Wikipedia talk:WikiProject Dates/Archive

Tagging article talk pages?
Should we even bother tagging article talk pages? The sheer number of articles with dates is overwhelming, and perhaps a list would be better suited to this task. On the other hand, template tagging can be automated by a bot and makes this project more visible, which will (hopefully) bring in more interest from the community and help to build consensus around some of the longer-range plans, such as deciding whether to improve date autoformatting or to disable it entirely. --UC_Bill (talk) 19:56, 4 September 2008 (UTC)


 * I should clarify that this project won't be tackling the issue of keeping/disabling DA. But the people participating in this project will presumably be debating this issue somewhere, and will be making use of the data gathered here.  Also, once a consensus has been reached on what to do about DA, this project will probably take part in the implementation of the solution.  --UC_Bill (talk) 17:08, 5 September 2008 (UTC)
 * Um ... sorry, Bill, consensus was reached last month. You just happen to disagree with it. Tony   (talk)  17:23, 5 September 2008 (UTC)
 * Tony, you're wrong. Consensus has not been reached to disable DA.  Please stop trying to disrupt things.  This project is to work on the stuff that we all agree on &mdash; cleaning up date formats.  Stick to that, and argue about DA elsewhere, please.  --UC_Bill (talk) 17:45, 5 September 2008 (UTC)
 * Please assume good faith: I was not trying to "disrupt things"; I merely mistook your statement to refer to the ''removal of DA" rather than its disabling. My apologies for that. Tony   (talk)  17:49, 5 September 2008 (UTC)
 * If you mean the unlinking of dates, then there's also no consensus to do that, either. The change to MOS (which was done without proper consensus, given the number of people who disagree, some vehemently) was to deprecate the linking of dates, not to encourage the unlinking of dates.  Regardless, we (this project) should start with the articles that have a mess of conflicting date formats, since those are the ones that we all agree need fixing.  Sapphic was working on generating some stats to that end, but given the recent fit she threw over the whole issue, I may take up that task myself.  Once we have a list of articles with mismatched date formats, we can proceed with fixing them.  There's no need to bring DA into this task at all; it can be discussed elsewhere.  --UC_Bill (talk) 18:06, 5 September 2008 (UTC)
 * Oh, it's always easy to bandy about accusations that there wasn't consensus: how long is a piece of string? In terms of the length and depth of the debate, the number of people participating, and the wider enthusiasm for the move (considerably greater than the isolated patches of complaint), I have to reject your claim. If it weren't for the guidelines against canvassing, I could easily have asked some of these people to weigh in. And please take into account those who opposed initially and were soon won over once they saw the improvements. As for the removal of DA, I could equally claim that there's no consensus not to do so. Mark's assertions that it causes widespread "disruption" are hard to take seriously; where is it? (No answer.) And where are the reversions? Almost none of about a thousand articles. That alone says a lot. And Mark raises the bogey of edit-warring; where is it, please? I ask you to accept that the removal of DA has been met with generally positive reactions, aside from isolated brushfires among people who have an unfathomable emotional investment in it.
 * There haven't been many reversions of your non-consensus removal of date linking because some of us don't feel like disrupting Wikipedia to make a point. I just reverted one of your unlinkings but then decided to refrain from any more because it is disruptive.  DON'T BE A DICK.  Have some patience and hold back on the unlinking (there are plenty of messed up dates that aren't linked you could work on instead) while we gather some data and plot a good course of action.  I have some suggestions for UC Bill regarding some improvements to his patch, that might make this whole issue moot.  In the meantime, play nice and stop with the unlinking.  --Sapphic (talk) 00:04, 8 September 2008 (UTC)
 * You're right, this is not the place to expand the debate, but your claims here needed to be rebutted. As for cleaning up dates, well, that's well underway. Tony   (talk)  14:44, 6 September 2008 (UTC)

