Talk:Wikimedia News/archive 2

From Meta, a Wikimedia project coordination wiki
Jump to navigation Jump to search
Archive 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.


Report project closures as well?

Lately, some projects have been closed: Kanuri Wikipedia, Low Saxon Wikiquote, Old English Wikisource. Shouldn't this be reported as well? --Johannes Rohr 07:22, 31 May 2007 (UTC)

I think that would be a good idea. Information on what projects no longer exist has the possibility to be helpful, and wouldn't take much work to write a little blurb about, so I see no reason why it shouldn't be included. I suppose the format would be along the lines of
"* The Project X has been closed down due to reasons Y. See [[Proposals for closing projects/Closure of Project X]] for more information."
Picaroon (Talk) 22:42, 31 May 2007 (UTC)
Supporting the above views. --Purodha Blissenbach 18:55, 28 March 2008 (UTC)

Too frequently update?

Recent reports contain "31000 articles", "26000 articles" ... etc and the page become thin. The recommendation is "1k, 2k, 5k increments to 20k". How do you think? We are better to weed out those minor milestones? Or keep them? --Aphaia 08:12, 27 July 2007 (UTC)

I think we should remove them - the page is kind of useless if every thousand is reported. — Timichal 08:45, 27 July 2007 (UTC)
I think we should keep them because a milestone shows a progress a one project, minor or major. A-yao 14:37, 28 March 2008 (UTC)

What the Wu?

What the heck is going on over at Wu? Pages for temperatures in 1°C increments?? - dcljr 05:46, 17 February 2008 (UTC)

Rename this page?

Since this page only deals with milestones now, rather than general news, I think it should be renamed to Milestones or something similar. At the moment the name is confusing and overlaps with Goings-on. The sidebar link should obviously be changed too. I'm not sure what to do about the archives though, some of the older ones do contain news other than milestones. What do other people think? the wub "?!" 13:14, 4 August 2008 (UTC)

Actually, given that Milestones exists already, what about Recent milestones? the wub "?!" 07:27, 6 August 2008 (UTC)

I think that maybe a merge is required. Having both "Recent milestones" and "Milestones" seems superfluous. --Anonymous DissidentTalk 08:55, 6 August 2008 (UTC)
Well, if we were to move, yes, we should certainly merge the pair.
As to whether to move this page in the first place, well. This page is for news about the various Wikimedia projects; we didn't intend originally for that news to be purely for "milestones". Maybe we might want to add other items in (e.g. like for the Commons section, which talks about the press releases too)?
James F. (talk) 19:34, 17 August 2008 (UTC)
That's what I would expect from the current title, but how is it different from Goings-on? the wub "?!" 08:43, 19 August 2008 (UTC)
Goings-on includes WMF news as well as WM news, I guess. But yes, there's a whole lot of over-lap. What do we want to do about them?
James F. (talk) 22:31, 19 August 2008 (UTC)

Maybe just merge Wikimedia News with Milestones and leave Goings-on by itself? Goings-on is a bit different from the other 2, which are quite similar to each other. Cirt 05:47, 20 August 2008 (UTC)

Cirt took the words out of my mouth :-) —Giggy 07:32, 20 August 2008 (UTC)
Fine by me. Should we action?
James F. (talk) 21:49, 20 August 2008 (UTC)
Milestones hasn't been updated very well, and this page looks to have been designed better. Maybe merge the data from Milestones to here instead? —Giggy 03:43, 21 August 2008 (UTC)

See the most recent comment is Aug 2008! Abiy lost here is this Meta or Betya Wiki?How about updating ?Great anyways fast connections useing thjisAndrevictor 00:09, 9 January 2010 (UTC)(4PMFriJan8210mytimePST@1stcnt.decded!

Sorani article explosion

The Sorani Wikipedia has jumped to over 1,500 articles recently. Because I periodically collect article-count statistics from the various wikis, I can say that the wiki's "raw" statistics was reporting 791 articles on Jan 23 at 20:07 UTC and then 1,492 articles 12 hours later on Jan 24 at 07:58 UTC. The count is now 1,523 (Feb 12 at 03:43). Initially ckb:Special:Allpages didn't seem to be showing nearly as many articles as the stats page count, but now I think it is (rough estimate based on how many pages of results there are). OTOH, ckb:Special:Newpages still doesn't show a large amount of activity over the same time period, so I can't explain why the number of articles has more than doubled. Does anyone else know anything about this? - dcljr 03:58, 2 February 2010 (UTC)

How about this conversation [1] (bottom of page)? Apparently, imported articles have been there for a while w/o having been included in the statistics... (up to now) Seb az86556 04:41, 2 February 2010 (UTC)

Aro-mania

The Aromanian Wikipedia has gone completely crazy. Bot activity in the last 24 hours, for example, just added almost 4,000 articles on (mostly) obscure medical conditions with no actual content on them except two links to "index" pages that listed only the titles of the articles that were being added! In other words, no real content was added to the wiki at all, just a bunch of titles. Is this what the other 45,000 articles that have been added in the last 3 months are like? - dcljr 06:57, 5 July 2010 (UTC)

To answer my own question, yes, it appears so. Hitting the "random page" link 30 times over there turned up 27 articles looking very much as I described above and 3 that consisted of one brief sentence. - dcljr 07:22, 5 July 2010 (UTC)
It's a fact that random gives nearly always a blank article with only titles and a link. I think this kind of edtis are worse for Wikipedia. Also happens with waray-waray language robot edition. Maybe they should be stopped. -Theklan 01:17, 8 January 2011 (UTC)

'book modules'

By that do you count full books or including subpages? Kayau WP WB WN 07:42, 26 July 2010 (UTC)

Well I don't know,I also not have nothing the look with it. @KnuxD 13:43, 31 August 2010 (UTC)

