Wikipedia talk:Manual of Style/Dates and numbers/Archive 70

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 65

←

Archive 68

→

Coordinates section

The section about coordinates was recently changed. Was there some discussion of this? -- User:Docu

The Geographical coordinates section was recently updated to say

As of April 2007, templates {{coor d}}, {{coor dm}} {{coor dms}} are deprecated. they have been superseded by {{coord}}.

Here are the relevant changes [1]. Was this change debated somewhere? If not, does it truly reflect the consensus of the community? I ask this as there appears to be some disagreement about whether to use {{coord}} at Template talk:Infobox lake#hCard microformat (and coord) -- Patleahy 07:17, 19 May 2007 (UTC)

Yes. I'll dig out refs later; it was a while ago. {{coord}} does everything the other templates do; and some. Andy Mabbett 09:58, 19 May 2007 (UTC)

Discussion ranged over a number of talk pages, but was chiefly at Template talk:Coor dms (the "parent" talk page for the family of superseded templates), the talk page for {{coord}} and at Wikipedia talk:WikiProject Geographical coordinates#New template replaces "coor" family. There were pointers to one or more of those, from various project and VP pages Andy Mabbett 11:03, 19 May 2007 (UTC)

The discussion was not concluded and the community did not reach consensus. As long as people editing Wikipedia and people reusing the content do not support the change (with Google most visible), the existing templates should not be declared deprecated. --Para 11:24, 19 May 2007 (UTC)

Consensus was reached. There were no objections to replacing the old template with {{coord}}. Once that was done, it was found that it caused issues with the parser for the Google Earth layer, and so the redirection of existing templates was reversed, pending a fix. The editor who liaises with Google Earth has failed to respond to requests for an update, for over a month (I've just, again, prompted them to do so). I'm unaware of any other dissent, apart from two editors who revert the use of coord but who have failed to respond to multiple requests that they articulate their reasons for doing so. The Google Earth issue does not preclude the use of coord for new coordinates additions. Andy Mabbett 11:47, 19 May 2007 (UTC)

Oddly neither this page nor Wikipedia talk:WikiProject Geographical coordinates shows a consensus. It looks mainly like pigsonthewing declared "it replaces", but it didn't quite convince. Besides, pigsonthewing himself added a notice to the usual templates stating that it's a proposal Template:Main_coordinates_templates&diff=125615811&oldid=125592116. -- User:Docu

The wording on that was changed to "proposed" at your insistence. Andy Mabbett 12:01, 19 May 2007 (UTC)

I don't see any discussion of this change prior to the change. The discussion that took place after the change clearly does not show a consensus for making this change. This change to the MOS should have been discussed on this page first. I think this change should be reverted until consensus for making the change can be determined by a discussion on this talk page. -- Patleahy 14:41, 19 May 2007 (UTC)

How about a compromise solution:

1. Have the MOS instruct writers to use of the {{coor *}} family of templates. This will allow existing machine readers continue to function.

2. Update the {{coor *}} templates to include Geo microformat tags in their output.

-- Patleahy 15:07, 19 May 2007 (UTC)

"I don't see any discussion of this change prior to the change" - probably because no changes to the style of Wikipedia have been made; only the underlying technology.

"How about a compromise solution" - I'm not sure what problem you're trying to solve here, but I see no point in making extensive changes to templates which are intended to be replaced, when their replacement already exists, and when, in come cases, it would not be practicable, if indeed it is possible.

Andy Mabbett 15:39, 19 May 2007 (UTC)

Changes in the underlying technology should be discussed too.

There is a difference of opinion as the whether these templates should be replaced. That is what we are discussing here and that is the problem my compromise is trying to solve. -- Patleahy 15:52, 19 May 2007 (UTC)

"Changes in the underlying technology should be discussed too" - indeed. but not here.

