User talk:Para/2008-08-08

Welcome to Wikipedia!

Welcome to Wikipedia, ! I'm Celestianpower. I noticed that you were new and/or have yet to receive any messages so I just thought I'd pop in to say "hello". Hello. Wikipedia can be a little intimidating at first, since it's so big but we won't bite so Be Bold and get what you know down in microchips! If you do make a mistake, that's fine, we'll assume good faith and just correct you: it'll take a few seconds maximum! Here, however, are a few links to get you started:


 * How to edit a page
 * Editing, policy, conduct, and structure tutorial
 * Picture tutorial
 * How to write a great article

There are lots of policies and guidelines to get to grips with but they all make your life easier and your stay more fun in the long run. If you have any questions, feel free to leave me a message on my talk page or add  to your userpage - someone will come very, very quickly to your aid. Please be sure to sign your posts on talk pages using four tildes (&#126;&#126;&#126;&#126;) to produce your name and the current date, along with a link to your user page. This way, others know when you left a message and how to find you. It's easier than having to type out your name, right? ;)

I hope you enjoy contributing to Wikipedia. We can use all the help we can get! Have a great time, all the best, sayonara and good luck! --Cel es tianpower háblame 20:04, 8 February 2006 (UTC)

Please stop adding joke pictures to articles
Please refrain from adding nonsense to Wikipedia, as you did to Animal communication. It is considered vandalism. If you would like to experiment, use the sandbox. Pete.Hurd 04:45, 17 April 2006 (UTC)
 * Why should an encyclopedia be done with such a serious mind? Pictures in articles make them much nicer to read, and everything I have added are on topic. Am I alone with the view that Wikipedia is different from conventional encyclopedias? --para 04:53, 17 April 2006 (UTC)

Dupes tool thumbnails
OK, should work with thumbnails now. Note that the script itself is slow because it has to copy images over the web. --Magnus Manske 16:27, 1 November 2006 (UTC)

coord & display preferences.
Please read the documentation for coord, as I asked you to do before your last revert. The template displays coordinates differently (DMS or decimal) depending in the user's preferences, as expressed via changes to their monobook.css. Also, with regard to usage, please note that the editor you cite says that they stopped counting at 5,000 - that's not the full figure. Andy Mabbett 10:06, 21 May 2007 (UTC)


 * Thanks for this note, your edits did not make it obvious you were talking about CSS. Out of the 4,448,000 registered users we have (from Special:Statistics), only 2,810 monobook.css files exist in the wiki (according to the month old toolserver database). --Para 10:39, 21 May 2007 (UTC)


 * I thought "please read its documentation", where you will find a "display" subheading, was sufficient. In any case, you've just reverted for a fourth time, You might like to revert yourself, before someone WP:3RR you. Andy Mabbett 10:47, 21 May 2007 (UTC)


 * If anything should be reverted, it's to revision 119222782. I am surprised no one noticed your attempt to change site wide policies earlier. A big change like that needs months of preparation from when the technical side is acceptable, you can't just do it overnight. --Para 11:03, 21 May 2007 (UTC)


 * I haven't attempted to change any "site wide policies"; nor can I see what that has to do with 3RR. Andy Mabbett 11:42, 21 May 2007 (UTC)


 * Combining back information that you deleted and reverting based on bare facts is just editing, not edit warring. It might have been better on the talk page, but your single handed attempt to change site wide policies had to go as soon as possible. What was it based on? How many people use a microformat supporting browser compared to the number of people using Google Earth for example? Unfortunately I can't give numbers, but the results of a web search for the Wikipedia geo microformat show much less interest than the results for Wikipedia in Google Earth. You simply can not deprecate a template over another that you happen to have a personal preference for. --Para 12:22, 21 May 2007 (UTC)


 * I'm not going to reply to all the red herrings and straw men in your comments; but I haven't attempted to change any site wide policies. You, though, have breached 3RR and I have given you a friendly warning of that. Andy Mabbett 12:29, 21 May 2007 (UTC)


 * "Since the rule is intended to prevent edit warring, reverts which are clearly not such will not breach the rule." The style guide now lists the available templates objectively based on verifiable facts. Community consensus on the preference issue remains to be seen, but looking at the discussion so far it seems that people would prefer the old templates. --Para 14:33, 21 May 2007 (UTC)

The_Anomebot
The_Anomebot is now adding coord - see The_Anomebot contribs (since 20 May - it was coor title d before that). The Mab-bot is adding coord to info boxes eg here. This all seems a little premature. -- roundhouse 00:58, 2 June 2007 (UTC)


 * I agree, and it's sad to see information being hidden like that, but also too much work to start fighting for single articles. When the community isn't well-informed or interested enough to comment on the problem, I don't know how to make these people stop. --Para 08:38, 2 June 2007 (UTC)


 * No information is being "hidden". New information is being added to Wikipedia. If the community isn't well informed, perhaps it's because of disinformation like that. Andy Mabbett 08:46, 2 June 2007 (UTC)


 * Anomebot's additions in this unsupported format are unfortunate, but still alright, since the data has never been on the English Wikipedia before. What you're doing though, as has been said repeatedly, by converting uses of supported templates into a non-supported one and making that information inaccessible for those who rely on the standard templates being used, is just that, hiding information. If you're not sure that the coordinates in your preferred format can be used by others, converting data to it despite numerous complaints is disruptive, and you should stop doing it until the community agrees with you. --Para 18:32, 2 June 2007 (UTC)


 * The coord template is not unsupported. No information is being "hidden". No information is being made "inaccessible". Please stop spreading disinformation. Andy Mabbett 18:48, 2 June 2007 (UTC)


 * If Google doesn't support the coord template, and Wikipedia information Google Earth users used to see disappears due to your actions, what would you say you have done to it then? --Para 19:00, 2 June 2007 (UTC)


 * Please cite evidence of "information disappear[ing] due to [my] actions". Andy Mabbett 22:26, 6 June 2007 (UTC)


 * You would have to look for the the latest addition of the coor templates visible in Google Earth to know the date of their latest import, and then list all articles and templates converted before that date. There is no tool for such a search. But it is evident that as coord has never been supported on Google Earth, any coordinates changed to coord before the latest import date are no longer visible for any of the people who would potentially read the Wikipedia data there. --Para 22:46, 6 June 2007 (UTC)


 * So, no evidence, then. Andy Mabbett 23:28, 6 June 2007 (UTC)


 * "The Mab-bot" - That's unacceptable. Kindly mind you tone. Andy Mabbett 08:46, 2 June 2007 (UTC)

Sillyness
I don't know whether you were making a point, but posting a 140+Kb list of irrelevant templates, most of which do not and never will use coordinates, to Template talk:Coord/conversion was unhelpful; not least because it made the page almost impossible to read, let alone edit. If you really must re-post the list, please do to a separate page. Andy Mabbett 14:17, 8 June 2007 (UTC)


 * You're right, that set wasn't the best to work on. Maybe Category:Geography infobox templates is more up to date, I'll see what I can get out of it. --Para 14:29, 8 June 2007 (UTC)


 * The list you have just added is indeed much smaller; but still irrelevant, as many of those you list do not use coordinates. Please remove it again. Andy Mabbett 15:58, 8 June 2007 (UTC)


 * Sorry, that's the best I can do. Currently there is no other way to know whether a template contains or suggests insertion of coordinates other than having someone take a look at it. I'm not going to do this project by myself, and I've done a fair share already. Go ahead and revert if you like, but I doubt you'll get better data anywhere. --Para 16:03, 8 June 2007 (UTC)


 * Do you have any evidence that any of the templates in your first table use coordinates at all; or that any in the other two tables use a 'coor *' template? Andy Mabbett 16:28, 8 June 2007 (UTC)


 * None, other than them being in a category called "Geography infobox templates". Perhaps we should also categorize all the templates touched, so that the next time around ( ?) things will be easier. --Para 16:32, 8 June 2007 (UTC)


 * All templates emitting a geo or hCard microformat have been categorised accordingly. Andy Mabbett 16:42, 8 June 2007 (UTC)


 * I've moved then to Template talk:Coord/templates Andy Mabbett 16:47, 8 June 2007 (UTC)

Coord warning
coord is now being parsed by GeoNames, and Google Earth and Wikipedia-World will both be parsing it within a week. Kindly remove your unnecessary and factually incorrect warning from its documentation page, ASAP. Andy Mabbett 12:44, 9 June 2007 (UTC)


 * Just calm down and wait until everyone announces the support. There's no need to be hasty, like you have been for the last few months. All you have is promises that something will be done, and even that is unverifiable. That's very different from something having been done already. It's common in the software industry to miss deadlines, and in this case with no money involved, there is no incentive either. Please try to concentrate on something else for a while, you'll get what you're after eventually. --Para 20:58, 9 June 2007 (UTC)


 * I'm quite calm. Removing the outdated, misleading and unnecessary warning now would not be "hasty", such action is already well overdue. I do not have "promises", I have statements of fact. And I'll concentrate on what I please, not what you decide. Andy Mabbett 21:05, 9 June 2007 (UTC)


 * Do you (Para) have any views on the implementation of coordinates in Dartmoor reservoirs? (The advantage to naming the locations, done here in each row, is that using Firefox with the Operator add-on gives 8 named locations in the toolbar rather than 8 pairs of coordinates.) -- roundhouse0 15:16, 25 June 2007 (UTC)


 * In fact I see you have coincidentally been working on the date of birth template (your sandbox version works for me). Is the coord template similarly name-able? If so it would solve a variety of difficulties. (If it would take the pagename by default that would be best.) -- roundhouse0 16:43, 25 June 2007 (UTC)


 * I brought up the template used in that article at Wikipedia talk:WikiProject Geographical coordinates, but the discussion stalled for the usual reasons. In principle I wouldn't support any template that adds yet another coordinate entry format without bringing any additional value for the coordinates themselves, so in this particular case it would be best if coordinates in that template could be given with coord directly. I also think that wrapping some indeterminate bits of information together with a single purpose template is not scalable, and as such shouldn't be introduced here. I can see the advantage though, re microformats, and I guess we have to let these things sprout like weeds while waiting for a better alternative, but still we shouldn't allow people to create just anything thinking of microformats only and nothing else. Semantic MediaWiki for example would be progress towards the right direction, as it allows people to tag information in articles with meaningful predefined names used throughout the wiki, and then display any subset of those tags for whatever purpose the reader might have for them.
 * Adding the name to coord is a good idea, and I would expect it to be easily done when you know what in the template calls what and where the microformat html is best generated. I don't feel like looking into it myself though since the template is very complicated, but I'm sure User:Quarl wouldn't have any trouble adding the name parameter. --Para 22:15, 25 June 2007 (UTC)


 * I find I had read the de-migration stuff a few days ago and had forgotten about it. -- roundhouse0 12:24, 26 June 2007 (UTC)


 * I have asked User:Quarl, with predictable results so far - perhaps you can follow AM's latest answers. -- roundhouse0 19:42, 26 June 2007 (UTC)


 * As cooperative as ever, I see. Quarl seems to have been inactive lately, that's too bad, someone else will have to do the edits. I took a quick look at the template myself then, and it looks like the html comes mostly from Template:Coord/link, so that one needs to edited along with the parts that call it to pass the name parameter there. I'll try and see if I can figure it out, but frankly, the template is a mess! Sandboxing any tests may be more difficult than just planning them well and editing the original templates directly. --Para 20:32, 26 June 2007 (UTC)

OTRS action on Gator Engineering
Hi Paradisal,

(Quick intro to avoid misunderstanding: I'm not disputing your actions on the article below, and I don't need information on how to dispute an OTRS edit. I'm just trying to get information on how things work.)