Georgian 40k milestone date

I'm using an analysis tool, Qlikview, to analyze growth rates of the language wikipedias. The program can parse out the data from the list of wikipedias as well as this page and hopefully merge the two. The end result would be a continual log of all of the datapoints on the list of wikis with the ability to slice the data a bunch of ways and calculate growth rates over varying periods. This page acts as a great set of datapoints in that I can get point-in-time values in the past where we knew what the article count was. I have only one sticking point, the milestone date for the Georgian wikipedia. Does anyone have a more refined date for the 40,000 milestone? I can always fudge unknown dates to the beginning of the month, but when we don't know the month either it's a little more questionable. Between January and now, Georgian grew at a rate of about 28 articles/day. In the past few days it grew at a rate of about 15 articles/day. That puts the estimated date between March and May. Any thoughts? Aremisasling 15:41, 31 August 2010 (UTC)

Nevermind. I found it on Wikipedia Statistics. I didn't find a specific date, but it hit the mark in early May. Aremisasling 15:54, 31 August 2010 (UTC)

French Wikipedia 1-million date

I'm curious which time zone the report that French passed 1 million on Sep 21st is based on. These three pages at the French Wikipedia (pointed out by en:User:Adam78, BTW) seem to corroborate this. However, I checked the raw statistics given by http://fr.wikipedia.org/w/index.php?title=Sp%C3%A9cial:Statistiques&action=raw at the following times, UTC, and none of them showed French having reached 1 million (I don't have a record of what the counts actually were, however):

  • Mon Sep 20 08:44:01
  • Tue Sep 21 18:50:23
  • Wed Sep 22 07:37:18

I always try to use UTC for dates that I report here. What time zone was used for this French milestone? - dcljr 05:28, 25 September 2010 (UTC)

Hello,
There was a bug with the count of articles. This bug was fixed the 23rd, during the sprint for the million. For the reason the counter reached 1,003,000. Then a developper looked into the database, a saw that the millionth was - at this time - fr:Louis Babel. That's how the millionth was decided. So we now the date of was the 21st at 06:54 (CEST). We use the time of France.
Regards
--Hercule 20:26, 25 September 2010 (UTC)

Colors

For years now, the table of milestones (at the bottom of the page) for Wikipedia has used a different color scheme from those for all the other projects. I just made an edit (which I then immediately reverted) in which I changed the Wikipedia table to more closely match the colors in the similar table at en:Wikipedia:Milestone statistics. (Full disclosure: the color scheme there was my idea. This one here had to be a bit more complicated to account for the entries for multiples of 1.5, 3, 4, 6, 7, 8, and 9, which aren't used in the other table; I decided to just use linear interpolation between the colors for the multiples of 1, 2, and 5, which are exactly those used in the other table except they're all "shifted down" since this table starts at 100 instead of 1000. It sounds more complicated than it really is...) Anyway, what do people think about using this color scheme (or something similar) for all the tables? Do the additional levels just make the colors too distracting? Can anyone suggest a better color scheme? - dcljr 12:28, 19 December 2010 (UTC)

Fair warning: If no one says anything soon, I'm going to make the change. - dcljr 02:00, 21 December 2010 (UTC)
be bold :) Przykuta 08:15, 21 December 2010 (UTC)
I have made the changes for the tables up to Wikisource, which has an "Updated" column, unlike the others. Most of the rest of the tables have a little different selection of milestones, so I have to think about what I want to do with them... - dcljr 08:10, 11 February 2011 (UTC)

Stats in MW 1.17

For those who were using the "action=raw" URL option to track wiki statistics, as of MediaWiki 1.17 that option is deprecated. You now have to use an API query such as:

http://meta.wikimedia.org/w/api.php?action=query&meta=siteinfo&siprop=statistics

See the helpful links at the above URL for other options. For example, if you append "&format=xml" to the end, you get just the XML output without the HTML wrapper. Thanks to User:Hydriz for this info. - dcljr 23:29, 17 February 2011 (UTC)

Which stats and which milestones

After tracking just Wikipedia article-count milestones for a few years now, and then adding Wiktionary entry-count milestones for the past few weeks, I finally have a general-purpose script that can track any desired milestones for any stats returned by the type of API query discussed in the previous section.

My question is, which milestones should be announced here? For the last couple of days, I've only announced the ones that seemed to be the "most important", but today I posted all 8 milestones detected by my script. This seems to be a pretty typical daily amount (based on what I've seen over the past few days), and it seems excessive to me.

To cut back on the number of announcements going forward (although it wouldn't have affected the number posted today), I propose that only the "1-", "2-", and "5-" milestones be announced for the stats other than articles (this includes pages, edits, users, activeusers, admins, and images, but not jobs, which is generally uninteresting). Or maybe I should cut some of these stats back to only powers of 10 (like maybe users and edits). Opinions? - dcljr 07:15, 21 February 2011 (UTC)

I think that users, admins, and edits milestones are only important when they are powers of ten and considerably large, and that pages (not content pages) milestones aren't important at all. Also, I don't think the "dropped below" parts are important enough to list. --Yair rand 07:56, 21 February 2011 (UTC)
For the moment, I've settled on powers of 10 for pages, edits and users; "1-/2-/5-" milestones (which I like to call "canonical") for images and admins (keep in mind, this covers "all" Wikimedia wikis, not just the Wikipedias); and all currently used milestones for articles (although, if I had my way, I'd make those canonical, too ;). Because it's so variable, I have stopped tracking activeusers (first time I included it, my script found 23 milestones for that one stat, in only one day!). I never entertained the idea of tracking jobs (although it might still be worth checking for unusually large values of it). Once I start tracking activeusers in a reasonable way, I may drop users entirely (look through Special:ListUsers on a typical wiki and note how few look "legitimate" — I guess the same could be said for edits, but that stat seems a little more informative). - dcljr 04:15, 26 February 2011 (UTC)
Ever since CentralAuth was implemented, the user count has been pretty much useless. Unfortunately, there doesn't seem to be a way of determining how many users have a particular wiki set as their home wiki. – Minh Nguyễn (talk, contribs) 08:38, 11 March 2011 (UTC)
FYI, to anyone who cares: I've modified my script to stop announcing wikis that just bounce up and down repeatedly around a single milestone (actually, it just reports them differently so I can decide myself whether it's worth posting). This should cut down on the number of "useless" milestone announcements. - dcljr 03:37, 19 March 2011 (UTC)

Using Wikipedia Statistics to fill in gaps

As there are a lot of unknown milestone dates, I've determined a possible way for filling them in. Is there a reason why we couldn't use the below pages to at least target the month and year a milestone was reached? They are listed in the referring pages as "Article Count (official)"

[2] - wiktionary [3] - wikipedia [4] - wikiquote

There are corresponding pages for the other projects as well.