"There is a difference of opinion as the whether these templates should be replaced" - cite? As I said above, I'm unaware of any other dissent, apart from two editors who revert the use of coord but who have failed to respond to multiple requests that they articulate their reasons for doing so. Even if that were the case, this is not the forum to debate technical issues which do not affect the style of Wikipedia.

Andy Mabbett 15:57, 19 May 2007 (UTC)

Changes to this page should be discussed on this page. That is both a general practice on wikipedia and a specific rule for the MOS.

The difference of opinion is demonstrated both in this conversation and at the conversation at Wikipedia talk:WikiProject Geographical coordinates#New template replaces "coor" family.

What do you think the benefits and problems with my proposed solution are? -- Patleahy 16:11, 19 May 2007 (UTC)

We are discussing the change to this page, aren't we? There's no requirement for prior discussion of such changes. There is no outstanding opposition to coord on the page you cite, only a request that bot-changes and re-directs be held over until Google catches up (which, for all we know, may already have happened); and no-one is this discussion has stated a reason why coord should not replace the other templates. I see no advantages to your proposed "compromise", and have already outlined some of the pitfalls, above. Andy Mabbett 16:21, 19 May 2007 (UTC)

The top of the page states "Before making any major changes to these guidelines, please use the discussion page to ensure that your changes reflect consensus."

The core issue here is not the forum for the conversation but whether the change should be made. I'll ask a simple question from everyone. Should this MOS page tell writers to use the {{coor *}} family of templates or the {{coord}} template? -- Patleahy 16:39, 19 May 2007 (UTC)

That's "major" changes. Andy Mabbett 19:09, 19 May 2007 (UTC)

On reflection, I don't see any reason why just the later should not be used, but if others still want the former to be mentioned, why not tell eple to "use one of the {{coor *}} family of templates or {{coord}}" with an explanation of the differences? Andy Mabbett 09:54, 20 May 2007 (UTC)

That sounds ok. So please stop migrating other templates to coord, it is making those articles inaccessible for all reuse where coord isn't supported yet. --Para 22:12, 20 May 2007 (UTC)

Can one of you suggest some wording to consider? -- Patleahy 22:54, 20 May 2007 (UTC)

I wonder if having the MOS give two different recommendations will be confusing. See my answer to User:Pc1dmn's (aka roundhouse) question below. Will the differences be meaningful enough for writers to allow them choose? -- Patleahy 03:02, 21 May 2007 (UTC)

Quite. The recommendation to use coord replaces the previous nine options... Andy Mabbett 08:18, 21 May 2007 (UTC)

I have now combined the instructions to include both styles. Looking at it again, I'm not sure people can make the choice, but it's still better than saying that the predominant style is deprecated. I'll have to think about it, but I'm starting to lean towards Patleahy's compromise solution above to include the microformat tags in the coor family of templates, and wait for everyone to support coord before starting to use it. --Para 09:25, 21 May 2007 (UTC)

I have re-worded your changes, since some of what you wrote was simply not true. As I noted above, adding microformats to the old templates "would not be practicable, if indeed it is possible". that's one of the reasons coord was created in the first place. The development of Wikipedia should not depend on - let alone be hamstrung by - the behaviour of external services. Andy Mabbett 09:33, 21 May 2007 (UTC)

Why do you think it "would not be practicable, if indeed it is possible" to add microformat tags to the {{coor *}} templates? -- Patleahy 17:48, 21 May 2007 (UTC)

{{coor *}}. As long as the most visible reuser Google does not mention the new template in their Geographic Web Layer FAQ and Google Earth doesn't have a Wikipedia placemark in a location that was marked with the coord template before the latest database dump, we can say that the reusers are not sufficiently supporting the change and thus I don't think we should support it either. The last database dump started on April 6, when this new template was still beyond broken. When the new template is sufficiently supported, it will be easy to convert everything with a bot and deprecate the old ones here, but until then {{coor *}} it is. --Para 17:16, 19 May 2007 (UTC)