I added a db-copyvio tag to Gator Engineering, as it was a direct copy of the UFL Engineering Dept. "About" page. The creator of the article evidently emailed the UFL and got their permission to publish it under a GFDL license. You then added a ConfirmationOTRS tag to the talk page, and removed the db-copyvio tag. I'm quite surprised UFL gave their permission, but if they did, I have no problems with this outcome.

So, my question, which I promise I'm not trying to be snotty about, is: "How do I know you have the authority to do this?" I looked on the OTRS/personnel page, and User:Paradisal isn't listed as an OTRS member (probably because that's on Meta?). So, assuming I'm not a trusting person, how would I verify that all this is above board? I'll admit I was initally sceptical enough to look at your edit history and see who you were, and I am 100% confident you're not part of some elaborate criminal conspiracy of counterfeit OTRS operatives, bent on world domination. But technically, if i wanted to, how would I verify this was kosher?

Pointing me towards the approriate WP or Meta page explaining this is fine.

Thanks for your time. --barneca (talk) 20:52, 15 June 2007 (UTC)


 * Someone was bound to bring this up at some point, I'm surprised it took this long. :) I linked my main accounts together now. Have to see whether I'll usurp the name here to avoid misunderstandings, or just wait for SUL. Thanks for ensuring OTRS reliability. --Para 21:54, 15 June 2007 (UTC)


 * Took me a second, but I see what you did. OK, a smarter person than me would have linked Para and Paradisal (especially now that I notice your sig), but I was, after all, interested in the theory, rather than unmasking a fraud.  It just concerned me when I realized that there was nothing technically stopping me from using the ConfirmationOTRS template, and it got me thinking.  Anyway, thanks. --barneca (talk) 22:01, 15 June 2007 (UTC)


 * To give some pointers for the future, the long and extensive meta:OTRS/personnel page is generally updated by people themselves, while the Wikimedia Volunteer Coordinator only keeps the list on meta:OTRS up to date. I would assume that unauthorized additions to meta:OTRS wouldn't go unnoticed, while the other page may well contain incorrect information. --Para 09:45, 16 June 2007 (UTC)

GeoHack
Hi, GeoHack is just that - a hack based on someone else's MediaWikiextension to run on the toolserver. I have wanted for it to become an extension again, so it would just be another special page in the respective wiki(pedia). But, alas, none's listening to me...

Anyway, if you want a quick fix: The generated GeoHack page is just a transcluded Wikipedia page, and, as such, still contains the tags. So, if you want Andorra to bring up the stuff under heading "Europe" first, link to http://tools.wikimedia.de/~magnus/geo/geohack.php?language=en&params=42_33_30_N_1_33_19_E_type:country#Europe and it will appear without scrolling.

Changes to the software will most likely affect all wikipedias and other Wikimedia projects alike, so this should better be planned and discussed on the mailing list or a relevant talk page. --Magnus Manske 13:28, 6 July 2007 (UTC)


 * Well there are two basic issues:
 * Technically, it would require checking for point-in-a-polygon from a database, which IMHO is not feasible at the moment given the high usage of the page and the fact that the toolserver is a shared location. Additionally, it would probably be a bitch to code, certainly a lot of work, which I am frankly not willing to do just to save some people two seconds of scrolling or clicking on the TOC.
 * It may well be that people complain about "having to scroll" (ridiculous as it sounds, users are strange;-) but what if we'd put the local section first? Woudln't just as many people complain of not seeing googlemaps first anymore? The reason people have to scroll is that the page is too long to fit on a single screen. So the fans of what ever is not in the "first view" will probably complain no matter what.
 * What might be feasible is that some sontents is "folded away"; if your location is in China, you don't want to see anything under the Europe and North America headings anyway, so they could be hidden (unfolded by some JavaScript, maybe). This would require either a parameter that specifies which headings to show/hide, or a lookup table region<=>heading. If any of these exist, let me know.
 * From my POV, all that needs to be done is to move the TOC up to the top of the page, which can be done by editing the Wikipedia template. That way, most of the local stuff will be in plain sight and reachable throuh a click. If anyone is bothered by that, send them to me, and I'll explain to them that one can only put so much on a single screen :-) --Magnus Manske 15:34, 6 July 2007 (UTC)

Coords
Is it possible, with the coord template in the 'traditional' position used by the Anomebot2, to pick up the lat and long variables from it and use them in the infobox? (eg to place that red dot on maps, to fill in a coords row if reqd, etc).

Further, is it possible, with the coord template in the 'traditional' position used by the Anomebot2, to give it automatically the name of the pagename? (cf "|name=") -- roundhouse0 14:17, 13 July 2007 (UTC)


 * Short answer: No and no. Sorry.
 * Long answer: Only templates called by the parent template can use parameters given to it. It would be cool to be able to have an abstract template call using the syntax, where the location of the descendant would depend not on the parent template, but where the parent template is used. As far as I know that's not supported, so unfortunately there's no way to combine the data like that. Visually it might be possible to have coordinates from the end of the wikitext displayed inside an infobox using CSS the way title coordinates are, but that still wouldn't allow actual processing of the data.
 * Having coord use the pagename automatically when none is given could easily be done in the template, but because of the light-headed idea of having templates generate the microformat instead of MediaWiki, doing that would break some of the other microformat implementations here. The reason for that is that some infoboxes and probably other templates too create a single complete microformat entity, and if one of their content elements wants to create one too, the page ends up with nested microformat html, which probably isn't allowed in the specifications. Templates also don't know anything about their containers, so without yet another additional parameter they can't behave differently depending on them being inside an infobox or not. --Para 10:41, 14 July 2007 (UTC)


 * "the light-headed idea of having templates generate the microformat instead of MediaWiki" - please fell free to implement microformat generation in MediaWiki. Andy Mabbett 11:01, 14 July 2007 (UTC)


 * I am not interested in implementing microformats, but I will try my best to limit their introduction to Wikipedia in cases where they're detrimental to ease of editing. If you care to continue your microformat crusade in a more thoughtful manner, you need to first have a general method to tag information in a way that can be used not only for microformats but for other purposes as well. When data is organised in such a way and output of a subset works in some format, it should be trivial to generate a microformat output as well. I think Semantic MediaWiki is doing it correctly and even a partial installation that only works on a single page at a time would be progress towards a more correct framework for semantic markup on Wikipedia. --Para 13:41, 14 July 2007 (UTC)


 * OK and OK - a pity. There is Persondata on a subpage in which you might be interested. Carcharoth tends to make good sense. -- roundhouse0 14:36, 15 July 2007 (UTC)


 * I wouldn't put much effort into the persondata project. When Semantic Mediawiki or another similar improvement makes it possible to tag information in a way that explains what it is and what it's related to, there will no longer be any need for a separate set of single purpose information hidden away somewhere. While waiting for that, I'd let the persondata people run their project as they feel is best, as long as they keep the data "hidden" and don't otherwise complicate article editing. --Para 20:41, 22 July 2007 (UTC)

Redirects
There is discussion about creating and categorising redirects (eg for ZIP codes). It occurred to me that a location redirect could be given coords (display=title) and the various parsers could be alerted to this. Eg there is List of United Kingdom locations: Ma-Maq with many redlinks - I am not claiming to have thought this out completely. There is also an existing partial categorisation of UK post codes (see eg MK postcode area) and one might imagine redirects (or indeed articles) for the sub-codes such as MK1, MK2, ..., suitably coordinated and sub-categorised. I have created such a redirect - MK89DP UK (this is part of one side of the road Holyrood in Great Holm) - a first step would be to agree on how this is best named and given coordinates. -- roundhouse0 15:00, 15 July 2007 (UTC)


 * That's a nice idea in theory and I can see how it would be useful, but I think it wouldn't be well accepted for a couple of reasons. Though the purpose would be reverse, such information might be compared to telephone numbers for example, and in turn to What Wikipedia is not. The existing postcode categorisation looks fine to me, if not even too detailed. Such static lists might be better in Wikipedia namespace as a to-do list of locations that still need an article, or as a storage of information to eventually go to those new articles. Anyhow, when a location is notable enough to have a paragraph or an article of its own, people will find the name of the location - the Wikipedia entry point - by other means than a postcode. If they can't, there's another problem that would be nice to have a solution for, but in my opinion best outside Wikipedia. --Para 20:41, 22 July 2007 (UTC)

Invite
Given your very frequent work on inline templates, I think you should definitely join WP:WPILT. —  SMcCandlish  &#91;talk&#93; &#91;cont&#93; ‹(-¿-)› 17:23, 27 July 2007 (UTC)

Can an image be moved?
Hello Paradisal! I noticed you were an admin at Commons, and that's why I decided to ask you about the following problem. It's regarding Image:Map of New Mexico highlighting DeBaca County.svg, I want it to be renamed to Image:Map of New Mexico highlighting De Baca County.svg because the county is named De Baca and not DeBaca. Is it possible to move images? If not what should be done? Thanks. -- Crzycheetah 02:24, 16 August 2007 (UTC)