Aremisasling 18:05, 14 March 2011 (UTC)

A good idea, but unfortunately it looks like the figures in those tables are being rounded (sometimes up) to the nearest tenths decimal place, since the table for the Wiktionaries lists French at "2.0M" in "Feb 2011", whereas the French Wiktionary is not yet to 2,000,000 entries as of the time I'm posting this. Bummer... Still, if used properly, the tables can probably be used to get the date to the nearest couple of months, anyway, in most cases. - dcljr 02:07, 1 April 2011 (UTC)
I've been using some of these "official" article-count pages (linked to above by Aremisasling) to give date ranges for some of the entries in the tables. As I've been going through them, every once in a while I'll come across a language that has wildly different article counts compared to the "s23.org counts" given in the various "/Table" pages here at Meta. For example, the Wikisource Statistics page at stats.wikimedia.org lists the Chinese article counts so far this year (through March, anyway) at around "55k", but a quick glance through the page history of Wikisource/Table reveals that the Chinese article counts have been over 102,000 during the same time period. Is it just a coincidence that the amount of error is approximately a factor of 2? (Chinese text requires 2 bytes per character—but how could the article count be affected by that??) I'm confused! - dcljr 07:05, 28 April 2011 (UTC)
Wikistats uses xml dumps, and builds all stats fresh on each run. So for wikistats deleted articles have never existed, it relates history of current content. For most wikis it counts as articles: pages in namespace 0, with internal link on the page, which are not redirect. For some wikis, like wikisource, a few other namespaces are considered so crucial they are now included into article count. These are for wikisource 102/104/106, for commons 6/14 for strategy 106. Erik Zachte 13:00, 28 April 2011 (UTC)
For the benefit of anyone stumbling across this discussion who doesn't know how to look them up, these namespaces are:
  • Wikisource: 102 = Author, 104 = Page, 106 = Index
  • Commons: 6 = File, 14 = Category
  • Strategy: 106 = Proposal
Note that you can see what namespace corresponds to what number by adding, for example, "{{ns:106}}" to a page on the wiki of interest (e.g., the Sandbox, or any page if you just use "Show preview" instead of "Save changes"). - dcljr (talk) 11:00, 11 March 2012 (UTC)
Also note the an API query such as this one (run on the wiki you're interested in) will show you which namespaces count as "content" (search for that string in the results), and thus contribute to the "article count". - dcljr (talk) 05:24, 3 April 2012 (UTC)

Bot-dominated wikis

The Malagasy Wiktionary is but the latest example of a tiny wiki suddenly rivaling the largest projects in only one metric: article count. This time, ‎Bot-Jagwar has mostly copied entries verbatim from other wikis (see the majority of this list). The same bot has also been creating entries for declensions or conjugations of words that aren't defined anywhere in the wiki.

While plenty of other Wiktionaries have benefited from bot-generated stubs in the past, these wikis have always coupled the stubs with imports from real dictionaries. But with these nonsensical stubs alone, who can seriously argue that the new entries inform visitors, or that they will encourage a commensurate number of volunteers to improve the wiki? These are the standards that bot operators must meet when applying for bot flags at well-established wikis. Unfortunately, when the Malagasy Wiktionary rises to the top of Wiktionary's front page, its dearth of useful content reflects very poorly on the entire project, belittling the hard work of countless non-programmers. As a French Wiktionary contributor has noted, the mass imports also lack attribution and thus infringe on the original authors' copyrights.

The multilingual portals and pages like Wikimedia News offer a strong incentive for bot operators to game the wiki rankings, as we saw with several Wikipedias long ago. As a first step, to prevent Wikimedia News from being dominated by one wiki, I propose noting only power-of-ten milestones for fast-growing, bot-dominated wikis like the Malagasy Wiktionary. We should make sure power-of-ten milestones are accompanied by an explanation of how the wiki managed to add the latest spate of articles.

Thoughts?

 – Minh Nguyễn (talk, contribs) 07:33, 16 March 2011 (UTC)

Removed ak

I have taken the liberty of removing Akan (ak:) from the Wikipedia table because it fell sharply from 119 articles to 68 in one day, back on 3 Apr 2011, and has only gotten to 74 articles as of the time I'm posting this. This fact indicates to me that the 100+ milestone was most likely not based on "legitimate" content, so the wiki shouldn't be listed in the 100 row of the table. - dcljr 08:18, 12 July 2011 (UTC)

Wikimedia Argentina

The opening of "the Argentine and Mexican Wikimedia chapter wikis" was just announced here. What is the relationship between the first wiki just mentioned and "Wikimedia Argentina", which is linked to from m:Wikimedia coordination and other projects? - dcljr 01:24, 15 July 2011 (UTC)

Malagasy Wiktionary moved to 500K in table

I have taken the liberty of moving wikt:mg: to the 500,000 row of the table, since its rise to over 1 million was based entirely on "controversial" bot-created stubs that have now been deleted. - dcljr 13:34, 18 July 2011 (UTC)

:it Wikipedia

Shouldn't the current situation at :it (just visit their mainpage) be mentioned on the Wikimedia News page? --Túrelio 18:12, 4 October 2011 (UTC)

article-count changes

Can anyone offer an explanation for the jumps? 1.18? Seb az86556 01:02, 8 October 2011 (UTC)