"Just get rid of the links.--HJensen, talk 22:57, 6 September 2008 (UTC)" "I would suggest that delinking continues, until other WP:TENNIS members come forward, and that Tennis Expert hops off the fence, is WP:BOLD and tries to reflect his own views and wishes. –MDCollins (talk) 00:47, 8 September 2008 (UTC)" "I've wasted enough time on this, but you [User:Tennis expert] are the only one talking contrary. It's true exceptions can be made, but wiki projects do not have the ability make a consensus that overides a general one.... It's a nonsense idea to link unrelated dates. Please stop beating a dead horse, and don't revert edits that follow widespread consensus, common sense, and your own strange insistence that you agree with the edits! 2005 (talk) 07:42, 7 September 2008 (UTC)" "Please follow the guideline, and don't revert those who do..... I don't know if you just being deliberately contrary, but naked years will continue to be delinked throughout the encyclopedia. 2005 (talk) 06:05, 7 September 2008 (UTC)"
 * Your emotive reaction to the removal of DA in articles is puzzling, Sapphic. Have you taken note of the widespread satisfaction with the move. Yes, there is isolated opposition in a few quarters, such as tennis articles, and there by only a single editor against the wishes of his colleagues). Let me quote a few of these other editors from Wikipedia_talk:WikiProject_Tennis over the past day or two:

Tony  (talk)  04:43, 8 September 2008 (UTC)

No endorsement
I'd like people to understand that by joining this project, I am not endorsing autoformatting of dates, and most especially not endorsing autoformatting of dates that are not marked up as dates (whether with wikilinks, or otherwise). --Gerry Ashton (talk) 20:44, 4 September 2008 (UTC)
 * I've added some text to the project page to make it clear. I agree that this project should focus exclusively on the cleanup work, and leave the arguments for/against DA for elsewhere.  --UC_Bill (talk) 15:36, 5 September 2008 (UTC)

Quick survey to see scope of work
I glanced at 40 random articles (not counting disambiguation pages, but counting stubs). I put each article into exactly one category, depending on what I thought was the most interesting feature of the article. (I made up the categories before I started.) The results are: --Gerry Ashton (talk) 20:44, 4 September 2008 (UTC)


 * We can just use the latest database dump to do a comprehensive analysis of all articles. I'll either do that myself within the next few days, or ask somebody at the database analysis Wikiproject to do it.  What would be useful is to know what kind of statistics we're looking for.  Gerry, your list is a great start &mdash; is there anything else we'd like to know?  --UC_Bill (talk) 21:32, 4 September 2008 (UTC)


 * It might be interesting to find out the relative popularity of various date formats. It would be especially interesting to flag articles with a mixture of month-first and day-first. You might also confirm my suspicion that there will be many articles that have month-first or day-first in the text of the article but all-numeric in the references. --Gerry Ashton (talk) 21:45, 4 September 2008 (UTC)


 * I'll take care of this. I've downloaded the latest database dump file for enwiki, and am starting the processing right now.  It'll take a while to fine-tune the report, but here is a very rough breakdown after about ten minutes of analysis:


 * MDY_raw => 17806
 * MDY_linked => 8131
 * DMY_linked => 4590
 * DMY_raw => 1845

Those are just raw counts of dates in those two formats, both linked and unlinked, for however many articles my script has processed so far. The percentage breakdown should be fairly accurate, but that's about it. I'll get more over the next couple days, including per-article stats. --Sapphic (talk) 01:25, 5 September 2008 (UTC)

Date format statistics
I've generated some preliminary statistics on various date formats and how often they are used.

These numbers are generated from an analysis of about 11 million dates from within Wikipedia articles. I'll provide more details over the next few days, including the first of our "work lists" of articles that have a mix of conflicting date formats. --UC_Bill (talk) 22:30, 5 September 2008 (UTC)


 * Those statistics are somewhat off. I suspect you're having the same difficulty I am with eliminating dates in references (the "accessed on such-and-such date" mentions shouldn't count, right?) but here are my numbers:


 * Total pages analyzed (from 2008-07-24 database dump) &mdash; 7,278,279 (includes template pages and project pages, but not talk pages or user pages)
 * Total pages with dates of any kind (linked or unlinked) &mdash; 1,580,914 (21%)
 * Total pages with a mixture of date formats (linked or unlinked) &mdash; 245,746 (3% of total pages, 15% of pages with dates)
 * Total pages with only unlinked dates &mdash; 746,171 (10% of total pages, 47% of pages with dates)

Here are some per-article statistics:

Those statistics still aren't entirely accurate, because of the difficulty in excluding dates from within templates or references, which probably shouldn't be counted as part of the page. The regular expressions used for matching were exactly the same ones as in the DateFormatter.php code in the MediaWiki software, so they contain things like optional spacing and optional commas and such. I also included the linked MD and DM formats, since they're autoformatted independently of the presence of a linked year. As soon as I get the bugs worked out in the script so that it excludes references and template dates, then I'll generate some lists of articles with the "most mismatched" date formats (articles with more than one date format, sorted in descending order by the number of uses of the second-most-popular date format on the page.) At that time, I'll also make my script (PHP) available to anyone that wants to duplicate the results or check over my methodology. --Sapphic (talk) 23:47, 7 September 2008 (UTC)

Oh, and here are my numbers for the equivalent of UC Bill's table (% of date references rather than % of articles) --Sapphic (talk) 00:13, 8 September 2008 (UTC)


 * The example 2005-12-25 is used too many times; it looks like a minor error.


 * Dates in instances of templates that are also contained in the master template probably shouldn't count, but instances that are filled in by editors should be. Also, references should be included. If you wanted to also calculate the results for references separately, that might be interesting.


 * Finally, before drawing attention to these statistics, you might want to pick a more neutral example date. Not everyone celebrates Christmas. --Gerry Ashton (talk) 00:40, 8 September 2008 (UTC)


 * Sapphic used Christmas because her table was just a modified version of mine, and I picked that date arbitrarily. I'm not sure what you mean about it being an error; those aren't examples from the script, they're just an example of the format that I entered by hand.  I'll go back to using the current date, since it seems to bother you so much.  --UC_Bill (talk) 15:20, 8 September 2008 (UTC)


 * The error seems to be that for the example " 2005-12-25 " there are three different rows in the table; the "Format" entry for the three rows are Linked MD, Linked DM, and Linked ISO1. I think the examples for the Linked MD and Linked DM rows should be different. --Gerry Ashton (talk) 16:01, 8 September 2008 (UTC)


 * I see it, thanks. Fixed.  UC Bill was right, I just copied his table and added a few rows, but forgot to change the example.  My bad.  --Sapphic (talk) 00:30, 9 September 2008 (UTC)

Statistics discussion
My first reaction to the statistics is that there are so many unlinked dates already that the fuss about whether or not to automatically unlink the remaining dates is moot; it just wouldn't make much difference. Yet, the people who say they like autoformatting seem to be having a satisfying reading experience, so how can we reconcile the statistics with the experience of some reader/editors? --Gerry Ashton (talk) 13:58, 6 September 2008 (UTC)


 * I find these figures very hard to believe, particularly the proportion of unlinked vs linked dates, and international vs US unlinked dates. What kind of sample was used (I mean, what spread of article topics)? Is the comma-less US syntax included? A large proportion of US-format is linked without the comma (which works). Do you have figures on wrong syntax? Tony   (talk)  14:48, 6 September 2008 (UTC) PPS I presume that talk-page signature dates are excluded.  Tony   (talk)  14:50, 6 September 2008 (UTC)

I would like to thank UC_Bill for the statistics. It is the first time that I have seen anything like it. It is only when statistics are offered that they can be challenged or counter-offered. Lightmouse (talk) 18:04, 6 September 2008 (UTC)


 * I agree with Lightmouse. I hope UC Bill looks at things the way I do: discussion of one's work is in itself a compliment. --Gerry Ashton (talk) 18:09, 6 September 2008 (UTC)


 * My numbers aren't that different than UC Bill's, so I suspect we're using a similar methodology. Talk pages are not included, so the unlinked DMY formats are actually in articles, and are not from signatures (I know that to be true for my stats, and would assume the same for UC Bill's.)  A lot of them probably come from references though, or are passed into templates.  I'll clean those out of my stats and hopefully then can generate some lists of specific articles with whatever properties we like.  --Sapphic (talk) 00:18, 8 September 2008 (UTC)


 * The statistics are welcome, though like Tony I find them a little difficult to believe. My gut feel is that a much higher proporion of dates are linked than these numbers show. Also, they're missing one very important item - links to bare years, decades and centuries, which are almost always undesirable. The current far-from-intuitive autoformatting method has led to a belief among naive editors that all time periods should be linked; any replacement (if any) absolutely must have a syntax that can be understood without careful study of the MoS's description of how it works, or we'll just be perpetuating the current confusion and all the valueless links that result from it. I had to read it a few times to get the hang of it, and I'm a professional software designer. That's why I've come to the conclusion that just changing the current system to remove the wikilinking effect, although an improvement, is not the answer to the problem. It's an inherently confusing method and it needs radical reform or scrapping, not tinkering. Colonies Chris (talk) 08:12, 8 September 2008 (UTC)