I have posted a notice about this discussion at Wikipedia:Requests for comment/Policies. -- Patleahy 23:26, 20 May 2007 (UTC)

I also added this notice at Template talk:Coord and Template talk:Coor dms. -- Patleahy 23:35, 20 May 2007 (UTC)

So at present if I wish to add {{coord}} what should I do? And how does this affect the various infoboxes such as {{Infobox UK station}} which call {{coord}}? -- roundhouse 01:46, 21 May 2007 (UTC)

As the conversation above now stands you can choose to use {{coord}} or one of {{coor }} family of templates. -- Patleahy 03:02, 21 May 2007 (UTC)

My understanding from the above is that Google may not yet have recognised {{coord}} whereas it has definitely recognised at least 2 of the others, so one should perhaps use the others and await further developments + a consensual bot. (I personally welcome {{coord}}.) Is there any obvious central page to discuss the implications of the activities of the microformats group, who seem to have decided to implement their ideas via info boxes, info boxes being already controversial? (See eg Wikipedia talk:WikiProject Composers + various archived pages, all in the last month.) -- roundhouse 08:58, 21 May 2007 (UTC)

WP:UF, though no such decision has been made, nor enacted - it's simply easy to add them to infoboxes first, since they already apply to large numbers of articles. They're also present in tables, and in-line, in article prose. The ongoing discission at WP: composers does not mean that there is an controversy about the use of infoboxes in other areas. Andy Mabbett 09:29, 21 May 2007 (UTC)

There is in some areas controversy about info boxes (as has been demonstrated), but not in many others, where info boxes exist harmoniously without any dispute, and I agree ought (given a diplomatic approach) to be susceptible to hcarding without too much opposition. I haven't seen many examples of microformats being added to prose to Andy's satisfaction ... I would think, given a neat unobtrusive way of doing this, it might well be uncontroversial. I have seen tables eg in Netherton Tunnel Branch Canal - I can see how that is done but it is arguably intrusive. -- roundhouse 10:22, 21 May 2007 (UTC)

Google support

Is Google working to support the {{coord}} template. If so do we have a commitment from them on when it will be done? -- Patleahy 17:51, 21 May 2007 (UTC)

That would be for our Chief Research Coordinator [2] to say, but unfortunately he has been unavailable lately. I believe the message in this edit still stands, and as to it being related to this style guide issue, I agree completely. Google's general policy seems to be that they don't comment or speculate about future changes, at least not in public. But since the database dumps are months apart, there's no hurry. --Para 22:25, 21 May 2007 (UTC)

"I believe the message in this edit still stands" - no, it was superseded by his post, time-stamped 00:53, 3 April 2007. All the issues identified in this side have been resolved; and he hasn't responded to requests for an update for over a month (but was posting, a week ago). Andy Mabbett 22:33, 21 May 2007 (UTC)

While the problems with the coord template may have been fixed, nothing has changed related to the issues mentioned: (1) the used templates are even moore uncoordinated now that the "deprecated" mention had been in the style manual for almost two months, (2) machine readability is now broken for the 21,500 articles using coord, as opposed to the 173,000 using standard templates, so (3) a part-way transition has been made and reusability is ruined and follows no logical rules. What needs to be done is have all notable reuses support the change, then convert all the uses of the templates at once, and deprecate the old templates. It seems to me that people (or a single contributor really) are trying to do this in the opposite order. --Para 23:16, 21 May 2007 (UTC)

I have come to the conclusion that {{coord}} should not have been added to the MoS two reasons:

1. This change was made without "active consensus, not simply a failure to object"[3].

2. The outstanding issues of machine access have not been resolved.

-- Patleahy 00:36, 22 May 2007 (UTC)

The change pre-dates the comment you cite. Andy Mabbett 16:29, 22 May 2007 (UTC)

It's irrelevant when such a comment was made, as there has never been any consensus to start using the coord template instead of the standard templates. Converting existing uses of coordinate templates into that is disruptive. --Para 19:29, 22 May 2007 (UTC)

There was consensus at the time; and coord is as much a standard template as any of the others. Andy Mabbett 20:31, 22 May 2007 (UTC)

Declaring consensus on your own does not make it so. Please point everyone to the discussion where the community agreed to start using coord over the coor templates and decided a plan for transition. --Para 20:42, 22 May 2007 (UTC)

"Declaring consensus on your own does not make it so" - good job that's not what happened, then. Andy Mabbett 20:53, 22 May 2007 (UTC)

"Please point everyone to..." - see above.