This is referring to the Oct 6th changes, BTW. - dcljr 01:57, 8 October 2011 (UTC)

Minor question about chronology

Since I often check the wiki stats and update this page a little after midnight UTC, sometimes a news item will have to be added to the "next" day's entry (literally, I guess I'm referring to the "current" day's entry). Then when I update it again (approximately 24 hours later), I might have more entries to put under the same day. When this happens, I add the entries at the bottom of the list. Other people, I have noticed, add new entries at the top. The latter way would match the reverse-chronological nature of the days themselves, but for some reason I prefer doing it the other way. Does anyone have a strong opinion (but hopefully not too strong ;) about which way is "best"? - dcljr 00:14, 17 October 2011 (UTC)

Loud and clear... [g] - dcljr 04:16, 6 December 2011 (UTC)

Neapolitan WP

Someone just tried to demote the Neapolitan Wikipedia to 10K in the Wikipedias table, except they didn't change the date (from the one when it reached 40K), so I reverted their edit. Apparently, in early August 2011, User:Vituzzu deleted a bunch of articles from that wiki, saying, "pagina praticamente vuota e in italiano" (which Google Translate informs me means, "virtually empty page in Italian"). So I've re-demoted it and restored the date that it first reached 10K, 20 June 2006. - dcljr 04:36, 8 November 2011 (UTC)

Standardizing interwiki links in archives

In the older archives of this page, many of the interwiki links are in "external-link" style (e.g., [http://ro.wikipedia.org Romanian Wikipedia]). Does anyone object to me changing these to "internal-link" style ([[:ro:|Romanian Wikipedia]]), as we prefer to use nowadays? - dcljr 00:55, 3 January 2012 (UTC)

A news

We r happy to announce Movement. We love movement cause reality. We love to be unedited. We r editors. We r in movement. We love movement. The vision has created. Now we r in vision. We can change the course of the humanity. The vision has created. One day fast of the 18ht Jan 2012 should going to be a milestone in the history of the Wikimedia Movement. —The preceding unsigned comment was added by Ansha (talk) 06:27, 18 January 2012 (UTC)

Emphasis on article milestones?

Since we're getting a lot more non-article-count announcements on this page than we used to (largely my fault), it's getting harder to pick out the article-count announcements (a.k.a. "entries" for Wiktionary, "book modules" for Wikibooks, etc.), I'm wondering if we shouldn't start emphasizing those announcements with, say, bold text on the article counts (which kind of makes sense, given that the introductory text at the top of the page uses bold text on the phrase "article-count milestones"). Interested parties can compare the current version of the page without bold article counts and with bold article counts (but only on announcements back to Januray 1st, since I had to add the bold markup "manually" and got tired by the time I reached that date — oh, and I included a file-count announcement for Commons, even though technically those aren't "articles" on that wiki). Comments? Objections? Should we use bold text on article counts from now on? - dcljr (talk) 05:14, 13 April 2012 (UTC)

I would support this proposal. Additionally, we could emphasize the names of "large" Wikipedias (having more than, say, 100k articles). - Ace111 (talk) 23:08, 13 April 2012 (UTC)
Very good idea. - Iketsi (talk) 16:33, 15 April 2012 (UTC)
OK, I'll make the change soon... - dcljr (talk) 03:19, 18 April 2012 (UTC)

┌─────────────────────────────────┘
Done. To be clear on the criteria I was using, I decided to use bold only on article counts for individual content wikis:

  • Articles include Wikipedia and Wikinews "articles"; Wiktionary "entries"; Wikibooks "book modules"; Wikiquote, Wikispecies, and Incubator "content pages"; Wikisource "text units"; Wikiversity "learning modules"; and Commons "gallery pages". So, not "total pages", "page edits", etc. The only exception to this rule is for "media files" on Commons (as mentioned in my opening comment above).
  • Individual means not groups of wikis, such as "Wikipedias in all languages" or "Slavic Wikipedias (20 languages)", which have gotten announcements here in the past. I have nothing against such announcements, I just felt that they shouldn't be emphasized the same as individual wikis.
  • Content wikis include all the projects just mentioned, but not Meta or things like the Wikimedia Foundation wiki, the various Wikimania wikis, and so forth.

This is just what I've decided to do. Others might have different ideas... Ace111, I have not emphasized anything else because that might dilute the benefit of adding the emphasis in the first place. - dcljr (talk) 05:54, 21 April 2012 (UTC)

May 10 article count updates

On May 10th a bug report requesting that the updateArticleCount.php script be run on all Wiktionaries and Wikisources was acted upon, resulting in 58 of those wikis surpassing or falling below one (or more) of the article-count milestones tracked on this page (not counting the new wikis whose milestones were already announced on May 10th). Some of the changes were quite large and therefore questionable.

Here's a summary table (initial order of the rows is by the percent change). Note that for each wiki the article count that's closest to my estimated count is highlighed.