 * Unfortunately MediaWiki doesn't support moving images; only pages can be moved. The only workaround is to re-upload the file using the same description, and then marking the incorrectly named file with  . Not surprisingly, this is a Commons FAQ and the developers are working on proper support for this much needed housekeeping feature. --Para 06:41, 16 August 2007 (UTC)
 * Thanks for the info.-- Crzycheetah 07:35, 16 August 2007 (UTC)

Coord MF sensitivity
You mention: "Closes the microformat wrapper in a way that it can only contain the coordinates and their name, and not the additional properties "address", "label" or "note". This breaks some articles where microformat markup has been written directly in the article wikitext."
 * But the "inline" option shows that coord can tinker with its microformat (well, I'm a template programmer so I know that). It would be possible for coord to be told it is being used within a MF and to not close the MF.  If that is correct, you could mention that if the WP MF create an hCard template which could include coord, coord could be modified to emit a compatible format.  So a decision now does not block future hCard use.  (SEWilco 21:11, 16 August 2007 (UTC))


 * That's true yes. I mentioned the possibility of such a parameter a couple of times in the discussion, but nobody showed interest so I didn't include it in the summary. It's a dirty solution as it would allow the name to be given but wouldn't solve any of the original problems; the template and the additional elements would still have to be surrounded with the presentation level markup to close the wrapper, and editors would have to ponder on the magic parameter which probably changes nothing for most people. But I suppose for completeness sake this possibility should be included too. --Para 21:56, 16 August 2007 (UTC)


 * Actually, editors would probably not ponder the magical parameter on coord much, because if it is being invoked as part of a MF then there probably is a different editor interface (such as a different template). We won't know until such an MF interface appears.  (SEWilco 21:22, 18 August 2007 (UTC))


 * But then, again, having a separate template for coordinate entry just to generate microformats breaks machine readability. Microformats are fine as long as common editors who don't know about them won't notice their existence, but anything past that is development to the wrong direction. If it's not possible to have these additional microformat features without stifling improvements elsewhere, then they'll have to wait until MediaWiki can generate them transparently. --Para 23:41, 18 August 2007 (UTC)


 * Whether MF affect machine readability also depends upon the machine reader. But I wasn't thinking of a situation which would create duplication, but rather something such as a "person" template which might include coordinates (for some reason); the person template would emit the info in some Wikipedia-accepted style while including MF markup.  Whether such a template invokes coord is dependent upon the implementation, but one option is for the "person" template to have its own coordinate parameters so the template can invoke coord with whatever incantation is needed to allow the needed MF to function.  So the editor may not know of coord details, they might just be filling in coordinate fields in a template or infobox.  (SEWilco 04:41, 20 August 2007 (UTC))


 * Of course, but the most notable machine reading on coordinates is done by external services like Google, who do not rely on microformats to get the information. It's the wikitext we need to keep consistent. If coordinates are entered as components to some template which may later pass them to one of the basic coordinate templates, everything that expects to find a standard coordinate template in the wikitext breaks for that article. The solution would be to generalise and make all those templates supported as well, not by naming all of them, but by using standard parameters as proposed on Wikipedia talk:WikiProject Geographical coordinates and saying that all templates that use those parameters are coordinate templates. --Para 09:21, 20 August 2007 (UTC)


 * It is a good idea to standardize the template terminology (in this case the parameter names). But Google's pattern matching on coord is due to their crawler apparently reading the wikitext so it doesn't see MF markup; Google's processing of Wikipedia's messy markup is more an indication of Google's skills rather than being relevant to standard formats.  Actually, Google's pattern processing fits right in with the original design of their original crawler's pattern-learning methods.  It doesn't seem to be working so well over at Commons, where apparently the Location template is not recognized by Google and neither is its geo MF.  If Google were looking at the geo MF it might have already picked up the thousands of tagged items automatically.  (SEWilco 21:51, 20 August 2007 (UTC))

I haven't finished reading the details, but another issue is whether the proposed change would instantly break existing articles. I think it would not. Only when someone adds the name= parameter would surrounding MF be broken, so the coord documentation for name should mention what hCard notations look like so editors could avoid breaking things. (SEWilco 21:11, 16 August 2007 (UTC))


 * That's right too. This way of implementing microformats in templates creates some unwanted dependencies which editors really shouldn't need to be thinking of when all they want to do is name the data element. But that'll be to worry about later when the existing microformat markup in articles is replaced. This annuls one of the opposing arguments, thanks for noticing! --Para 21:56, 16 August 2007 (UTC)

Geo serving
As you're interested in DB geo awareness, I point out PostGIS and GeoServer. The tools, not the articles. (SEWilco 21:37, 16 August 2007 (UTC))


 * They look useful, but they may be a bit heavy to be run on the toolserver as PostgreSQL isn't installed. I'm happy with MySQL and its basic geospatial features so far though, GeoCommons works without too much delay and even the proposed region lookup for GeoHack looks good. Perhaps at some point people will come up with more resource hungry ways to utilize our data and I'll have to take another look to the available platforms. Thanks for the barnstar! --Para 22:08, 16 August 2007 (UTC)

Coord questions
It looks like there is support for name= but there are some questions about it. I'd also like to know the proper usage in a list because I have 10,000 coordinates to emit and they may as well be in the properly named format so they'll look right when the change is made. Also, the requested change should be specified clearly for an editor, such as by doing cut and paste. The RfC is a bit messy; maybe a new section should be made with the requested change and mention the above RfC for support. (SEWilco 16:44, 23 August 2007 (UTC))


 * Thanks for the note, I've been away for a while and the watchlist isn't ideal if it's checked less than daily. I'll try to answer the questions and clarify the rest as soon as possible. --Para 17:01, 23 August 2007 (UTC)

Coord slight error
I think: Table row

|- class="vcard" |class="fn org"|Approx. tunnel mid-point |52.50435°N, -2.05932°W

into

Approx. tunnel mid-point 52.50435°N, -2.05932°W

forgot the table context, so an exact translation would be:

Table row

|- class="vcard" |class="fn org"|Approx. tunnel mid-point |52.50435°N, -2.05932°W

into

|-   | Approx. tunnel mid-point 52.50435°N, -2.05932°W

That is, the "vcard" example only functions within a table, so to produce similar wikitext the replacement would also be within table wiki markup. If I understand the desired change, coord with name= would also work outside a table (technically, coord would be directly emitting the vcard/fn_org markup rather than forcing wikisyntax to emit it within wikitable markup). So your example is faulty only because you took the new version out of the wikitable context into which the vcard was forced. You're making the wikitable context optional rather than mandatory. (SEWilco 19:06, 23 August 2007 (UTC))


 * True, the row and column separation markup is of course still needed if an editor wants to construct a table. The table row/cell and the coordinates with their generated microformat are two independent elements that can be used wherever, except in the case where someone has created an undocumented dependency in a table by writing the microformat markup there directly. I must have thought that the lines are not relevant for comparing the generated output, but for completeness sake I guess they should be there. You can go take the credit for that one. :) --Para 21:18, 23 August 2007 (UTC)

Stepping through coord changes
Please proofread my proposed changes to coord. Then ask an admin to wade in. (SEWilco 19:15, 26 August 2007 (UTC))


 * Thanks, I corrected a few slips and did a last minute change to move the pagename to Template:Coor URL after all. That way the title parameter for GeoTemplate is never empty and all services that support at least a single name will always get something regardless of the coordinate template used. --Para 20:27, 26 August 2007 (UTC)


 * The admins took control of the templates, now an admin gets to sweat the templates. I'll be off doing magic in a valley while they're busy.  (SEWilco 04:28, 28 August 2007 (UTC))

Coord named (neatly)
Full marks for perseverance and remarkable patience in this matter. Now that we have a name parameter, do you favour using it on the typical article with a single location as tagged by the anomebot? (It would then appear as a name in Firefox rather than a pair of coords.) -- roundhouse0 14:23, 6 September 2007 (UTC)


 * Thanks. And I expected understanding the template and finding the correct modifications to be the most strenuous and time consuming part. :) So now that we do have it, you can use it wherever you're not seeing a name in Firefox yet. With some infoboxes for example a name is visible even though it hasn't been given to the coordinate template, so then naturally it won't be necessary. In my opinion adding a name to the only coordinates in an article is a bit redundant since the article name is visible in many places already, but  name=  can be used if you like. Too bad it wasn't possible to make that the default. Also, if the article is about a broad location, and pinpoint coordinates mark some small scale object around the middle, it might be a good idea to name that instead of using the article name. In any case, if one day MediaWiki gets a better microformat implementation, the article names can all be replaced by a bot. --Para 14:05, 8 September 2007 (UTC)


 * I was wondering if Google Earth could look for coord with name=XXX, also whether the Anomebot could be asked to add  name= . Also the documentation of coord doesn't say much about name=. (I am supposing there has been no great outcry by readers offended by metatext showing up when least expected.) -- roundhouse0 16:07, 8 September 2007 (UTC)


 * They could, but I don't know if it's in their interests. Google filters Panoramio images for example, so it would make sense if they only ever used notable coordinates, one per each topic that we Wikipedia people consider worthy of having a separate article. Otherwise it would get busy very fast, especially with people marking linear features with the same template like they're doing now. About the bot, any bot could do that really, since it doesn't require parsing the coordinates at all. The bot can just add the name parameter to any coord template with display=title that doesn't have a name yet. I don't know what else to say about the parameter in the documentation, but if you can sell it better than it is now, feel free! All the supposed problems were theoretic, so it shouldn't be too hard to do that once you find good use for the names somewhere. --Para 22:24, 8 September 2007 (UTC)