Andy Mabbett 20:53, 22 May 2007 (UTC)

Since Gmaxwell (talk) has been quoted a number of times I have asked for his input. [4] -- Patleahy 16:14, 22 May 2007 (UTC)

Perhaps you missed this. Andy Mabbett 16:29, 22 May 2007 (UTC)

I am not myself in a position to establish (or indeed understand) whether or not machine readability is broken for coord. Andy remains quiet on this matter here, and no-one so far seems to have appeared here from the consensus he claims (and I must say I don't see any consensus at all in any of the links cited by Andy and others above - indeed where there is consensus it appears to be against Andy). Andy continues to change individual articles from the established coor family to coord despite local opposition (eg here) and I would ask for a moratorium on any such changes at least until Gmaxwell (talk) (who has made no edits at all since 14 May) supports the implementation of coord. -- roundhouse 12:36, 23 May 2007 (UTC)

"Andy remains quiet on this matter here" I don't think you have any grounds for that assertion, but allow me to assure you - machine readability is most definitely not broken for coord. There are no grounds or precedence for a moratorium such as you suggest; much less for one reliant on a single, apparently absent, editor. Andy Mabbett 14:31, 23 May 2007 (UTC)

My grounds are that, in this particular discussion, it is claimed above (Para 23:16, 21 May 2007) that machine readability is broken for coord and there had been no counter-claim in this thread for 36 hours. I am not reassured by AM's assurances and would like other opinions. In the meantime the introduction of coord at all, both directly and indirectly via info boxes, seems premature. -- roundhouse 17:03, 23 May 2007 (UTC)

"it is claimed above (Para 23:16, 21 May 2007) that machine readability is broken for coord" - it is not. Andy Mabbett 21:18, 24 May 2007 (UTC)

The Wikipedia community has agreed on a set of rules for entering geographical coordinates, and these rules or specifications have been followed everywhere on the site. Wikipedia has given the specifications also outside the community to make it easier for people to manage the content we edit for everyone to use, and make Wikipedia an even more valuable resource. The data that follows these specifications is machine readable, anything else is not. There is nothing wrong with the coord template itself, and it can be read with a computer just as any other set of characters, but it is not what the community has decided to enter coordinates with, and it is not the format everyone else expects the coordinates to be in. We do not need to wait for any single editor to make a decision, neither to go along with a change nor to stop us from changing. It has to be a community decision. Wikipedia is criticised for being uncoordinated where "anyone can edit" really means it throughout the site. Please let's not make that the reality. Until we agree on a plan for change, have a proposition for the style manual and related instruction pages, have infobox and other templates ready for replacement by tested new versions, decide on a date when a bot converts the templates, and notify interested parties of all this, geographical coordinate information with coord is not machine readable. --Para 00:55, 25 May 2007 (UTC)

This seems to be a fair position - we can't deprecate the working one until the new one works. I am in favour of the coord template ultimately becoming the standard because of its ease of use, however, we do not have a consensus yet and this Google issue (along with other reusers of content) is outstanding. Orderinchaos 19:07, 29 May 2007 (UTC)

Agreed, so where do we go from here, go back to the Geographical coordinates section is this edit: [5]? -- Patleahy 20:46, 29 May 2007 (UTC)

No, {{coord}} is a live, viable template, already used on tens of thousands of articles, To pretend it doesn't exist would be ludicrous. The current wording is an adequate compromise. Andy Mabbett 21:38, 29 May 2007 (UTC)

"This seems to be a fair position" - not while it includes the wholly bogus claim that coord is not machine-readable, it's not. Andy Mabbett 21:42, 29 May 2007 (UTC)

"The Wikipedia community has agreed on a set of rules for entering geographical coordinates" - really? Can you cite them, and point out where they prohibit the use of {{coord}}?