Proj. Wiki May 9
count1
Change May 10
count
Percent
change
May 9
level2
May 10
level
Content
namespaces
May 10
activity3
Estimated
articles4
WT Nepali Wiktionary 4,821 −4,748 73 −98% 2,000 0 ns0 none 28/30 * 4,937 = 4,608 est. (~93% linked) [2012-05-11]5
WT Pashto Wiktionary 4,082 −3,882 200 −95% 2,000 200 ns0 +4 22/30 * ~7,638 = ~5,601 est. (~73% linked) [2012-05-13]
WT Tsonga Wiktionary 363 −344 19 −95% 200 0 ns0 none 20/30 * 345 = 230 est. (~67% linked) [2012-05-12]
WT Hindi Wiktionary 105,353 −99,173 6,180 −94% 100,000 5,000 ns0
WT Tigrinya Wiktionary 784 −673 111 −86% 500 100 ns0 none 10/30 * 561 = 187 est. (~33% linked) [2012-05-13]
WS Slovak Wikisource 229 −194 35 −85% 200 0 ns0 none 28/30 * 230 = 215 est. (~93% linked) [2012-05-12]
WS Azerbaijani Wikisource 2,381 −1,999 382 −84% 2,000 200 ns0 30/30 * 2,577 = 2,577 est. (~100 linked) [2012-05-12]
WT Kyrgyz Wiktionary 3,375 −2,516 859 −75% 2,000 500 ns0
WS Bosnian Wikisource 1,643 −1,214 429 −74% 1,000 200 ns0 none 30/30 * 1,687 = 1,687 est. (~100% linked) [2012-05-12]
WT Bengali Wiktionary 713 −477 236 −67% 500 200 ns0 none 28/30 * 802 = 749 est. (~93% linked) [2012-05-13]
WS Thai Wikisource 13,599 −8,548 5,051 −63% 10,000 5,000 ns0 none 16/30 * 11,101 = 5,921 est. (~53% linked) [2012-05-12]
WS Ukrainian Wikisource 4,563 −2,616 1,947 −57% 2,000 1,000 ns0 +1 28/30 * 4,757 = 4,440 est. (~93% linked) [2012-05-11]
WT Telugu Wiktionary 49,812 −26,739 23,073 −54% 40,000 20,000 ns0
WT Amharic Wiktionary 373 −200 173 −54% 200 100 ns0 none 19/30 * 398 = 252 est. (~63% linked) [2012-05-13]
WT Cherokee Wiktionary 376 −198 178 −53% 200 100 ns0 none 18/30 * 302 = 181 est. (~60% linked) [2012-05-13]
WS Breton Wikisource 3,535 −1,654 1,881 −47% 2,000 1,000 ns0, 100 [Index], 102 [Page], 104 [Author] ns102: +1 ns0: 30/30 * 1,734 = 1,734 est. (~100% linked); ns100: 71 (100% linked); ns102: 0/30 + ~11,175 = 0 est. (~0% linked?); ns104: 62 (79% linked) [2012-05-13]
WT Divehi Wiktionary 138 −59 79 −43% 100 0 ns0 none 14/30 * 172 = 80 est. (~47% linked) [2012-05-13]
WS Turkish Wikisource 5,156 −2,166 2,990 −42% 5,000 2,000 ns0, 100 [Author] none [2012-05-12]
WS Estonian Wikisource 717 −282 435 −39% 500 200 ns0, 102 [Page], 104 [Index], 106 [Author] ns0: +2; ns106: +2 ns0: 26/30 * 720 = 624 est. (~87% linked); ns102: 0/30 * 321 = 0 est. (~0% linked); ns104: 7 of 8 = 7 (88% linked); ns106: 17 of 18 = 17 (94% linked) [2012-05-14]
WT Lao Wiktionary 60,674 −23,072 37,602 −38% 60,000 30,000 ns0
WS Welsh Wikisource 271 −97 174 −36% 200 100 ns0 none 25/30 * 337 = 281 est. (~83% linked) [2012-05-13]
WS Limburgian Wikisource 2,336 −811 1,525 −35% 2,000 1,000 ns0
WS Macedonian Wikisource 2,734 −888 1,846 −32% 2,000 1,000 ns0
WT Chinese Wiktionary 1,202,474 −376,510 825,964 −31% 1,000,000 800,000 ns0 +5 29/30 * ~1,234,526 = ~1,193,375 est. (~97% linked) [2012-05-13]
WT Persian Wiktionary 78,864 −20,744 58,120 −26% 70,000 50,000 ns0
WT Sundanese Wiktionary 238 −56 182 −24% 200 100 ns0
WS Dutch Wikisource 5,524 −1,298 4,226 −23% 5,000 2,000 ns0, 102 [Author] +6 [2012-05-12]
WT Greenlandic Wiktionary 1,067 −230 837 −22% 1,000 500 ns0
WT Gujarati Wiktionary 518 −107 411 −21% 500 200 ns0
WT Macedonian Wiktionary 2,304 −333 1,971 −14% 2,000 1,000 ns0
WT Sanskrit Wiktionary 219 −31 188 −14% 200 100 ns0
WT Sindhi Wiktionary 1,073 −120 953 −11% 1,000 500 ns0
WT Oriya Wiktionary 103 −10 93 −10% 100 0 ns0
WT Tswana Wiktionary 100 −6 94 −6% 100 0 ns0
WS Arabic Wikisource 20,844 −1,186 19,658 −6% 20,000 15,000 ns0, 102 [Author], 104 [Page]
WT Samoan Wiktionary 102 −3 99 −3% 100 0 ns0
WT Somali Wiktionary 203 −4 199 −2% 200 100 ns0
WT Afrikaans Wiktionary 14,993 +338 15,331 +2% 10,000 15,000 ns0
WT Luxembourgish Wiktionary 4,971 +150 5,121 +3% 2,000 5,000 ns0
WS Hungarian Wikisource 19,293 +835 20,128 +4% 15,000 20,000 ns0, 100 [Author], 104 [Page], 106 [Index]
WT Turkish Wiktionary 283,030 +18,640 301,670 +7% 200,000 300,000 ns0
WS Portuguese Wikisource 87,931 +7,478 95,409 +9% 80,000 90,000 ns0, 102 [Author], 104 [Index], 106 [Page]
WT Sinhalese Wiktionary 935 +131 1,066 +14% 500 1,000 ns0
WT Russian Wiktionary 327,449 +109,838 437,287 +34% 300,000 400,000 ns0
WT Malay Wiktionary 1,465 +630 2,095 +43% 1,000 2,000 ns0
WS Spanish Wikisource 52,159 +25,071 77,230 +48% 50,000 70,000 ns0, 102 [Page], 104 [Index]
WT Greek Wiktionary 191,251 +99,440 290,691 +52% 150,000 200,000 ns0
WT Interlingue Wiktionary 184 +99 283 +54% 100 200 ns0 −1 25/30 * 328 = 273 est. (~83% linked) [2012-05-12]
WS Venetian Wikisource 1,974 +1,327 3,301 +67% 1,000 2,000 ns0, 100 [Author], 102 [Page], 104 [Index] none ns0: 21/30 * 2,836 = 1,985 est. (~70% linked); ns100: 9/30 * 126 = 38 est. (~30% linked); ns102: 3/30 * 4,902 = 490 est. (~10% linked); ns104: 61 of 62 = 61 (98% linked) [2012-05-18]
WS Italian Wikisource 67,136 +48,308 115,444 +72% 60,000 100,000 ns0, 102 [Author], 108 [Page], 110 [Index]
WS English Wikisource 369,323 +398,350 767,673 +108% 300,000 700,000 ns0, 102 [Author], 104 [Page], 106 [Index] ns0: +84, −10; ns102: +4; ns104: +528, −6; ns106: +5, −1 ns0: 28/30 * ~256,071 = ~239,000 est. (~93% linked); ns102: 29/30 * ~12,947 = ~12,515 est. (~97% linked); ns104: 3/30 * ~732,203 = ~73,220 est. (10% linked); ns106: 30/30 * ~5,017 = ~5,017 est. (~100% linked) [2012-05-13]
WS Polish Wikisource 35,682 +47,828 83,510 +134% 30,000 80,000 ns0, 100 [Page], 102 [Index], 104 [Author]
WS Norwegian (Bokmål) Wikisource 1,667 +2,478 4,145 +149% 1,000 2,000 ns0, 102 [Author], 104 [Page], 106 [Index] ns0: +2; ns104: +8; ns106: +2 ns0: 18/30 * 2,456 = 1,474 est. (~60% linked); ns102: 25/30 * 223 = 186 est. (~83% linked); ns104: 1/30 * ~38,389 = ~1,280 est. (~3% linked); ns106: 24/30 * 612 = 490 est. (~80% linked) [2012-05-17]
WS Swedish Wikisource 17,732 +29,048 46,780 +164% 15,000 40,000 ns0, 104 [Page], 106 [Author], 108 [Index]
WS Hebrew Wikisource 34,144 +57,112 91,256 +167% 30,000 90,000 ns0, 104 [Page], 108 [Author], 112 [Index]
WS Catalan Wikisource 5,707 +10,754 16,461 +188% 5,000 15,000 ns0, 102 [Page], 104 [Index], 106 [Author] ns102: +1 ns0: 20/30 * 2,780 = 1,853 est. (~67% linked); ns102: 5/30 * 13,013 = 2,169 est. (~17% linked); ns104: 30/30 * 350 = 350 est. (~100% linked); ns106: 30/30 * 385 = 385 est. (~100% linked) [2012-05-17]
WS German Wikisource 86,340 +166,943 253,283 +193% 80,000 200,000 ns0, 102 [Page], 104 [Index]
WS French Wikisource 281,166 +819,297 1,100,463 +291% 200,000 1,000,000 ns0, 102 [Author], 104 [Page], 112 [Book] ns0: +~2,000, −3; ns102: +1; ns104: +~250; ns112: +7 ns0: 26/30 * ~89,424 = ~77,501 est. (~87% linked); ns102: 27/30 * ~5020 = ~4,518 est. (~90% linked); ns104: 0 * ~1,061,910 = 0 est. (~0% linked?); ns112: 30/30 * ~8,451 = ~8,451 est. (~100% linked) [2012-05-12]
Note 1: This and the count for May 10 are based on what the API-query equivalent of {{NUMBEROFARTICLES}} returned.
Note 2: Milestone table row the wiki belongs in (assuming correct article count); level 0 means not in the table because it's below 100 content pages
Note 3: Any article creation or deletion activity on May 10, as seen in Special:NewPages (not checked for links) or Special:Log
Note 4: Estimated article (non-redirect content page with at least one [[link]]) counts based on check of 30 Special:Random articles and census (or ~estimated count) of Special:Allpages; in rare cases (small wikis or small namespaces on larger wikis), a complete census of articles (non-redirects with links) was taken (these lack "est.")
Note 5: Date actually checked