Picture request browsing
Actually I was thinking of browsing more like this  I don't know how the URL links behave in the GE embedded browser. Yes, Super-Regions would be a nicer user interface but for the test your kml URLs are sufficient for the concept test. (SEWilco 14:08, 18 September 2007 (UTC))
 * Well, that seems to have worked better than I thought. Clicking on the balloon URL does load the KML.  I see Bolivia's icon ended up in Brazil due to the coordinate in Madidi.  And I see there are some requests near me that I hadn't noticed.  (SEWilco 16:30, 18 September 2007 (UTC))
 * Fixed Madidi. Coordinates were way off, starting with N instead of S.   But that KML is just a test so I'm not rerunning my messy script.  Other than that, simply using the averages of the coordinates seems to have placed the icons well enough.  (SEWilco 16:49, 18 September 2007 (UTC))
 * I could have created a KML with all the points. But generally one would browse specific locations, probably where you or acquaintances are.  That's also why I'm seeding the National Historic Landmark list pages, so it's easier to find an NHL near you.  (SEWilco 16:52, 18 September 2007 (UTC))
 * I wonder if there is an invisible version of the geo microformat which could be put on the Category pages to mark a browsing icon location. Then a bot could ensure each of the relevant Cat:Wikipedia_wanted_* pages has such a location, whether detected by a geo-sensitive web browser or a script.  (SEWilco 16:49, 18 September 2007 (UTC))

That looks good, it's somewhat easier browsing places you might visit on a map than trying to find all the possible categories on-wiki. That could be just us though, I don't know how many Wikipedia users have GE or like using it. I like the averaging idea and the location can be precomputed in the database on every update, but I don't like tagging categories with a location, except maybe if we can mark areas and not pushpin points, but even then there probably exists an article where the area would be better defined. Anyway, I could easily(ish) implement a viewpoint based kml query tool using the same method I did for the Commons image layer, but I don't know if I'd be stepping on someone's toes there. This kind of functionality would be great for Wikipedia-World, making it more popular and centralising the internal Wikipedia kml data to a single place. Categories also don't need very high resolution imagery, so WikiMiniAtlas would be an ideal platform as well. Whether it's me or anyone else who writes the code, it would be nice to make it as general use as possible. We can look at: Filtered by Plotted in ...for starters. It'll be a bit of work. --Para 19:24, 18 September 2007 (UTC)
 * Articles
 * Categories
 * A single category
 * A parent category recursively
 * Template used
 * Each article location
 * Location averaged from all the articles in the closest parent category
 * Location averaged from all the articles in the nth level parent category


 * It will be a lot of work if you try to do it all at once. There already are single-article tools which are usable, so we can try to deal with other problems and see if general-use tools appear.


 * I averaged the coords because it was a quick test and I didn't want to code coorD DB lookup and disambiguation. That trick may be useful for bot use but the purpose is to get the SUMMARY icon near the data and the icon does not have to be always at a data centroid; indeed, that could cause icon jiggling with popular topics.  If icon locations are stored at bottom of Categories they could be adjusted by humans or bots (a bot override template could keep bots from repeating mistakes).


 * There are no parent Cats, technically, which makes some tasks hard. I simply grabbed all Cats with a certain prefix, except the null-named one and Tokyo prefectures which had Unicode (which my primitive script did not like).  To view all picture requests, Cats for WP:RP should be included; we could campaign for replacement of RP and subpages with Cats but RP does have some advantages.  One solution with the present tool would be a Cat whose members would be assembled in this WorldOfWantedPictures file (on demand or on schedule is a minor issue until requests are being filled much more quickly).  The Cat:WantedPicMap could be added to articles such as WP:RP and Reqphotoin/Reqphoto (the templates would attach one template for the global cat and the existing cat for the region); I think the DB has the sorting name and your tool could pick up the "in" name from that).  The (KML?) data could be displayed by various tools once assembled.  (SEWilco 03:03, 19 September 2007 (UTC))


 * The above would be useful for Wikipedia editor find-a-photo-subject use. Could it be useful in article space?  There are some coordinates spread across several articles, and browsing summaries of those locations could be similarly useful.  For example, List of U.S. National Historic Landmarks by state has some National Historic Landmarks (NHL), some states have separate NHL lists, and I think some counties have their own NHL lists (some states are full of history).  If your tool picks up all coords in an article then it will work for list articles; the name of the article could be used for a Cat name or your tool could detect the article name as a Sort key to a single Cat.  Maybe GeoGroupTemplate or a similar template could provide access to the summary map.  Let's see.. marking articles, collecting data, and providing access from an Article.  OK, could work.  Next I examine one detail more closely.  (SEWilco 03:21, 19 September 2007 (UTC))


 * The present tool grabs all articles in a Cat. My summary used each Cat as a region tag and emitted a summary icon for each region.  (Actually some Cats are by topic rather than region)  A summary tool could be told or detect the relevant Cats (such as by manually adding a Cat to each member Cat).  But I suspect the Cat map tool is looking at the table categorylinks, which has both cl_to and cl_sortkey.  So all members of related Cats could also point at a Cat used by the tool, and by using the SortKey the related articles could group themselves together.  The template marking all Category:Wikipedia_requested_photographs_in_Avon could also add Category:Wikipedia_requested_photo_map|Avon and the tool could detect the Avon grouping (if I understand the table correctly).  The tool might emit only a summary file or might emit several files (can you store KML files on the toolserver for Network Link retrieval?).  The present tool handles only members of a single Cat, but if it can detect SortKey tagging it should be easy to have it create summaries.
 * A related issue is that WP:RP is not a member of the Cat:Wikipedia_requested_photographs_* group but would be nice to include it in the pictures map. The page and its subpages could be manually marked with the appropriate Cat.  Contents would be worldwide rather than regional, but at least those would also be available on the map.  Altitude of summary icons can be increased based on max difference of member coordinates so national/worldwide icons will be easier to find.  Or if you want to work hard at programming the tool, an article such as WP:RP might NOT provide a SortKey and the tool would try to group its coordinates within regions defined by other SortKey-named groups.  Either way (global or regional), articles such as WP:RP could be used by the SortKey-modified tool.  For WP:RP to work the tool would have to pick up all coords in the page and not only one.
 * Article space: Summarized grouping based on SortKey or article name would reduce displayed points. Articles being used the same was as WP:RP should work with manually defined Cats; should also work for topical template-created Cats.  Preceding mention of "work" only refers to the tool being able to detect members in a Cat.  Actually using the data for Editors merely means some backstage access to the data.  To use the data in Article space it would have to be made visible someplace.  GeoGroupTemplate or a similar template could emit links to URLs for accessing the map data.  Or if a topical template ("Minigolf courses of Orlando navbox"?) is emitting the grouping Cat, it could also emit links to the map URLs.  OK, I think it could be used in Article space also.
 * Summarize everything: What if every article is stuffed in the same MapGroupCat? The DB table should be able to handle it.  There would probably be many summary icons, depending upon how many were chosen, but there would probably be over 10,000 summary icons or 10,000 icons within one region, so many maps would be overly crowded by the result.  Depending upon the coding of your tool it might be able to process the data (processing one Cat at a time and emitting its data and summary placemark before moving on to next one).  Processing time will still be significant.  So there will be an upper limit, whether it's map real estate or processing resources.  (SEWilco 05:14, 19 September 2007 (UTC))

Wow, you're full of ideas. I don't follow this sortkey idea though. Sortkeys are for sorting category lists, and any other use is bound to break. What we need here is information about category inclusion. A category is a parent category when something links to it. If we want to put together articles linked from a page with articles in a certain category, we should put them both in the same parent category: all photo request categories should be made a child category of the area category they're requests for. I don't know if there are any categorisation rules for mixing Wikipedia categories with content categories though.

Though if it's enough just plotting articles, and categories can be ignored, a query interface to find articles linking to a template will be enough, and it's much easier to implement than traversing categories. In addition to templatelinks, articles linked from a certain page can be included in the results. The high volume and density of requests in some areas can be helped by scoring them based on the date of last edit, date of last 10th edit, number of total edits, etc, and then depending on the viewpoint altitude, showing only the top articles, perhaps also with a filter to get sparse results.

Still about plotting categories: If it can be computed automatically, it should, and not start entering artificial data to Wikipedia. Since live access to current article coordinates isn't available, and most coordinates are static anyway, the location will change only when new articles are added to the category, and rarely when some are corrected. To make it less prone to "jiggling", we can use median coordinates instead of average, so that outliers won't have so much weight. Or since the intelligibility of point coordinates gets worse the bigger the area is, we could do better and instead of a single icon, use the convex hull of the article locations as a multi-point polygon.

For use of more than one set of coordinates per article we should really work together with User:Stefan Kühn from the Wikipedia-World project to make that data available. --Para 15:22, 21 September 2007 (UTC)


 * Mixing Wikipedia categories with content categories: There are many categories used for things other than article topics, but you're right that links to things other than categories could be used.
 * Articles linking to template: Can simple template parameters be parsed so labels/groups can be extracted?
 * Child categories: I tried to avoid child categories as a requirement because there is no automatic way to define the relationships. Using a new parameter location in Reqphoto wouldn't be detected until a suitable cat link gets manually created.
 * Sortkey: I was using that as a way to get the grouping key. If Sortkey is created by a template then the special grouping meaning only needs to be understood by the template editors; if an article gets linked to a category without a grouping key then it would probably be displayed as in a group of its own (unless the tool autoassigned the data to groups).
 * Spreading by article characteristics: Good point. Data can be spread through the 3rd dimension based on a number of attributes; it could also be spread through time if there is need to limit the time window visibility.  I'm not concerned about which attributes as we probably need to see some data before adjusting the user interface.
 * (SEWilco 22:43, 21 September 2007 (UTC))


 * Ok, I forgot that categories may have to be linked manually after a request template has been added. I doubt the whole tree is ever up to date for all requests around the world then, so the category structure can't be trusted.
 * It would be technically possible to parse template parameters, but if there are many pages to parse, it would not be practical as each page would need a possibly expensive fetch request, and the results would have to be written into a separate database that is updated periodically by retrieving and parsing all the pages again. Not a good solution. What kind of parameters did you have in mind though, and for what purpose?
 * Seems to me like all that is needed is a query for where request templates are used and where the request page links to, joined with the coordinates table, and selected either by the viewport bounding box or as nested regions with the scoring if the results are often too heavy. This doesn't need any additional data stored in the database and can be done on application request. Now someone just needs to start coding. :) --Para 17:05, 23 September 2007 (UTC)


 * If you're going to use template links to find articles, then either the map tool would display articles which are linked to the template (requiring one template for each map or each mapped region) or the map tool would have to identify parameters given to the template (either by parsing the template call or identifying a side effect such as a category link, which would probably require telling the map tool what category link pattern is to be followed, so the map tool may as well be directly told to look for the categories which follow a certain pattern; the difference depends upon database speeds of multiple table link access as compared to searching the categories for those which match a certain pattern). Grouping by bounding box may work when the region is known, but grouping according to the existing templated picture request categories for, say, individual New England states allows more focused browsing than grouping by grid boxes.  (SEWilco 03:00, 24 September 2007 (UTC))