"...anything else is not - as I have already told, you, coord is machine readable (and not just as a text string). Andy Mabbett 21:42, 29 May 2007 (UTC)

"Silence equals consent". The prevailing coordinate notation has stayed the same for two years. The only suggested change was the introduction of {{coord}}, which has been opposed everywhere where all the coordinates in Wikipedia are handled as a whole, and where people are expected to understand the implications of changing the notation in such a short term without any planning.

Most uses of the coord template are in infoboxes, where they were largely placed by User:Pigsonthewing. The user has been banned from doing such edits in the future. This alone shows that the community has not approved the use of coord. But unfortunately the damage has already been done, and reverting all the changes would probably be more trouble than it's worth.

Since existing coordinates are still being converted to coord, my suggestion is to prohibit this in the style guide and move the mention of coord to a subpage, from where it can be brought back at a later date. --Para 16:37, 30 May 2007 (UTC)

Nearly everything in your first two paragraphs is false. Your third is therefore redundant. Andy Mabbett 17:21, 30 May 2007 (UTC)

Paradisal, are you suggesting removing the Coordinates section from the MoS or removing {{coords}} from that section. -- Patleahy 17:33, 30 May 2007 (UTC)

Neither is necessary. The current wording is both adequate and accurate. Andy Mabbett 18:03, 30 May 2007 (UTC)

I meant the removal of {{coord}} from the MoS for now, and have it on a separate temporary page where it can be prepared and discussed for later reintroduction. This would help stop people from destroying existing reusable information by converting it to coord. The coordinates section is not long enough to justify a separate page of its own, but a temporary page just for the coord notation would be useful to show people how Wikipedia plans to enter coordinates in the future. So in addition to going back to revision 119222782 for the coordinates section, the last paragraph could be edited to mention the "suggested change" page. --Para 20:29, 30 May 2007 (UTC)

Nobody is " destroying existing reusable information ": that's hysterical hyperbole. Andy Mabbett 20:44, 30 May 2007 (UTC)

I would support the removal of all mention of coord from the MoS (per Para) pending its full approval, in particular pending full and demonstrable google-compliancy; I am also concerned about the present use of coord in infoboxes. Andy's current state vis a vis blocks/bans is here (last line - max of 1 revert per article per week, edits otherwise allowed). -- roundhouse 19:16, 30 May 2007 (UTC)

Andy Mabbett is continuing to replace coor with coord in info boxes, despite concerns expressed here - see recent diff. -- roundhouse 15:10, 1 June 2007 (UTC)

There is no such things as "Google compliancy", and no requirement for us to obtain "permission" from a third arty before deploying improvements to Wikipedia. Andy Mabbett 10:41, 2 June 2007 (UTC)

Two ends of an egg

I'm distinctly unimpressed by this whole controversy. As the operator of User:The Anomebot2, a geotagging bot, I'm currently by a large margin the single biggest user of geotags on Wikipedia.

Both the {{coor *}} and {{coord}} templates are completely machine-readable, and there's not much difference between them from the point of view of a data consumer. Each requires several different cases to be recognized and processed differently -- it's just that one discriminates between these cases in the tag header, and the other in the arguments within the tag body.

I recently started using {{coord}}, following a request from Andy. I'm disappointed to discover that there was no consensus behind that request, some 2700 article tags later. However, the argument here is not helping progress things in any direction, and reminds me of the argument between Lilliput and Blefuscu.