I'm putting this here mainly for my own benefit, as I investigate these changes to see whether they're actually reflecting correct article counts. If I find out there's nothing wrong with the new counts (and hence the new milestones), I'll add the info to the May 10th announcements and, I guess, change the Wiktionary and Wikisource milestone tables to reflect the new corrected milestones. (This means more dates become unknown in the tables... urgh!) Comments? - dcljr (talk) 06:59, 12 May 2012 (UTC)

Now that I've written a script to automate the process of checking these milestones against the official database dumps, I'm abandoning this "sampling-based" table in favor of exact counts that I'm collecting at User:Dcljr/Article counts. I've finally announced (see May 10 entry) all the May 10th promotions that I've verified in this manner, but I'm still holding off on the demotions until I get more info. Some May 10th milestones based on the on-wiki/API counts are actually too low, and I've announced these separately and included what the dumps indicate. In the Wikisources table, however (the only one affected by this "milestones too low" problem), I've elected to use the on-wiki levels for now, until the updateArticleCount.php maintenance script can be fixed (bug report) and we can get correct on-wiki counts for all wikis. The Battle Continues... - dcljr (talk) 10:50, 2 June 2012 (UTC)

Wiktionary and Wikisource article counts are correct

OK, I can finally confirm that the article counts after May 10th on the Wiktionaries and Wikisources are in fact correct (see #May 10 article count updates for more context). So, I guess we need to "demote" the wikis that lost enough articles to change to lower milestone levels (a list of which is now given under the May 10th entry at Wikimedia News#May 2012) in the Wiktionaries and Wikisources tables (the promotions to higher levels have already been taken care of). I'll start doing this someday "soon"... Also, note that I plan to post a very lengthy description of what happened on May 10th (and earlier) to the page "Article counts revisited", so we (Wikimedians in general) can discuss whether the current state of affairs is really what we want. (I'll announce the creation of that page in a bunch of different places, not just here.) Stand By... - dcljr (talk) 18:32, 14 June 2012 (UTC)