 * I'm not quite sure I understand, so for a proof of concept I put together the simple query interface I had in mind: try temp/photorequests.kml. Perhaps it'll be easier to discuss when there's something to see. To me categories and geospatial browsing are two parallel methods for choosing which articles to look at. When a user has already chosen an area of interest without resorting to categories, additional categories won't be much help unless they categorise the articles by some other criteria than their location, but off-hand I can't think of any that would be useful for someone who is thinking of going to take photos of an area. --Para 14:14, 24 September 2007 (UTC)


 * Yes, using KML Super-Regions reduces the transfer overhead. I don't see why some points are being chosen for being visible from further away.  You're right that the need to label by categories on a map is not very important when viewing by location.  On the Wikipedia side of the tools, what is needed are ways to mark information for presentation, and ways in article and WP:pages to make geobrowsing available.  For "requested pictures", there are templates which emit Categories with certain names and there are WP:RP and its subpages.  How can KML-emitting tools be told to find all those for one browsable map?  A template which you pass several parameters which can be a Category or an article?  And does a link to an article include transcluded pages?  (SEWilco 18:39, 25 September 2007 (UTC))


 * Whatever you're doing for summarizing is causing some states to seem totally empty. One has to zoom rather close in for something to become visible.  Simultaneously, a large number of items in surrounding states have to appear to make items appear in sparse states.  The encoding also needs some adjustment (not surprising for a test) to deal with the period in "St. Paul, Minnesota".  (SEWilco 19:28, 25 September 2007 (UTC))

Arbitrary section break in placemark test

 * With this test the placemarks are a limited pseudorandom subset of all the points in the viewport. That's as good as any criteria we've talked about so far. ;) In addition to bringing down the transfer and refresh delays, regions would help get more evenly distributed results as the queries would be independent. I'll have to start coding that at some point, now I just took the GeoCommons codebase (with its stripping of file extension, ie. everything after the last dot, heh).
 * Combining many types of queries in a single request won't be difficult, we just need to think of all the types people are likely to want to use, and parametrise those. It could be something like "templates=a,b,c|articlelinks=x,y". Would it for example be useful to have functionality to plot links to an article as opposed to links from an article? The transclusion difference also needs a parameter. All these need to be reflected in the template parameter "language". I also see from the logs that people have tried to use kmlexport with pages that don't have any coordinates, so they probably expected it to magically use links instead like it does with Category pages, so we'd need to make it clear from the template parameters and (probably the same) php script parameters what kind of results can be expected. --Para 15:03, 26 September 2007 (UTC)


 * Parameters: I think pagenames would be better than the tool requiring that links be separated based on namespace, as it is easier for programming languages to parse Namespace:Name phrasing than it is for templates to manipulate them. We can already see a need to be able to show in WP:RP a message "Click here to see locations of requested pictures" which involves referring to both articles and categories, and one template can probably achieve that incantation more easily without namespace sensitivity.  The namespace sensitivity (and error messages) can be dealt with in the invoked tool.
 * Links to/from articles: I hadn't thought of that yet. There are links for many purposes.  One can draw lines with arrowheads, so I recognize what could be expected to be displayed for a country's article would be links to/from all the surrounding countries ("bordered by..."), links to/from all sister cities, links from cemeteries and ships (person/ship buried at coord and article names a country), and various lists (volcanoes, mountains, oil producing countries) unless the source article must have a title coord.
 * Transclusion: Transclusion processing could be limited to transclusion within the same namespace, which could reduce problems such as template or standard message usages. Should transclusions be automatically processed, or should transcluded pages have to be separately tagged (and with visible or invisible tags)?  It seems the design of transclusions in articles is to make them behave as much as possible as if they are part of the article, so automatic processing of transclusions would be nice.  WP:RP is the only example which comes to mind at the moment, and a single slow-changing item like that could be handled through several means (including simply passing names of subpages to the tool invocation).
 * Kmlexport failures: The linking ability of kmlexport has only been mentioned in two Talk pages, as far as I know, so it is more likely that people are using kmlexport invocations for other purposes. Some possibilities are to invoke it in articles where they intend to add coordinates, that they're invoking kmlexport in articles with unrecognized coordinates (but you probably looked for that already), or that people are adding kmlexport invocations to all articles the same way some people add reflist to articles in case someone later adds relevant material.
 * (SEWilco 16:42, 26 September 2007 (UTC))

Get the popcorn
Oh, this should be good. Someone's trying to argue without reading things. This shouldn't take long. (SEWilco 05:07, 27 September 2007 (UTC))
 * Yup, his 3RR got thrown out on the simple timing issue. He's not reading the 3RR instructions, so he also probably was not aware that my alterations to my reversions probably break the nonresponsive-to-Talk clause.  Now he's trying to force a 3RR for a content dispute which is due to his own refusal to read or accept what is in the Talk page.  The way he's going he'll get banned for what he's doing in another article anyway.  (SEWilco 18:42, 27 September 2007 (UTC))
 * Fellow must be asleep. Another editor added a link to an online copy of the column; I tried a Google search for the column's name and the new link is in the first page of results.  So it was easy to find that way too; I'd used other means to get a copy so hadn't done that particular search.  (SEWilco 02:49, 28 September 2007 (UTC))


 * Right, more eyeballs needed indeed. On this side of the pond I don't remember hearing much about that incident though, and reading the intro of the two articles didn't really refresh my memory too well, so looks like referencing isn't the only problem with those articles. --Para 17:33, 28 September 2007 (UTC)


 * There's no backlink to Dan Rather? Good point, I'll look at that.  (SEWilco 00:36, 29 September 2007 (UTC))
 * OK, there is a summary of the situation already but what it probably loses in translation is that CBS had a decades-old reputation for news coverage, and "60 Minutes" had a long history as a news magazine TV show, with Dan Rather being anchorman of the CBS Evening News and leader of "60 Minutes". His being caught presenting a forged smear on the President just before the election was a tad unsettling.  Particularly because the opposing political party had been informed of the story and was going to follow up right away with a "favored son" series of personal attacks on the President.  Over here the BBC also has a strong reputation for news, so imagine something similar involving a BBC news personality.  (SEWilco 00:48, 29 September 2007 (UTC))

Showing picture requests
Looks like someone agrees (diff) a map of requested pictures can be useful. Any progress in making a map of all the requested pictures? (SEWilco 17:05, 6 October 2007 (UTC))


 * Well, on the Google Earth side I implemented regions and noticed that it makes only sequential http requests with a single host. That makes loading the placemarks painfully slow when there are many regions to load at once, such as is the case when the area is divided into a grid and nested vertical levels. So a single database query with the viewport based refreshing is quickest for now, and I managed to pinch a few milliseconds off of that too. Criteria for which placemarks to show first is now the page length.
 * With kmlexport I implemented a linksfrom=1 parameter, which places links from an article into the section folders the coordinates are already in. See for example how it works with Requested pictures/Architecture. Support for the parameter should probably be added to the template using a similar parameter to invoke the functionality per page, as the default for most pages is probably to map coordinates only.
 * I would've started looking at recursion into transcluded pages as well, but couldn't actually find it used with any location related pages. Perhaps instead of recursion into the very specific "main" links or something like that, I should make the prefix idea of yours happen. Will see. --Para 11:09, 9 October 2007 (UTC)


 * I'm focusing on the pictures situation because I think we almost have something usable, and it is just complicated enough that solving a few details will probably also create tech useful for several things. A non-article task is also a good place to test it.  To make a single map for requested pictures, can kmlexport handle links to several articles/categories?  If so, markup in that one page could contain the kmlexport URL with the various pages hand-crafted into it; to suck in all the reqphoto categories, reqphoto could include a single category for kmlexport's needs.  Would that create a usable map, or is there a technical problem with that?  (SEWilco 04:46, 10 October 2007 (UTC))


 * I added support for multiple articles/categories to kmlexport now, just add more  parameters to the url. Google Maps probably has some arbitrary limit on the number of placemarks it shows at a time, so you will run into problems trying to feed all the points to it at once. Their Mapplets would probably help with that and enable dynamic queries like Google Earth does, but I still haven't looked into all that though it would be useful on Commons too. Perhaps to make all the requests visible at the same time without having to worry about Google's limitations, we should look to Dschwen and the WikiMiniAtlas. The map doesn't need to be too detailed with such wide ranging data, so WMA's base map would work fine, it'd just need support for kml or some other format we'd both implement. --Para 16:06, 10 October 2007 (UTC)


 * I think Google Maps can be made to retrieve new data when zooming in to an area (yes: a server request upon clicking, so you need a Google Maps servicing script on a web server), although it would also be possible to emit a generic map and when a click is made, emit a map for only that region. Or to splatter the fewer "Requested pictures" coordinates across a map while showing only one icon for the "Category:Requested pictures in..." groups until zoomed in to an area (showing at least one icon in an area shows there is something to zoom in on).
 * For Wikipedia's needs, the clickable map should include a link to the relevant article someplace, so one can browse down to a city view and get a link for the details about a point. We can add next to the "Requested pictures" GeoGroupTemplate (properly labeled so they can choose original or extra chunky) a more complete version with whichever map tools can include the subpages and categories.
 * For kmlexport, then, we can craft a request with all the needed subpages and categories handcoded. I have editors which can create that if you want me to.  (SEWilco 16:43, 10 October 2007 (UTC))

In my experience Google Maps does make a number of kml requests to the server in the beginning of a session, but only in the beginning, and doesn't display a huge number of placemarks at a time unless requested to do so with Javascript (it offers a "show more" link at times). But even then it only uses the results of the first requests, from what I've seen. It's perhaps possible that the behaviour could be controlled with http caching and expiry headers and kml directives, and the current additional requests could be just Google saving the interweb data for search purposes. I don't know how GM chooses which to show at first, but it's probably the order they're received in. Now that GM started supporting network links in kml files though, the situation might change somewhat, and I could maybe work on improving the Google Earth interface we were testing above... but that had its own problems.