 * No, actually a much lower percentage of dates are linked, because the excluded reference and template variables are mostly unlinked. A higher percentage of in-article dates are linked though, because neither Sapphic nor myself has managed to separate out all the reference/template uses, yet.  We're intentionally not including bare years, centuries, etc. because those dates are not subject to autoformatting (or even manual formatting, come to think of it.)  I guess maybe a better name for this project would have been "Dates Formatting WikiProject" because it's really all about date formatting (auto or manual) and not dates, per se.  But then again, maybe it should be about dates in general.  Anyway, I have some questions regarding the stats that I'll start in a new section below.  --UC_Bill (talk) 15:24, 8 September 2008 (UTC)

Desired statistics
What type of statistics are actually desired? We've been looking at the various types of date formats that are subject to autoformatting, or which would be subject to autoformatting if they were linked. Bare years and centuries, etc. that have only one element and are thus not technically capable of being formatted, have been excluded. Dates within references or passed as parameters to templates have also been excluded, but could be added back in and counted separately. We're also planning to produce lists of articles that are "most mismatched" in terms of their date formats. "Most mismatched" is being defined as the list of articles sorted in order of the number of dates in the "mismatch" format for that article (i.e. the second-most-popular date format, working on the assumption that the most popular date format for a given article is probably the format the entire article "should" be in.)

Does that pretty much sum up what we're looking for, so far? Please feel free to add suggestions/requests. --UC_Bill (talk) 15:32, 8 September 2008 (UTC)


 * I don't really care whether BC or BCE is used, but it might be interesting to look for articles that are inconsistent.


 * Looking through material from around 2003 at the WP:MOSNUM page, and at meta and mediawiki, I've come to the conclusion that autoformatting was consistently described as being the ISO 8601 format, and that there was an intent to apply it to years starting with AD 1. Yet, ISO 8601 format requires the Gregorian calendar, but autoformatting provides no date conversion capability. Furthermore, it isn't supposed to be used before 1583 without mutual agreement. So the autoformatting facility failed to meet its design objectives.


 * Then we come to the issue of how a reader will, or should, interpret a date such as 1500-01-01 in an article. The casual reader might just suppose the format conveys no information about which calendar, and consider it equivalent to 1 January 1500. The reader who isn't sure and checks recent reference works or who has done work with ISO 8601 as a software developer may think it has to be a Gregorian date. So it might be useful to identify articles that use the YYYY-MM-DD format or which link any date before 14 September 1752, the date the Gregorian calendar was adopted in England and the British Empire. --Gerry Ashton (talk) 16:18, 8 September 2008 (UTC)

I am re-running my processing script, taking into account most of the features mentioned above. It just completed one run but I found some errors (it's much harder to find dates that aren't linked &mdash; one more reason to leave them untouched for now!) and it takes about 40 hours to run, so expect some updated stats by the weekend. --Sapphic (talk) 02:34, 11 September 2008 (UTC)
 * But where are these statistics leading to? Convince me that any data you generate will be useful, and how. Tony   (talk)  15:17, 11 September 2008 (UTC)


 * The "most mismatched" articles list will be the first concrete result of this data mining. User:Sapphic has already created several useful lists (see WP:DBA) that are used in other WP maintenance tasks.  Working from a list of articles sorted in order of how "badly mismatched" they are in terms of date formats is much more effectively than randomly selecting articles.  Nobody here disputes that the mismatched formats need to be fixed, and producing such lists will help with that fixing.  --UC_Bill (talk) 15:33, 11 September 2008 (UTC)
 * By "mismatched", do you mean among the ~ 60% of general articles that are faulty? (See the pie graph. Doing them at random not only cleans out the DA, but date faults in nearly 2/3 of articles. FAs are much better, of course; GAs are a little better. Tony   (talk)  15:49, 11 September 2008 (UTC)


