, but it sounds like a line could be added to what should be linked to say a generic geohack link. This then does not prevent a more specific map service that adds more "amount of detail". 2005 (talk) 05:42, 24 February 2008 (UTC)

At one time the geohack-like tools linked to pages that had five or ten location-specific links to map services; I don't believe those were controversial. Currently, however, the "Geohack thing" provides links to pages like this one (a small town in the United States), this one (Vienna), and this other one (a small park in the United States), which present a comprehensive but bewildering array of links to geographic information sites all over the Internet. In addition to the "standard" map/imagery sites for the world and the particular geographic area, it includes many links to derivatives of those "standard" map/imagery sites, various Google Earth applications, non-map geographic sites such as the Degree Confluence Project. Much of the page lists links specific to one part of the world, meaning that most of them will be irrelevant to any given location, and at any given time at least some of the allegedly global links do not serve information specific to the requested location. GeoHacks is amazing, but for the vast majority of users it is likely be received as Information Overload. (I suspect that most users who visit Geohacks will simply click on the first Google Maps link.) The pushback on its use comes from editors who want to provide smaller focused collections of map links that are specifically relevant to the article. --Orlady (talk) 17:45, 24 February 2008 (UTC)

The example I provided kind of proves the point:

• The article - Moai)
• The Geohack link - 26-6-51.23S 109-21-50.38W

It's an article about the Easter Island statutes. Most of the geohack links are useless because the services don't cover this part of the world in any detail. After lots of clicking you find one that actually has good map detail only to find the link goes to the geographic center of the island, not an aerial view or map of the statue locations. It would great in this particular case to guide the reader to a pre-chosen viewpoint or map if there is one. There's probably a way you could trick out GeoHack to do this (let the article editor choose a default view). You could also trick it out to let each user choose their favorite map view (e.g. Google hybrid, zoomed to maximum resolution) and have that come up as the default view. But so far, no. Wikidemo (talk) 18:17, 24 February 2008 (UTC)

The map provided in that Moai article is infinitely more useful than any of the Geohacks links. If I were involved with that article, I might be tempted to delete the link to the geohacks page, but it's likely that some helpful 'bot would restore it. --Orlady (talk) 18:51, 24 February 2008 (UTC)

I agree, I just lost an hour of my life playing around trying to find the Moai by satellite...but remember that the geohacks link is just a templated geographic coordinate. So it's worth having that coordinate information in that template, even if someday we change the way it gets displayed. The issue here is it's not really a single point location, so it ends up pointing nowhere special on the island. Some locations maybe ought to convey size or shape, e.g. the Las Vegas Strip is a couple miles long, or the Mississippi River is a couple thousand. Both of those have some really cool maps, but map links to specialized maps or even map services that have been set up with the right scale and view could be helpful. Wikidemo (talk) 22:59, 24 February 2008 (UTC)

How can we discourage people from adding links to their favourite map services with a positive statement? Nobody has argued that the general link shouldn't be included, but that one or more of the services behind the general link should be chosen subjectively and linked from articles directly. The 16 (or now 15) items in the section "links normally to be avoided" on the guideline page does not make an absolute ban on those links, but as the page says, "Each link should be considered on its merits, using the following guidelines". (Let's just talk about linked coordinates from now on instead of GeoHack, that's more descriptive than the name of the piece of software that shows map sources for a location.) --Para (talk) 18:31, 24 February 2008 (UTC)

How does it not? I explained it, and the point is rather obvious, so what do you have a problem with? I suspect your issue here is you are unfamiliar with this guideline. The bottom line is the guideline says you can't just add links because you want to. they have to go over and above what is in the article, or the official site. So it is quite obvious that an instruction for a generic map link makes it a violation of the guideline to link to specific maps in most cases, since the specific map will add nothing over and above. 2005 (talk) 00:30, 25 February 2008 (UTC)

The people supporting linking of their favourite map service from articles directly are opposing removal because of "ease of use", which in their mind adds to articles because the maps behind the links are additional information and related to the article, regardless of all the issues with such a practice and the availability of the same link behind the general link. See for example this revert. It would be very difficult to discourage such convenience links with a positive statement. --Para (talk) 09:14, 25 February 2008 (UTC)