Yeah... I never posted the description at Article counts revisited. You can read it at User:Dcljr/Article counts, though. One Of These Days, I'll post about it "for real"... - dcljr (talk) 03:02, 14 October 2012 (UTC)

Bashkir Wikipedia

Someone (at 2 IPs) keeps changing old entries about the Bashkir Wikipedia. I keep reverting them because my records of Wikimedia wiki statistics don't support the submitted information. Based on daily API queries, the Bashkir Wikipedia had:

  • 7,300 articles and 98,371 edits at 2012-01-20T05:43:09Z
  • 15,521 articles and 107,849 edits at 2012-01-21T04:00:41Z

This is why I posted the following milestone announcement under 20 January 2012:

  • The Bashkir Wikipedia has reached 15,000 articles (more than doubling in the last 24 hours) and 100,000 page edits.

(BTW, I would have checked the wiki itself before posting to verify the date I was using — i.e., that it happened on the 20th and not the 21st.)

So, I can't see how either of the two versions the anon edits are trying to get the page to say:

[under 15 January 2012]
  • The Bashkir Wikipedia has reached 15,000 articles (more than doubling in the last 24 hours) and 100,000 page edits.

or:

[under 20 January 2012]
[under 15 January 2012]
  • The Bashkir Wikipedia has reached 10,000 articles (more than doubling in the last 24 hours)

could possibly be correct.

(BTW, based on the 2012-07-01 dump of that wiki, the on-wiki article count is actually off by 30 articles from the "real" article count that would result from running the updateArticleCount.php maintenance script. FYI...)

So, to the anon user: please give an explanation of your edits here (preferably with a link to your source, if you have one), or I'm just going to keep reverting you. - dcljr (talk) 03:17, 10 July 2012 (UTC)

Great. Now an anon has changed my comment on this talk page! Lovely... - dcljr (talk) 00:12, 30 July 2012 (UTC)

In the future

What Wikipedias will have reached 500K articles by the end of 2012? What Wikipedias will have reached 200K articles by the end of 2012? What Wikipedias will have reached 150K articles by the end of 2012? What Wikipedias will have reached 100K articles by the end of 2012? What Wikipedias will have reached 75K articles by the end of 2012? What Wikipedias will have reached 50K articles by the end of 2012?

--M'encarta (talk) 01:58, 18 August 2012 (UTC)

It's been a while since I predicted milestones, but this comment spurred me to crunch some numbers. According to article counts from yesterday and approximately 2, 4, 12, 28, 44, and 77 days ago (nevermind why those particular numbers… unless you really want to know), the following wikis that are not currently above nK — for n = 500, 200, 150, 100, 75, 50 — are at least somewhat likely to be above those levels by 2013-01-01T00:00:00Z:
  • The Swedish Wikipedia will almost certainly reach 500,000, based on consistently high activity levels over the past 11 weeks (i.e., sufficiently high to reach the milestone).
  • The Vietnamese Wikipedia could possibly reach 500,000, but only if the recent (insufficiently high) activity levels were to increase to the levels seen 2–11 weeks ago.
  • The Arabic Wikipedia will almost certainly reach 200,000, based on consistently high activity levels over the past 11 weeks.
  • The Turkish Wikipedia will almost certainly reach 200,000, based on consistently high activity levels over the past 11 weeks.
  • The Indonesian Wikipedia could possibly reach 200,000, but only if very recent activity levels (seen over the past 2 weeks) continue.
  • The Kazakh Wikipedia could possibly reach 200,000, but only if activity levels were to increase to those seen 2–4 weeks ago.
  • No Wikpedia not currently above 150,000 is at all likely to reach that level, based on activity levels seen over the past 11 weeks.
  • The Estonian Wikipedia will reach 100,000 in the next few weeks.
  • The Burmese Wikipedia is not at all likely to reach 100,000 based on typical activity levels seen over the past 11 weeks, but if heavy bot activity seen 4 weeks ago were to resume and continue for a few months, it might reach that milestone.
  • The Uzbek Wikipedia could possibly reach 75,000, but only if activity levels were to increase to those seen 4–11 weeks ago.
  • The Belarusian Wikipedia will almost certainly reach 50,000, based on consistently high activity levels over the past 11 weeks.
  • The Tamil Wikipedia could possibly reach 50,000, based on having sufficiently high activity levels from time to time over the past 11 weeks.
- dcljr (talk) 21:51, 18 August 2012 (UTC)
25000? 10000? 5000? 2000? M'encarta (talk) 16:46, 19 August 2012 (UTC)
Uh, OK. The following predictions are based on the exact same data as before, and so are worded as if they were being posted at the same time (Aug 18th):
  • The Bengali Wikipedia could possibly reach 25,000, but only if the activity levels seen in the last week or so were to continue.
  • The Afrikaans Wikipedia will probably reach 25,000, based on sufficiently high activity levels seen from 11 weeks ago up until only a couple of days ago.
  • The Zazaki Wikipedia could possibly reach 25,000, but only if activity levels seen in the last couple of days were to continue for the next 3 months.
  • The Min Nan Wikipedia could possibly reach 10,000, but only if activity levels were to increase to those seen 2–11 weeks ago.
  • The Scots Wikipedia will almost certainly reach 10,000, based on consistently high activity levels seen over the past 11 weeks.
  • The Yiddish Wikipedia could possibly reach 10,000, but only if activity levels seen about 4 days ago were to return and continue for the next 3 months.
  • The Mazandarani Wikipedia could possibly reach 10,000, but only if activity levels were to increase to those seen 4–11 weeks ago.
  • The Kyrgyz Wikipedia will almost certainly reach 10,000, based on consistently high activity levels seen over the past 11 weeks.
  • The Nahuatl Wikipedia could possibly reach 10,000, but only if activity levels seen about 2 weeks ago were to return and continue for the next 3 months.
  • The Dutch Low Saxon Wikipedia will almost certainly reach 5,000, based on consistently high activity levels seen over the past 11 weeks.
  • The Pashto Wikipedia could possibly reach 5,000, but only if activity levels seen 6–11 weeks ago were to return.
  • The Rusyn/Ruthenian Wikipedia could possibly reach 5,000, but only if activity levels seen in the last 2 weeks were to continue.
  • The Mingrelian Wikipedia could possibly reach 5,000, based on having sufficiently high activity levels from time to time over the past 11 weeks.
  • The Zeelandic Wikipedia could possibly reach 5,000, but only if activity levels seen about 2 weeks ago were to return and continue for the next 3 months.
  • The Gagauz Wikipedia will almost certainly reach 2,000, based on consistently high activity levels seen over the past 11 weeks.
  • The Interlingue Wikipedia could possibly reach 2,000, but only if the activity levels seen in the last week or so were to continue.
  • The Lingala Wikipedia will probably reach 2,000, based on sufficiently high activity levels seen from 11 weeks ago up until only a couple of days ago.
  • The Veps Wikipedia will probably reach 2,000, based on sufficiently high activity levels seen from 11 weeks ago up until only a couple of days ago.
  • The Guarani Wikipedia could possibly reach 2,000, but only if recent activity levels were to increase to those seen 2–11 weeks ago.
  • The Shona Wikipedia could possibly reach 2,000, but only if activity levels seen about 2 weeks ago were to return and continue for the next 3 months.
And, before you ask:
  • The Samoan Wikipedia could possibly reach 1,000, but only if activity levels seen in the past 2 weeks were to continue.
  • The Lezgian Wikipedia could possibly reach 1,000, but only if activity levels were to return to those seen 6–11 weeks ago.
  • The Tetum Wikipedia could possibly reach 1,000, but only if activity levels seen about 6 weeks ago were to return and continue for the next 3 months.
  • The Cheyenne Wikipedia could possibly reach 1,000, but only if recent activity levels (last week or so) were to continue.
- dcljr (talk) 01:17, 24 August 2012 (UTC)