Another issue is that when the number of coordinates increases, so does the processing time, and Google Maps has a short timeout for how long to wait for a response. This happens with pages where it takes a long time to retrieve the page from Wikipedia (like a page Docu has been playing with, here), or long database queries like the Google Earth ones. One solution to the timeout problem could be to make an intermediate landing site for Wikipedia pages to link to, which runs the query, saving the results to a temporary file preserved for some short time, and after finishing the query redirects to Google Maps with the parameter to use the temporary file as a kml source. Might work, but with the browser interface only.

Retrieving new data would indeed work with a group placemark per category, but I think states and other boundaries used with categories may not be clear when people just want to see requests from a rectangular map area, which may cover parts from many categories. Also, the single requests to be splattered are probably less important than the ones from busy areas, so they shouldn't be made more easily available than the important ones (on a scale of page length or links to pages or something like that, which is more likely to be higher for places where there are many people and therefore many other placemarks or requests).

Handcrafting a list of pages and categories wouldn't be difficult, but if that's to be used with all the image request categories as an http parameter, it would probably hit an http request length limit. Easy solutions would be adding a common category to all the templates so there would be only that category plus a couple of pages to feed to the parser, or me implementing a recursive query with some depth limit, but that might again end with the timeout problem. --Para 17:24, 19 October 2007 (UTC)


 * Timeout: Cache the KML created for an article. When a request is made, provide the cached KML with a refresh time.  Then check if the Wikipedia article has been altered and if so then prepare a new KML and Update the Google Earth data.  A web page which is showing Google Maps can also be designed with a refresh timeout.  There are obvious variations on the idea when you start adding multiprocessing in order to try to present the newest data.  The toolserver could also be monitoring cached articles and preprocessing the KML cache, but each improvement is a separate step.  Because timeouts are already a problem it would seem to be a good idea to add the simplest cache; once that is working then the data becomes usable and cache adjustments become improvements rather than requirements which delay implementation.  (SEWilco 18:58, 19 October 2007 (UTC))


 * I think there are only two templates to which a common Category would have to be added: "Reqphotoin" and "reqphoto|in=". The Talk page of the Category should include mention of the reason for the blanket Cat.  (SEWilco 18:58, 19 October 2007 (UTC))


 * Aww. The RSS feed on a Category only applies to edits to the category's text, not to adding the category to an article.  Can't use RSS to detect article changes.  (SEWilco 13:04, 10 October 2007 (UTC))


 * CatScan shows recent changes to a category's articles in HTML format, that might help a bit. RSS output is on the todo list, unfortunately for quite a while already. Some more poking might help. --Para 16:06, 10 October 2007 (UTC)


 * Doesn't matter yet. I was just looking for options for the toolserver updating its info and caching maps before they are requested, so as to reduce load by simply serving cached data.  (SEWilco 16:43, 10 October 2007 (UTC))

Mapsources
The new implementation of the region parameter worked quite well (good work!), but it currently seems to break the page. Not sure what was changed in the last couple of hours. -- User:Docu


 * There's been some fine-tuning, and looks like the last change a few hours ago broke the whole thing and none of the links work anymore. That's what happens when changes aren't tested well enough. Sorry about that, we'll have to wait until later today when Magnus has time to fix it. Perhaps the tool url needs to be changed to something else for the time being. --Para 13:44, 14 October 2007 (UTC)


 * It works if the URL doesn't include a region parameter. -- User:Docu


 * That's odd, since the culprit is PHP's DOM Functions, which urlencodes the variables used in the links and is run with all requests regardless of parameters. Anyhow, it's easily fixed, but unfortunately only Magnus can touch the code. --Para 13:57, 14 October 2007 (UTC)

That's progress
Oh, good. My summary and list of coord proposals is collecting comments in some semblance of order. No wonder the discussion was so entangled, with so many variations of the issues. I just hope they start flowing in the same direction. (SEWilco 17:26, 18 October 2007 (UTC))


 * All good progress. There's still a number of other issues on people's minds, like GeoHack usability, coord vs other templates, and coordinate link placement in an article, and I don't know if they should be brought up separately or left as lonely comments in each proposal. With this run we're getting rid of the really external links and it would be a shame if that stalled because people can't agree on other issues. --Para 15:34, 19 October 2007 (UTC)


 * Yes, that's why I tried to focus on each specific point about the Geolink topic, so we'd try to resolve the EL issue so we can move on to the other issues. The points had gotten scattered among the discussion threads.  (SEWilco 15:53, 19 October 2007 (UTC))


 * Can you put "Oppose" ahead of your Eh comment? I think you're opposing.  It's because I am not certain of the intent of the phrasing that I'm seeking confirmation of the interpretation.  Incidentally, look at the bottom of that Talk page for questions about the specific phrasing to use.  (SEWilco 18:39, 22 October 2007 (UTC))


 * I don't understand what I'd be opposing. Would it be refuting the misunderstanding that that section is about replacing the contents of Geolinks? Can you edit the subsection to better explain what it applies to and why? The question at this point shouldn't be about the phrasing, but about what's to be included (text? links? coordinates? visible ones?), how, and where. It's not looking good if the people most active in the proposal aren't understanding each other... --Para 19:08, 22 October 2007 (UTC)


 * Looks like the one-liner is preferred but there aren't many participants. Because global replacement implies template deletion I splattered participation invitations across the Geolinks templates.  Let's see if anything changes in a week (it's a good idea to wait for the weekend participants).  (SEWilco 19:48, 23 October 2007 (UTC))

Merging Hcard-bday into Birth date and age
Hi Para, The merge notice you added to Template:Birth date and age/doc is now so old the decision has been archived. Do you still want to do this merge? -- PatLeahy (talk) 04:14, 23 October 2007 (UTC)


 * Certainly. The same improved semantic markup with coord has been put to good use. --Para 11:22, 25 October 2007 (UTC)

Google Maps KML change
Apparently more KML is supported in Google Maps now. (SEWilco 03:00, 24 October 2007 (UTC))

Maps of requested pictures
How come I can't get your amazing tool to work for other locations? What am I doing wrong? (By the way, the exciting possibilities of such a tool—which none of us suspected was already in existence—were raised at the recent New York City meetup) Thanks.--Pharos 17:29, 6 November 2007 (UTC)
 * OK, never mind. I got it working.  But is there a more user-friendly interface way to play with it than just changing the URL?  Thanks again for this amazing tool.--Pharos 17:42, 6 November 2007 (UTC)


 * OK, I've created Reqphotomap for use on category pages. Is this a good idea?--Pharos 17:53, 6 November 2007 (UTC)
 * Ech, I've deleted it; I can't get it to work right.--Pharos 18:09, 6 November 2007 (UTC)


 * Good to hear of a bigger crowd with the same idea. The tool is currently linked from GeoGroupTemplate, which can be used in articles to get a map of all the article's coordinates (or those of a request page), or on a category page to get a map of the locations of all the articles in the category. The previous couple of pages above, SEWilco has also been pushing for a view of all the photo requests, but I have been busy lately and the functionality hasn't seen much progress. Meanwhile, urls shouldn't be too hard to change by copypasting a category name from the Wikipedia url, but I can add a html form with the necessary urlencoding to the kmlexport page if you'd prefer? --Para 19:03, 6 November 2007 (UTC)

Stephens City, Virginia/OTRS
Thanks for posting that link on the talk page, I greatly appericate it. Is there anything that needs to be posted on the article page itself? I have already sourced the link to the article but want to make sure I am not missing anything. If so, please let me know. Take Care....NeutralHomer T:C 07:54, 19 November 2007 (UTC)


 * It's fine as it's now. In all Wikipedia articles the GFDL licensing is mentioned at the bottom of the page already and all attribution is in the list of authors on the history page and OTRS notices. --Para (talk) 11:27, 19 November 2007 (UTC)

Utility of a utility
It hadn't occurred to me that GeoHack might be of use to others. I guess its becoming more closely tied to Wikipedia articles may also protect the server from non-Wikipedia traffic. (SEWilco (talk) 03:25, 22 November 2007 (UTC))


 * Most tools related to coordinates would probably be useful for the world, at least if generalised a bit. That's kind of what the Wikimedia projects are all doing isn't it. If external use of useful tools isn't significant, it's not a problem, but the conflict is that the main servers in the toolserver cluster have been donated to help use and maintain the information in the projects, but not to provide new content themselves. Independent tools that don't use Wikimedia data or generate it for Wikimedia projects, but for some external use, wouldn't fit that requirement. To take GeoHack as an example, it's an independent tool even though the map sources page is collaboratively edited and stored on Wikipedia for convenience. It's open source though so outside interest wouldn't hurt, since non-wikipedians can't just have "us" implement features that aren't needed here, but they'll have to do it themselves and release the code. Who knows, maybe they might even come up with something wiki-utile. --Para 16:05, 22 November 2007 (UTC)


 * It will be nice if someone can use the tech, but I'd rather flood the Wikipedia servers with Wikipedia work. (SEWilco (talk) 00:10, 23 November 2007 (UTC))

Template:Geolinks-US-colorphoto
If you're going to remove the template (which isn't a bad idea), please propose it in Templates for deletion, using the proper procedure. If consensus is reached, a bot can remove them faster than you can. &mdash; Arthur Rubin | (talk) 21:04, 5 December 2007 (UTC)


 * Please see Wikipedia talk:WikiProject Geographical coordinates, this has a bit of history. All the geolinks templates will be deleted soon, but there's some work left before that. Manual review was needed as some articles didn't have other coordinates than with this template, and there were only a few dozen uses altogether. --Para 21:11, 5 December 2007 (UTC)


 * Sorry. I traced the references back, and found the project link.  (I still think a bot could have changed the template to undefined, and keep track of duplicates, but I could be wrong.)  &mdash; Arthur Rubin |  (talk) 21:14, 5 December 2007 (UTC)

Hey, thanks
For the vandalism revert you made on my page. Cheers mate,  Dfrg_ msc  22:14, 12 December 2007 (UTC)

Geolinks-cityscale
Please note that I have reverted this page to its pre-discussion state. As far as I can see there has been no concensus about the way forward for these templates but simply a 7:4 vote. This was both non-unanimous and non-democratic (I cannot see a way to contribute to this discussion as it's an "archive".) Unless there is consensus, the status quo should remain.

The reason for cityscale (I was the originator) is the ability to access Google Maps etc with one click. The geohacks page is fine for those who want other mapping systems but an unneccessary extra step otherwise which is the reason to have two ways of accessing Geohacks (top right and co-ords) and one each for the non-intrusive short list of links.