Why would a consensus on this page be needed, when there is already consensus that Wikipedia is an encyclopedia for the dissemination of information, that Wikipedia has a neutral point of view and is free content, that Wikipedia should be accessible to the widest possible audience, and that Wikipedia content should be consistent, not arbitrary. I have claimed that those points are against direct inclusion of external map links in articles, and nobody has justified why they would not be relevant or sufficient. The only argument against all that was that Wikipedia should have related information available as easily as possible, which was shown not to have limited external links cleaning in Wikipedia previously, so it shouldn't now either. Since most of the community is reluctant to participate in discussion, should we start counting how many people have removed external map links and how many of those have been reverted? --Para (talk) 18:31, 24 February 2008 (UTC)

I would argue that if the matter has been specifically considered before, or elsewhere, or is clearly the common practice, then we know there is consensus. However, it is not consensus to say that a certain result follows from policy, based on your interpretation of policy, when other people disagree. Interpreting policy takes consensus. As to your rhetorical question, you could try to say as specifically what you mean and see if that can go in the guideline or commentary, something like: "don't add a link to a mapping service simply because it is your favorite; to justify a link to a mapping service, that specific link must have some special encyclopedic value." Wikidemo (talk) 18:38, 24 February 2008 (UTC)

People may disagree, but their interpretation needs justification as well. The disagreement to the above points came down to idontlikeit. If nobody can substantiate why an interpretation of policy would not be accordant with that policy and there are not enough contributors to show a numeric consensus, then the interpretation is clearly a clarification of that policy. I added a section to demonstrate the prevailing consensus. --Para (talk) 23:46, 24 February 2008 (UTC)

No they don't, and you are just being dismissive about this anyway. Before the past day or so only three people really commented on this, and the person opposing you has written essays in opposing. You want to add something. You need a consensus to add it, period. There has not been one. 2005 (talk) 00:35, 25 February 2008 (UTC)

Para, you are quite single-minded in your views on this, and that's fine, but the fact that you disagree with other editors over policy does not give you the right to unilaterally declare a consensus. It's exceptionally tedious to have you continue to claim that this discussion amounts to a consensus in your favour, when all it amounts to you is you dismissing the views of everyone who disagrees with you. You can open an RfC/policy if you want to get a broader view on things, but for now it would be much more productive to concentrate on the points where there is agreement. --BrownHairedGirl (talk) • (contribs) 02:02, 25 February 2008 (UTC)

Please see #Issues with inclusion or exclusion of map service links. Nobody showed reason to dismiss my interpretation, and all claims to the contrary are therefore entirely false. You cannot dismiss a justified interpretation with a simple "no". The opposing person is indeed quite verbose with her personal opinion, but the single concern she has doesn't become any stronger with that method, and it was already shown not to have been a problem on Wikipedia before, therefore not blocking the originally proposed guideline. --Para (talk) 09:14, 25 February 2008 (UTC)

Para, there are plenty of things that could be said about that, but I for one am not prepared to try to waste time discussing the substance of these points with you ... because I tried at great length to do so on useability issues, and it was one of the least productive discussions I have ever had on wikipedia. You firstly claimed that my concerns didn't make sense and needed expansion, so I put in a lot of time explaining them in great detail, only to find you simply dismissing them as "false" or "nonsense". We then got into a bizarre situation where you claimed that you were nothing to do with the development of the GeoHack, then chided me for not having noticed some recent tweaks to the GeoHack system, which it turned out had been done by you ... and you then proclaimed that the problems identified had been "resolved" without bothering to listen to any input from the people who had described the problems, let alone having the courtesy to ask whether the changes had helped.

Your interpretation of policy involves stringing together a long list of policies to reach a conclusion not stated in any of the individual policies. It's up to you to find support for that synthesis, not to dismiss objections simply because other editors are not prepared to devote hours of their time to explaining the logical flaws to someone whose modus operandi is to simply to dismiss all objections as wrong and then unilaterally proclaim a consensus. --BrownHairedGirl (talk) • (contribs) 17:10, 25 February 2008 (UTC)