 * I don't think it's anywhere near 60% (not even that many articles have dates in them in the first place, including in the references) but yes, those are the articles I'm talking about. Working at random has the disadvantage of becoming increasingly difficult as you fix more articles, since you'll end up hitting the same articles you've already fixed with increasing frequency as time goes on.  It also doesn't scale as well as worklists do (each person can "claim" a particular section of a list and won't have to worry about duplicating the effort of others &mdash; this is how most WP maintenance tasks work, and those that don't use lists use categories, so one way or another people doing this kind of work generally avoid just picking articles at random.)  --UC_Bill (talk) 16:00, 11 September 2008 (UTC)


 * Minor issue, I think. This very wikiproject could be used to list who is to do what. When a substantial proportion of the low-hanging fruit in any topic area are done, the rest should follow as editors get it. Tony   (talk)  16:09, 11 September 2008 (UTC)

Statistics
Okay, the script has finished running and other than a few minor errors (about 37,000 pages out of 7.2 million in the XML dumpfile that had a blank first line were excluded, I've fixed the problem and they'll be in the next run) everything seems to check out. A spot check of some randomly selected articles give the same count for each date format as I find manually, so I'm fairly confident in these results. One minor miscount I found is that dates in raw ISO format sometimes appear in URLs in external links (example: News story ) so the count for unlinked ISO is artificially high. I'll need to do some sampling of the matches to figure out a good way to exclude them, and then the next run should be even more accurate.

I had some stats here, but I just found a mistake in the script that produces the tables from the data output by the first script (linked to above) so I've removed those tables. They should be on their own page anyway, for easier reference. --Sapphic (talk) 02:36, 13 September 2008 (UTC)
 * These statistics are for distinct topic areas, yes? That's the only way they could be remotely beneficial. Tony   (talk)  03:47, 13 September 2008 (UTC)


 * You're a riot. If you want a feature, you have to ask for it before the statistics are generated.  I'm sorry if you find that confusing, but it's just the way things work.

My hosting provider (where my servers are located) seems to be unreachable at the moment, but I'll be back later to continue adding more detailed statistics, and some lists (since I bet you're just dying to know which 680 articles have dates in raw YDM format in them.) --Sapphic (talk) 04:18, 13 September 2008 (UTC)
 * Please don't issue personal attacks ("you're a riot"). There have been continual attempts to bully me on the basis that my actions somehow interfere with the gathering of such statistics or pre-empt findings that will be based on them. When I ask what bearing the statistics will have on the DA issue, you tell me not to be "a riot", that I'm somehow asking for too much. Tony   (talk)  04:32, 13 September 2008 (UTC)


 * Whatever.


 * Ah! They came back online. And as it turns out, many of the 680 articles with YDM_raw dates in them are just strangely formatted date intervals, which are extremely difficult to parse properly in the absence of any date linking.  For example: 44th Fighter Squadron contains the string "18 November - 9 December 1973, 30 December 1973 - 20 January 1974, 10 February - 2 March 1974" which can be mistaken as having the YDM_raw dates  "1973, 30 December" and "1974, 10 February" embedded in it.  Also, I forgot to break out the pages in the Wikipedia, Template, and Image namespaces from the stats, so I'll do that now and revise the numbers.  --Sapphic (talk) 04:39, 13 September 2008 (UTC)
 * Whatever. Tony   (talk)  04:41, 13 September 2008 (UTC)

task: Develop a set of criteria for standard date formats for articles
It seems that page Wikipedia talk:Manual of Style (dates and numbers) is the place where this task is happening, so we probably shouldn't try to usurp it. Teemu Leisti (talk) 03:46, 20 September 2008 (UTC)
 * So, after weeks of arguing and polling, no consensus was found for any option. Do we want to continue the discussion here?  If not, does this project have a point anymore? Teemu Leisti (talk) 09:15, 21 September 2008 (UTC)


 * The consensus is to leave the MOSNUM as it is. There is much to be done, because many articles have inconsistent dates. A few articles have a date style that is not appropriate for the topic. --Gerry Ashton (talk) 18:02, 21 September 2008 (UTC)