I have been helping to keep this template efficient and uncluttered for three years or so. The discussion other people have had is reminiscent of the planning application for Douglas Adam's Vogon hyperspace bypass. --Scotthatton (talk) 13:46, 20 December 2007 (UTC)

*Ting* Mmm, toasty icons...
You assignment Graphic_Lab/Images_to_improve is complete. Please take your new icons and leave. :) Seans Potato Business 03:29, 5 January 2008 (UTC)

Google Earth, Coord template questions...
Hi Para, I've been lurking around and researching the coord suite of templates for a couple of days, and you seemed like someone in the know. I work on an intranet that has a full enterprise google earth server, as well as a mediawiki instance. I'm trying to find information about how, exactly, the google earth server apparently leverages the geohack services to pull geopositioned wikipedia pages out and display them as a base layer in GE, it seems like there isn't much information out there on the specifics. I've learned a lot about how to actually build the logic expressions from the existing Coord templates. I am thinking, however, about just enabling the GIS extension but am not sure how i could hook up the google earth server to that. any ideas would be helpful! cheers, Corazon Vacio (talk) 23:01, 14 January 2008 (UTC)


 * Hmm, this question has been asked often enough for someone to write something about it on WP:GEO. I don't know anything about Google Earth Server, but the last time I remember talking about this was at Talk:Google Earth#Infrastructure & implementation of wiki integration. Google can probably help better with their enterprise products than I can. But before that, you need to have the coordinate data from a MediaWiki database available, and for that the database dump needs to be parsed for strings that look like coordinate templates, or if the database is small enough, fetch the pages directly and parse on the fly. WikiMedia projects have the dumps in XML, and User:Dschwen has a reference implementation on http://tools.wikimedia.de/~dschwen/commons_geo/ for parsing a dump using Perl and regular expressions. That ought to get you started. --Para (talk) 00:30, 15 January 2008 (UTC)


 * Thanks, that's helpful. If and when I get this working, I'll contribute some documentation to the WP:GEO project. Corazon Vacio (talk) 01:13, 15 January 2008 (UTC)


 * I've seen mention, I think in a Google Earth blog, of the search engine googlebots being programmed to recognize two or three coordinate formats. The note mentions that the coord format is being recognized.  But I think it's the source code which is being recognized, so it's not enough for coordinates to look in HTML like what coord emits, it has to be coord which is being invoked in the article.  -- SEWilco (talk) 01:28, 15 January 2008 (UTC)

WikiMiniAtlas
This is deadly, as we'd say in Baltinglass. Can you add a little message to the template to say "click on globe to view map"? That would solve everything and we could all stop adding maps that you don't like! Sarah777 (talk) 14:43, 17 January 2008 (UTC)


 * (note that I'm not Para responding) Put your mouse over the globe for several seconds. Making the message more visible would make the coordinate label much more intrusive; discussion about such messages might belong at WT:GEO, but there is a different messaging discussion going on there at the moment.  -- SEWilco (talk) 16:23, 17 January 2008 (UTC)


 * Well, I've been reading, looking at and even adding that coordinate template for ages and never realised that it could produce a map. Needs a message to be useful to the average reader. (Sarah777 (talk) 19:07, 17 January 2008 (UTC))


 * (I am!) Glad you had such a positive first reaction, though User:Dschwen deserves all the credit for his hard work in creating an entire map service from scratch. I've seen the comment somewhere else too, that it may not be obvious to people that the globe icon is clickable and leads to more information. Unfortunately I can't think of a solution for the problem.
 * Adding instructive messages to articles wouldn't be a good solution: Wikipedia uses hypertext in quite an exemplary way by linking descriptive words to pages with more information about the topic, instead of linking something like click here. This good rule of writing articles is easy to keep in mind if you think how the page would read if you printed it, and it seems to be covered in Self-references to avoid already. With locations, the names of the locations are reserved for Wikipedia articles about the locations, but coordinates don't really have such meaning other than the location itself. Like you've noticed, its purpose may not be obvious, and I don't know if Wikipedia can or if it should change that situation. Using a descriptive icon such as a globe is a good hint already of what the mystic numbers might be related to, together with the word "coordinates".
 * Wikipedia generally is a bit of an exception to other websites by not using icons to link somewhere else than a bigger version of the icon. With the WikiMiniAtlas it's also an exception in showing an icon next to text that before clicking seems to represent the same thing, but in fact links to two different places. That you and many others haven't noticed that clicking on the globe actually has an effect could be a result of our Wikipedia experience, and it may be difficult for us to say what a random reader thinks of the icon. SEWilco mentioned somewhere that it might be good to have a section in Wikipedia's Help / Browsing Wikipedia / Basic navigation about using map services from Wikipedia articles, similar to a "key" of the page layout in printed encyclopedias.
 * There have also been plans to add a slightly bigger WikiMiniAtlas on the GeoHack page, so then a reader will always see the Wikipedia map service before anything else, if only first they realize what coordinates are. --Para (talk) 19:51, 17 January 2008 (UTC)

Thanks for coordinates for Zhaozhou Bridge
Thanks for your quick response to my little request (in my edit history for Zhaozhou Bridge) for coordinates. I followed the coordinate link to GeoHack and then to Google Earth, and the view is very nice. Shows how the bridge has been turned into a park after the diversion of the river and the highway; there's even a freeway bypass under construction to the west.

A question: why did you use

instead of
 * 37.72017°N, 114.76325°W ?

Both seem to work, and I gather that "coord" is the new standard, with the plan being for bots to convert all coor at dms to coord.

I'm also curious what tool you're using to find such requests at mine so quickly. Wdfarmer (talk) 17:08, 20 January 2008 (UTC)


 * Mapping for China is exceptionally badly available online for people who don't read Chinese, and I hope Wikipedia and other community contributed projects can help with that. I based my guess on Panoramio images geocoded to the vicinity of the bridge, and those together with the satellite images are indeed very nice.
 * The two coordinate templates don't really have any significant difference so I wouldn't care much about which to use, but yes a bot run will hopefully be done at some point to have just a single template. Until then, when I enter coordinates to Wikipedia I usually just use the template that's quickest to type or paste, and that's not always the same template.
 * As for finding your request, there has been some discussion lately on map service links and so I've been monitoring the use of coordinate templates and comments about it through the toolserver database, and that's how I saw your summary. In the future if you'd like help from others, you can add the article to Category:Articles needing coordinates or place the LocateMe template on the article's talk page. For general use though, I could put up some tool on the toolserver to filter recent changes based on the summary, maybe that would be useful for other purposes as well? --Para (talk) 18:15, 20 January 2008 (UTC)
 * If you look at my list of contributions you'll see that I didn't edit Main coordinates templates until after you'd responded to my request. All I'd done prior to your response was just put "Coordinates are still needed" in the comment of an edit to Zhaozhou Bridge.  So I figure you must have been doing some intense monitoring of the Recent Changes display for comments containing the word "coordinates".  I didn't know about LocateMe, so thanks for that. Wdfarmer (talk) 22:00, 20 January 2008 (UTC)

External map links
Hi - Yes, I've loads of links to map sites on the French communes I've copied / translated from fr:wiki where the links seem to be standard. I'll cut them out in all future editions. Those 250+ I've done will receive attention as and when I've time (unless you'd like to help !? ) Dickie (talk) 12:10, 25 January 2008 (UTC)


 * It would be a bit tedious to do manually, but shouldn't be too hard to have a bot do the job, as all the links are in the external links sections and each on a single line. All we need to make a bot request is a list of articles and beginnings of the links to remove. --Para (talk) 16:28, 25 January 2008 (UTC)

Map service icons
I'm guessing my last post was too surreal for you to deal with effectively so I'll keep this one sane. You posted an assignment on the graphics lab ages ago: Graphic_Lab/Images_to_improve and it has been completed. Please do whatever it is you intended to do with the icons, and leave a note at the bottom of your assignment to let the graphics lab know that the assignment can be archived. Thanks. Seans Potato Business 11:56, 26 January 2008 (UTC)
 * Sorry, I see you did respond to my last post. I guess I could have found a way to have made this one surreal as well... :/ There's an unsigned message at the bottom of the assignment - did you type that? Is this assignment finished now? Seans Potato Business 12:01, 26 January 2008 (UTC)


 * The icons are ready as far as I'm concerned, but I don't know if the community will find them tasty, plus we don't have the CSS brand butter yet and I assume the picky lot won't take any other kind. That's what I've been waiting for, but it depends on what the Lab wants to do in cases that need community approval before they can be closed. Maybe it aims for fast turnover? The unsigned comments are from Blacklemon67. --Para (talk) 16:41, 26 January 2008 (UTC)

Notification of new post in "resolved" ANI thread
I've made a point about custom edit summaries in an ANI thread. See here. Notification left because the thread was previously marked "resolved" (I've removed the resolved label as I felt the issue is not resolved). Comments would be welcomed. Carcharoth (talk) 01:22, 3 February 2008 (UTC)

French communes/place-names
You wrote:

Hello. You have been identified as having added or removed direct external map service links in articles. There is a discussion at Wikipedia talk:External links about which should be done, and some more opinions would be good to find community consensus. --Para (talk) 23:07, 8 February 2008 (UTC)

Yes, I've added articles for some French place-names (not very many, no more than a dozen or so - mostly to avoid redlinks in articles I happen to be working on).

There is far too much to read (for now) in the discussion page whose link you provided. My initial reaction to your proposal is unfavourable, but I'd really need to read the whole thing to come to a firmer decision.