BrownHairedGirl has a problem with the interface, yes, that's quite clear by now and we can put the mention to rest. She is free to work on her issues somewhere else with someone else who doesn't mind listening to repeated hyperbole about having to scroll 13 screens. This is however not the place. As I have repeatedly pointed out, the issues she has with the interface were not a problem with the external link list interface for information about books, which set a precedent to follow with other topics where general links are possible, such as external links to map services. If a contributor has no time to get involved in discussion, it's better to opt out altogether rather than make uninformed comments. Unilateral opposition of link removal based on nothing but personal preferences can and should be ignored. --Para (talk) 17:49, 25 February 2008 (UTC)

A possible way forward on map service links

In extensive discussions on this page, there has clearly been no consensus to change the guideline to deprecate direct links to map services. Even Para (talk · contribs), who made the change merely claims that there "there were no valid reasons for ignoring existing policy" ... but that is based on Para's interpretation of a group of of policies, and that interpretation does not have consensus. If Para wants to push the point, then a policy RFC would be in order.

If I was trying push my take on the issue, I would add something to the guidelines to the effect that despite the claims of a few editors involved in its maintenance, the useability of GeoHack is still very poor, and that the policy concerns are grossly overstated, so direct map links may be added if an editor feels that they are particularly helpful. However, I don't think that would be at all helpful; it would merely be an inverse of Para's attempts to impose his POV, and since neither view has consensus, we shouldn't be amending the guidelines as if there was.

Most of the discussion to date has focused on the pints of (often heated) disagreement. I suggest that it would be helpful for us to try approaching things from the other angle, by first updating the guidelines to reflect the points where there is already agreement.

If other editors are willing to try that approach, I suggest that one way to do this would be to start drafting a small sub-section of the guidelines which would be solely about map services, something which would uncontroversially note the issues agreed in the discussion so far, something in the spirit of the following quick draft:

• On geographical articles, links to external map services should be provided through the GeoHack system, which provides a comprehensive and standardised list of links to many different mapping and imaging services
• There is no consensus as to whether it is also appropriate to add other links to external map services. However, the general principle of this guideline applies: that they should be kept to a minimum of those that are meritable, accessible and appropriate to the article

What I have tried to do in those two points is not impose one side of an argument, but focus on the points on which I think that there actually is agreement. (Sorry if I got that wrong)

The above wording is only a quick and rough draft, and would obviously need polishing if we were to try using it. We have debated the substance at huge length, and I don't see nay chance of an immediate resolution to the disagreement, so rather than get stuck straight back in to a debate about the substance, I have tried to focus on the points where I think that there actually is a clear consensus. I hope that those two points interpret other views correctly, and would welcome correction if I haven't got things right.

May I ask other editors to please consider this suggestion in the spirit in which I have intended it, of a compromise suggestion for a way forward by placing the initial focus on the areas where there is agreement, and leaving the disputed issues for later? --BrownHairedGirl (talk) • (contribs) 17:49, 24 February 2008 (UTC)

That looks good to me. Perhaps we can phrase the "no consensus" thing positively, instead of saying there is no consensus to include or exclude, we can say that there is consensus that there is no absolute rule. Meaning, something like "Links to external mapping and imagery services, if any, should kept to a minimum and limited to those that are accessible and add to the encyclopedic coverage of the subject matter." (I tried to define what "meritable and "appropriate" might mean, without getting too specific). Also, I don't know if we need to say it but it would be good if map and image links could all be kept together in a predictable place, perhaps even with a template (we could have a google map template, for example), rather than scattered randomly throughout the external links section. Wikidemo (talk) 18:24, 24 February 2008 (UTC)

What is the desired outcome of stating on a guideline page the current inconsistent status of map links? How does it guide people who have come to read about Wikipedia's external links? Why would they bother generalising the location information when they already have a link to the map service they are most comfortable with? --Para (talk) 19:20, 24 February 2008 (UTC)

In response to Para, the purpose is to set out the current state of play, so that anyone looking for guidance knows what the current consensus approach is, and where there are disputed issues. As to why bother generalising, I'm a little surprised that you question that! The purpose is to offer a choice to readers who want a choice, and/or may be familiar with the geohack style of links ... and also to make it clear that there is a consensus (so far as I can see) that geohack links should be included (distinguishing the consensus on that point from the lack of consensus on the removal of other links).