 * The topic is irrelevant to the date format, except as a rough guess as to the majority readership. The only reason we even care about the topic is because we (currently) have no way of tailoring the date format to users that haven't specified a preference (i.e. most readers.)  I'm mentioning this because it would be silly to waste too much effort coming up with a comprehensive policy for date formats that hinged on the topic, since it's very likely that a better solution (that autoformats according to geolocation or other determiners) will be produced at some point.  It's probably still useful to have some kind of rules until then, but I think the focus should be more on the consistency of date formats rather than the appropriateness.  In other words, we should be more focused on those articles that have a broad mix of formats, rather than ones that have an "inappropriate" format but nonetheless apply that format consistently.  --UC_Bill (talk) 15:53, 9 October 2008 (UTC)

Horrible Histories
Hi. We're having a discussion on the fate of Horrible Histories TV show at: Horrible Histories (2009 TV series). As a relevant Wikiproject, we would greatly appreciate it if you would voice your opinion on the talk page, or to have a crack at editing and improving it. Thankyou for your time. :)--Coin945 (talk) 13:22, 30 June 2013 (UTC)

RFC: Connection between ISO 8601 standard and YYYY-MM-DD date format
An editor has been proposing to remove the restriction in Manual of Style/Dates and numbers that the YYYY-MM-DD date format is restricted to Gregorian calendar dates, with the year range limited from 1583 to 9999 inclusive. I don't believe this change should be made without proper consideration so I have opened an RFC: WT:MOSNUM. Jc3s5h (talk) 23:34, 25 January 2014 (UTC)

Leaflet for Wikiproject Dates at Wikimania 2014
Hi all,

My name is Adi Khajuria and I am helping out with Wikimania 2014 in London.

One of our initiatives is to create leaflets to increase the discoverability of various wikimedia projects, and showcase the breadth of activity within wikimedia. Any kind of project can have a physical paper leaflet designed - for free - as a tool to help recruit new contributors. These leaflets will be printed at Wikimania 2014, and the designs can be re-used in the future at other events and locations.

This is particularly aimed at highlighting less discoverable but successful projects, e.g:

• Active Wikiprojects: Wikiproject Medicine, WikiProject Video Games, Wikiproject Film

• Tech projects/Tools, which may be looking for either users or developers.

• Less known major projects: Wikinews, Wikidata, Wikivoyage, etc.

• Wiki Loves Parliaments, Wiki Loves Monuments, Wiki Loves ____

• Wikimedia thematic organisations, Wikiwomen’s Collaborative, The Signpost The deadline for submissions is 1st July 2014 For more information or to sign up for one for your project, go to: Project leaflets Adikhajuria (talk) 12:08, 27 June 2014 (UTC)

Comment on the WikiProject X proposal
Hello there! As you may already know, most WikiProjects here on Wikipedia struggle to stay active after they've been founded. I believe there is a lot of potential for WikiProjects to facilitate collaboration across subject areas, so I have submitted a grant proposal with the Wikimedia Foundation for the "WikiProject X" project. WikiProject X will study what makes WikiProjects succeed in retaining editors and then design a prototype WikiProject system that will recruit contributors to WikiProjects and help them run effectively. Please review the proposal here and leave feedback. If you have any questions, you can ask on the proposal page or leave a message on my talk page. Thank you for your time! (Also, sorry about the posting mistake earlier. If someone already moved my message to the talk page, feel free to remove this posting.) Harej (talk) 22:47, 1 October 2014 (UTC)

WikiProject X is live!


Hello everyone!

You may have received a message from me earlier asking you to comment on my WikiProject X proposal. The good news is that WikiProject X is now live! In our first phase, we are focusing on research. At this time, we are looking for people to share their experiences with WikiProjects: good, bad, or neutral. We are also looking for WikiProjects that may be interested in trying out new tools and layouts that will make participating easier and projects easier to maintain. If you or your WikiProject are interested, check us out! Note that this is an opt-in program; no WikiProject will be required to change anything against its wishes. Please let me know if you have any questions. Thank you!

Note: To receive additional notifications about WikiProject X on this talk page, please add this page to WikiProject X/Newsletter. Otherwise, this will be the last notification sent about WikiProject X.

Harej (talk) 16:57, 14 January 2015 (UTC)

Proposal to merge this project into WikiProject History
See Wikipedia talk:WikiProject History. –&#8239;Joe (talk) 09:24, 14 May 2024 (UTC)