The articles I've added have been pretty much straight translations of the French, with the odd bit lifted from other language wikis, if available and useful; sometimes I've added a dablink or a short sentence or two relevant to other articles on en:. I've noticed that fr: has more-or-less the same links on each article (but I haven't checked in detail), and I'm inclined to leave them, on the assumption that participants in fr: know what they're doing, and we should follow their lead. Have you asked wikiproject France for their opinion? --NSH001 (talk) 01:01, 9 February 2008 (UTC)


 * Ok, thanks, I've posted there now. Should've thought of it earlier, as there indeed have been many users doing the exact same thing with French articles. I imagine the French articles have those links only because some editor chose to put them there and others haven't paid attention. Yet. --Para (talk) 02:33, 9 February 2008 (UTC)

External map service links
Hello. 18,000 external map service links? I strongly support developing a style guideline. Also, FYI, I responded here Wikipedia_talk:External_links adding some Exceptions after your request for comment that you left here User_talk:Rosiestephenson. Cheers Rosiestephenson (talk) 01:54, 11 February 2008 (UTC)


 * Hi, and thanks for the comments. The number of external map service links is down to about 12,000 now, thanks to reduction through some templates. I answered on EL, and it's getting complicated with all the possible data... --Para (talk) 09:27, 11 February 2008 (UTC)

Hi,

Thanks for the message on my talk page. I did click on the link that you provided, and saw the extensive debate that has been taking place there. I don't have time to review all of that or to join it, so I thought I would simply let you know that I don't have any massive objection to map links, although the argument that it is better simply to have a map or maps in the article itself does make sense me. I removed those map links because they were covered with banners, Adsense, and various other kinds of advertising. --Bcnviajero (talk) 13:04, 11 February 2008 (UTC)

OTRS apuja
voisitko vilkaista keskustelua kahvihuoneessa ja kommentoida sitä

voisitko katsoa nuo OTRS-tickettiä odottavat kuvat ja jos lupa ei ole kunnossa, palauttaa epäselvä mallinteen--Musamies (talk) 14:54, 14 February 2008 (UTC)

Geohash
I noticed Geohash, which is an interesting idea but I don't know if I have much use for it yet. The removal of precision idea is an old one, and I remember seeing it in the NAPLPS vector graphics standard decades ago. -- SEWilco (talk) 18:09, 27 February 2008 (UTC)


 * It's kind of tinyurlish. Cool, but not that useful when few interfaces these days are that limited on data size. A bit too much obfuscation just to get rid of separators and signs really. Even with the human mind it might be easier to use other commonly known mnemonics to make coordinates pronouncable. And to go back to the url comparison, there aren't many things people often refer to that have been measured so precisely that a sequence of letters would be considerably shorter than the original numbers. But good to know anyway in case people start spreading them the way they have with tinyurl. --Para (talk) 16:39, 14 March 2008 (UTC)

Samantha Brown Image question?
http://en.wikipedia.org/wiki/Image:Dots_SamanthaBrown.jpg You did the OTRS for this image, so I was wondering if you could help. An Admin has taken it upon himself to make this image as disputed even though I forwarded all the permission details to OTRS once I received them from the copyright holder. On the copyright holder's flickr account It says all rights reserved, but in the emails that were sent I clearly used the form letter for obtaining permission and he agreed to the releasing for gnu free license. I don't have the emails anymore, so what should be done. --Kolrobie (talk) 18:21, 9 March 2008 (UTC)


 * Don't worry, it won't be deleted. The permission is clear, but the admin probably doesn't have access to it, isn't aware of OTRS or just didn't notice the mention. I'll take care of it. --Para (talk) 19:08, 9 March 2008 (UTC)

Thanks--Kolrobie (talk) 19:32, 9 March 2008 (UTC)

I think we're doing better
I thought you might find this amusing:  -- SEWilco (talk) 16:04, 14 March 2008 (UTC)


 * True true, we probably do better in explaining many topics to common people. I suppose that can be expected when comparing to something from Springer, and again many other topics here would need complement on that more advanced side. But at least our project also has the "with" instead of just "of". --Para (talk) 16:24, 14 March 2008 (UTC)

Converting templates
I'm thinking of converting all the references of Geolinks-US-cityscale to coord. I noticed from one of your edits that you have some knowledge in this area. I have already done Astoria, Oregon.

I removed from the external links section and replaced it with  just before the infobox. Got any advice?

I have been working on some mountain articles and I'm starting to wonder about how many times you can find a reference to the coordinates of a locality in a single article. It often appears in the title bar, the infox, the external links, and sometimes in the article itself. In mountain articles I've been including the coordinates in the info box in dms format and in the title bar in decimal format and removing other references. I think decimal format is becoming more useful as people use GPS systems more often. Got an opinion? I'm good at taking suggestions and not fighting about it. --DRoll (talk) 20:50, 21 March 2008 (UTC)


 * Good project! I don't know what you've read about it yet, but like most things on Wikipedia, it comes with a bit of history and mixed preferences. Wikipedia hasn't always had all the structures available that we're used to now, so some information may seem to be in odd places sometimes. Though coordinates have been moving away from the article body lately, they may still have a place there if there's other similar information near them, like coordinates of another related location, geographical information, or links to maps, but I'm all for presenting information in a consistent way. Having the same coordinates on a page more than once though, especially when they're in different formats, may lead to deviation when someone decides to fix the article's coordinates a bit and doesn't notice the other mentions, so your workaround for avoiding clear redundancy may backfire there. There's lots of back and forth editing going on however, with title coordinates being added to one group of articles and other people moving them to infoboxes only in another, so I can't suggest any long term solution. But that's less likely to happen when the infobox has many decorations at the top to take space, like in the Astoria article. As for the format, I prefer dms myself because it's closer to the definition of the measuring system and good interfaces should allow the use of both regardless of the input format, but again there are editors adamant about their personal preferences and it'll be hard to get articles consistent without some real software support.


 * After all that blabla, here's some pointers to more concrete recent things: WP:GEO discussion with a group of editors insisting that articles that have linked to map services from the external links section must continue to do so with coordinates (see the surrounding sections for background), and some preparation for replacing the Geolinks templates at Wikipedia talk:WikiProject Geographical coordinates/geolinks replacement. Personally I think you're on the right track and can continue with that, but is it a bit much to do manually? --Para (talk) 00:34, 22 March 2008 (UTC)


 * Thanks for the advice and the time. --DRoll (talk) 17:31, 23 March 2008 (UTC)

Recent interest in reqphotoin maps
Discussion spilled over from elsewhere about reqphotoin maps to Template talk:GeoGroupTemplate. I don't know where there is documentation about the details such as how much delay exists. Sorry, I won't help with the automation programming until the ArbCom reverses its decisions involving my bots. -- SEWilco (talk) 17:51, 24 May 2008 (UTC)

Wikimapia-> Coord
Requesting a review of all removals/conversions because of some comments raised in disscusion...

I will however once the template docs been read look into migrating some links over to the new style coord +scale ones (possibly in reference to what was there before..)

Thanks for the support though..- And to be fair in some cases the information loss caused by the loss of zoom values IS significant. Sfan00 IMG (talk) 18:17, 20 June 2008 (UTC)


 * Postscript1: I note the new coord has a source param - That might solve one issue :) Sfan00 IMG (talk) 18:35, 20 June 2008 (UTC)
 * Postscript2: No objections to me checking or tightening existing coords then? Sfan00 IMG (talk) 18:35, 20 June 2008 (UTC)

The "coord" template
What should I do to fix them? Sfan00 IMG tells me that I should ask you for help. Yamakiri TC    [ §]  06-20-2008 • 22:10:26
 * You can leave a message for me on my talk, or reach me at our irc from 16:00 to 03:00 Wikipedia time. --Yamakiri

Brown Dog - Wikimapia
Thanks for that piece of advice. I was not aware that the photos had been lifted, nor did I give permission. It's disappointing. That said, I'm not going to pursue the issue with WM, in large part because I'm too idle to surmount the submission of an electronic signature which appears to me to be a barrier to raising a DCMA takedown notice - WM's preferred means of being advised of copyvios. I have my suspicions as to the wikipedian who caused it to appear on WM. Hey ho.

I'm more exercised about the WM links at the foot of BDA, but I've said all I have to say about them in BDA talk ... and have taken BDA off my watchlist, 'cause life's too short for conflict. That's a shame too.

On the plus side, the whole thing has reminded me that there are several thousand Dictionary of National Biography entries yet to be ported over to Wikipedia, so I'll aim to content myself by contributing to that activity. But thanks again for the advice. --Tagishsimon (talk) 19:08, 24 June 2008 (UTC)

Roosevelt Raceway (harness racing)
I'm curious why you changed the map link to a coordinate link, rather than just adding it on. Also you left the wording after the link, which was rendered meaningless. Is this a preferred form?  Citi Cat   ♫ 04:21, 25 June 2008 (UTC)
 * It's best to keep external linking neutral when there are so many providers of aerial images of the area (WP:ELNO#15), and just give the information that's usable with any map without relying on any single one. I didn't see the item mentioned after the link on the map and so didn't pay much attention to it, but looking at it again makes me think that the whole mention should be in prose. Maybe I'll fix that, and you can then add more detail with appropriate sources. --Para (talk) 10:33, 25 June 2008 (UTC)

Geoli.
Eventually, we should probably update [1] as it still includes the obsolete way of specifying coordinates. -- User:Docu


 * Yep. I found other links from WP:WikiProject Sydney/Suburbs, WP:WikiProject Scouting/Local Scouting articles (BSA), WP:WikiProject Piers, WP:WikiProject Microformats, WP:WikiProject Cities/Guideline, WP:WikiProject Bridges, WP:WikiProject Adelaide, WP:Template messages/Links and WP:Coordinate-referenced map templates. There may still be other pages with such mentions without links, which my database queries won't find. I remember talking somewhere that we should either make a list of all pages that describe how to enter coordinates and then watchlist those pages, or prepare short instruction snippets for a few different purposes and transclude them on the project page and all those other referring WP pages, so that all the instructions stay consistent. --Para (talk) 09:53, 28 June 2008 (UTC)
 * I updated WP Ohio recently, but that is an inactive page. -- User:Docu
 * BTW, I didn't find anything on WP:WikiProject Bridges needing updating -- User:Docu

Blocked for edit warring
As per your block log. I have also blocked Crum375. See [6]. Viridae Talk 01:13, 12 July 2008 (UTC)


 * Oh, an interesting choice of actions. If anyone seeing this actually cares about something else than drama, SlimVirgin kindly started a resolution at Wikipedia talk:External links. --Para (talk) 02:10, 12 July 2008 (UTC)

Para, please see WP:ANI#I_have_blocked_Para_and_Crum375_for_edit_warring_on_Brown_Dog_affair. Per that discussion, if you will agree (1) to stop reverting and (2) talk it out on the talk page, possibly using dispute resolution, then you will be unblocked immediately. Thank you. --B (talk) 02:12, 12 July 2008 (UTC) Mooted per another admin removing the block. --B (talk) 05:16, 12 July 2008 (UTC)