In response to Wikidemo, yes, I like the idea of rephrasing the lack of consensus in a positive sense as "no absolute rule". I deliberately took the "meritable, accessible and appropriate to the article" language as a direct copy of the text at the top of the page, presuming that language already had consensus. I think that idea of a more specific wording is a good one, and I like your phrasing "Links to external mapping and imagery services, if any, should kept to a minimum and limited to those that are accessible and add to the encyclopedic coverage of the subject matter."

The idea of a template for direct links to specific map services sounds to me like a good one, but I'm not so sure about the idea of centralising the other links. It's good in principle, but I struggle to think of cases where there would be more than one or two of them, or maybe three as an absolute maximum, and I'm wary of anything which might encourage editors to create a "Map and image links" section, which I fear would have a tendency to grow. Have you any thoughts on how the idea of keeping these links together might work? --BrownHairedGirl (talk) • (contribs) 19:39, 24 February 2008 (UTC)

We should probably table the question of exactly where the links go, because that's adding another open question in something that may be difficult to resolve to begin with. It's really a question of how we organize external links, something nobody's really addressed. Wikidemo (talk) 23:03, 24 February 2008 (UTC)

On both those points, isn't there a danger here of being over-specific, and generating instruction creep? --BrownHairedGirl (talk) • (contribs) 23:30, 24 February 2008 (UTC)

You might be interested in WT:GEO, where external map links through templates have been used for years, and the removal discussed for about half a year now. Wikipedia talk:WikiProject Geographical coordinates#Geolinks coverage shows how badly the template experiment failed, and the related consensus resulted in the removal of all external map links from templates. --Para (talk) 23:46, 24 February 2008 (UTC)

That discussion is about something else, namely an attempt to add specific links through GeoHack, which is problematic because it can't be effectively globalised. That's a rather different issue to what Wikidemo is proposing, which is a template for a specific direct link to a specific map service. --BrownHairedGirl (talk) • (contribs) 02:10, 25 February 2008 (UTC)

Please read the linked page more carefully. In all the discussions on that page, "Geolinks" refers to a set of templates with map service links chosen by a handful of editors, and placed in various location related articles. After years of narrowly maintained linksets, the templates and their use was inconsistent in just about every possible way, in addition to all the issues with such linking. It's impossible to standardise such a scheme to all location related Wikipedia articles when nobody is maintaining the localised and scale dependent linkset to work with all those articles. Multiple templates for single services are better than plain links only visually, and are contrary to all other efforts in standardisation and consolidation of Wikipedia's external linking. Wikipedia content should be consistent, not arbitrary. --Para (talk) 09:14, 25 February 2008 (UTC)

Please read this guideline. "ease of use" is not a criteria, so again your comment just avoids the issue. A positive statement quite obviously can cover the issue, and easily in fact. Your insistence that black is white does not help your cause. At this point your own insistence that neither existing guidelines and opinions of other matter in contrast to your assertion is what is keeping this rather easy to handle molehill into a moutain. 2005 (talk) 00:37, 26 February 2008 (UTC)

Evidence of existing consensus for removing map service links

I looked through some recent changes on external map service links, and found the following edits:

As Wikidemo pointed out above, if there is a clear common practice, then we know there is consensus. It's pointless to rehash the dispute endlessly on this page when the common practice is obvious, so we should just reinstate the removed guideline to hopefully reduce the amount of work on link removal. --Para (talk) 21:31, 24 February 2008 (UTC)

Para, selective list-making can be an interesting way of passing the time, but it doesn't demonstrate a consensus. --BrownHairedGirl (talk) • (contribs) 23:28, 24 February 2008 (UTC)

Feel free to add to the list if you think I haven't found all related edits. --Para (talk) 23:46, 24 February 2008 (UTC)

This list adds nothing to the discussion here. It's not even on the topic really, which is the phrasing of encouraging appropriate map links while discouraging inappropriate ones. 2005 (talk) 00:41, 25 February 2008 (UTC)

It shows clearly how the well informed Wikipedia community handles external map service links, and is in contradiction with the newly proposed "guideline" of inconsistent status quo. The prevailing well justified practices are to be followed, and alternate personal practices discouraged. --Para (talk) 09:14, 25 February 2008 (UTC)