The problem here isn't technical standards, it's communication between geodata producers and consumers. It doesn't really matter to me what standard we choose, so long as it is well-defined, readable by both people and machines, and accepted by both data generators and data reusers. At the moment, the opinions of data reusers are notable by their absence.

This is not a stylistic issue, it's a technical standards issue, and the Manual of Style is not the place to hold the debate.

Can we please move this discussion to Wikipedia talk:WikiProject Geographical coordinates, where it belongs? -- The Anome 09:43, 2 June 2007 (UTC)

"I'm disappointed to discover that there was no consensus behind that request" - This - facile - dispute has arisen since I suggested, on 11 April that you might "like to have a look at the new template and its documentation, and comment as you see fit?", and seems to relate to the replacement of existing coordinates with a coord version; not the addition of new coordinates as carries out by your bot. There is no reason why coord should not be used in such cases (nor, indeed, at all). As to "communication between geodata producers and consumers", I've been trying to achieve that for two months, and getting no response. Andy Mabbett 10:35, 2 June 2007 (UTC)

Everyone seems to agree that "the opinions of data reusers (on coord) are notable by their absence". I think it is crucial to establish the opinions of data reusers on coord before recommending its use in MoS, in particular where coord replaces the established coor, eg in info boxes. (I am impressed with the activities of Anomebot2; as coords are being supplied where none previously existed, this is undoubtedly several steps forward.)

Moreover there is already a long conversation at Wikipedia talk:WikiProject Geographical coordinates#New template replaces "coor" family which does not seem to me to reflect any consensus there to introduce coord. -- roundhouse 14:49, 2 June 2007 (UTC)

Adding geodata to Wikipedia is a huge project, and requires stable standards. This doesn't mean we can't ever change them, but when we do, we should take care to do so in a measured way.

What bothers me is the constant changes of course, as one group or another declares their proposal to be the one-and-only official format. I'd actually be quite happy with {{coord}}, providing that the major downstream reusers are happy with it too. At the moment, our major users of geodata are the Kolossus team at de:Wikipedia:WikiProjekt Georeferenzierung, and Google. If you want to get {{coord}} accepted, I suggest you get both those groups to buy into it.

User:Gmaxwell seems to have made some fairly reasonable and balanced comments about this. de:Benutzer:Stefan Kühn is probably the best person to talk to at the Kolossus team. Evan Martin from Google participates in Wikitech-l. If you can get all or most of these people happy with using {{coord}} data, I would be quite happy to support a proposal for standardizing on it. Otherwise, without consensus among major generators and users of geodata, sticking with the status quo, even if it's ugly, will sadly remain the best option. -- The Anome 14:58, 2 June 2007 (UTC)

Can we make this change now

There is some mention that {{coord}} is now supported by Google and other consumers of geometric coordinates here. There is also gathering momentum to convert {{coor *}} to {{coord}} here. I therefore ask are were ready to make this change to the MoS now? -- Patleahy 11:50, 8 June 2007 (UTC)

Not that it really matters, but I'd prefer waiting for a message from each of them announcing coord support, or coord marked data visible in their services. --Para 12:22, 8 June 2007 (UTC)

The Anome has made a fine contribution above so if TA were to endorse the change then I would too. -- roundhouse 19:25, 8 June 2007 (UTC)

I have already posted a link to a message from Google saying that they're happy for us to proceed; and I have already stated that I have an e-mail from GeoNames saying that they're parsing coord. Andy Mabbett 19:42, 8 June 2007 (UTC)

What I would like is a reference which satisfies the requirements of Wikipedia:Verifiability or failing that some consensus here, or as I said above an endorsement from The Anome (who seems both knowledgeable on these matters and impartial). -- roundhouse 13:25, 9 June 2007 (UTC)

You're making a WP:POINT by cut & pasting my words from [6]. Please point out where Wikipedia:Verifiability applies to talk pages. Meanwhile, Wikipedia-World have confirmed that they will also parse {{coord}}. Andy Mabbett 14:23, 9 June 2007 (UTC)