I don't think there's any dispute that there are 300,000 or so articles with map links, of which 12,000 or so link to map services, right? So we don't need to prove that the ratio is 25 to 1. The question, to me, is whether the 24 out of 25 ought to be stated as the general rule, and whether the 1 in 25 (or whatever the ideal number might be) is worth calling out so that people don't take the rule as absolute. And if we do call it out, how and in what level of detail? 1 in 25 is a small percentage, but 12,000 is a lot of articles (as is 50,000, the number that were apparently edited recently). Wikidemo (talk) 19:40, 25 February 2008 (UTC)

The fact many editors think many links to external map services don't merit linking is not relevant to the discussion. Pretending it does is just time wasting. The issue is forumulating a consensus wording to guide editors, particularly those editors not highly focused on this issue. You are taking the rather absurd position that because some editors remove externaal map links that YOU can add any wording to this document YOU want! C'mon. 2005 (talk) 00:47, 26 February 2008 (UTC)

ISBN similarity

In scattered discussions about GeoHack, aka {{coord}}, when compared to an article link such as "Click to see location in Google Maps", it has been pointed out that the GeoHack page serves a similar purpose as the WP:ISBN list produced by clicking on an ISBN. Linking to a commercial book selling site is discouraged in WP:EL, with an ISBN being permitted. Cellulose man argument: I notice that EL does not forbid linking directly to a specific library's entry for a book; would it be acceptable for articles to include "Click to see Library of Congress data" and "Click to see Springfield Public Library data" links? -- SEWilco (talk) 20:39, 25 February 2008 (UTC)

• Noted exception: EL does allow linking to media, so an article on The Battle of New Orleans might link to a map of the battle, or to a public domain copy of a book on the battle. That is different from the more general task of finding the location of the battle. -- SEWilco (talk) 20:39, 25 February 2008 (UTC)

Like GeoHack, the useability of the ISBN links is also pretty atrocious; it's so bad that I don't use it, because it's much easier to simply copy-paste the ISBN number into the search engine of a major library.

There is a very good reason for not allowing direct links to sites such as Amazon which exist to sell books, because that is inviting the reader to hand over their cash to a particular bookseller. The same doesn't apply to the free map services, where the reader is not being asked to fork out.

So the way I see it is this: the ugly information overload of the isbn linking system is a neccessary evil, partly because of the predominance of booksales sites, and partly because its major use is in references such such as {{cite book}}, which already form a generic template solution, making specific links impractical.

Also, I can see very little benefit in an direct link with an ISBN number, because they only need to be consulted if the reader is checking references; the casual reader of a wikipedia article is highly unlikely to want look up the ISBN number of a reference, and I for one have only ever used the ISBN number off a wikipedaia article to check a reference. On the other hand, a map link offers a lot to the casual reader of a geographical article, who can use maps and imaging services to learn a lot more about the subject of the article.

So the two situations appear similar only if one starts by focusing on the technology of GeoHack or the ISBN lookup system. Unlike ISBN lookup, map links are of direct benefit to even the casual reader, and unlike ISBN links to amazon, most of the map services liked to are not trying to extract money from the reader. --BrownHairedGirl (talk) • (contribs) 03:12, 26 February 2008 (UTC)

Video game external links discussion

There's a WP:EL-related discussion starting up at Talk:Baldur's Gate#Sorcerer's Place link that may be of interest to editors here. Additional input is welcome. --Muchness (talk) 05:55, 24 February 2008 (UTC)

Imdb

I am sure this has been discussed before, but since imdb is a .com site, should we be using it in external links? If not, why do we have Wikipedia:IMDb templates? Thanks. Student7 (talk) 17:37, 24 February 2008 (UTC)

There's no prohibition against .com sites. They are as welcome as any other in the external links section. imdb seems to have pretty broad community consensus for certain uses, so the template is an easy way for editors to include it in a standardized fashion. -- SiobhanHansa 18:17, 24 February 2008 (UTC)

Explanation request

Could someone explain to me External_links#How_to_link? Thank you--mrg3105 (comms) If you're not taking any flak, you're not over the target. 12:19, 25 February 2008 (UTC)

That section is a set of illustrations of the code to type in to make links look like the given examples. Which bit of it do you find needs clarification? -- SiobhanHansa 15:52, 25 February 2008 (UTC)