Semi-protection

After resisting it for a while now, I'd like to request that some admin who watches this page (Wikimedia News) go ahead and semi-protect it, so that only registered users can edit it. Some anon doofus enjoys submitting bogus news items here a little too much... (Objections?) - dcljr (talk) 01:36, 23 August 2012 (UTC)

Done temporary for two weeks, some hours later before you asked. Regards. -- MarcoAurelio (talk) 07:52, 29 August 2012 (UTC)

cbk-zam

cbk-zam: used to have more than 2,300 articles until late April when it dropped below 2,000, and then below 1,000 on May 1st. This was part of massive cleaup effort by cross-wiki admins / stewards. Should we "reset" this wiki's entry in the Wikipedia table (i.e., put it in the 1,000 row with date 29 September 2012) or leave it where it is (in the 2,000 row with date 28 July 2010)? Opinions? And what name should we use for the wiki, BTW? "Zamboanga Chavacano" (used in the List of Wikipedias), "Zamboangueño Chavacano", "Zamboangueño", or "Chabacano de Zamboanga" (all of which appear in en:Chavacano language)? - dcljr (talk) 04:36, 2 October 2012 (UTC)

I've gone ahead and demoted the wiki in the Wikipedia table as described above. This is how other significant drops in article counts have been handled here in the past (example). But if a wiki just happens to drop below a recent milestone through "normal" editing actvity, I don't demote it in the table. FYI. - dcljr (talk) 04:09, 9 October 2012 (UTC)

Mass demotions in Wiktionaries and Wikisources tables

I've finally gotten around to updating the Wiktionaries table, demoting the wikis that dropped to lower milestone levels after their on-wiki stats were corrected on 10 May 2012. I'm about to do the same to the Wikisources table. Since I usually ignore wikis that bounce down and up around a recently achieved milestone, I didn't notice that some Wiktionaries legitimately reached new milestones shortly after the May 10 count changes. I've promoted those in the table, too. Now things should actually be correct again. (I hope!) - dcljr (talk) 04:39, 15 October 2012 (UTC)

Ditto for the Wikisources table now. BTW, although demotions usually revert back to the date they first reached the lower milestone, since these demotions are based on fixing broken stats, I've dated all of them 10 May 2012, the date the stats were fixed. - dcljr (talk) 04:59, 15 October 2012 (UTC)

Pre-RFC discussion about changes to article counts

As promised, I've posted a little something at Article counts revisited about the changes in article counts seen in the Wiktionaries and Wikisources back in May 2012. I kept most of the details at User:Dcljr/Article counts, since that page is quite lengthy (and not really finished, unfortunately). I figure I'll eventually open an actual RFC about this, but I figured I could use some "pre-RFC discussion" first. Please see Talk:Article counts revisited for some of the issues I think need discussing, and comment there, if you want. - dcljr (talk) 07:25, 15 October 2012 (UTC)

Bold in Wikidata announcements

I see someone has been treating item counts at Wikidata the same way as content pages at the main content wikis — that is, marking the counts in bold. This is problematic at this point for two reasons:

  1. Wikidata is not listed as a main content wiki on the homepage of the English Wikipedia (for example) or here at Meta. Until it is listed as such, we probably shouldn't treat it as main content wiki.
  2. More to the point, the item count at Wikidata isn't even for content pages ("articles"). The content-page count there is stuck at 0, and will be for the foreseeable future, since pages only count as content if they include a local [[wikilink]]. Pages at Wikidata will never have wikilinks in them, by the very nature of that wiki.

Until these things change (the second problem could be addressed, for example, by submitting a request [somewhere...] to count content pages at Wikidata by the "all" method), I don't think we should mark the announcements in bold like we do for the main content wikis. If these things do change, we can go back later and mark the announcements in bold. - dcljr (talk) 04:00, 12 November 2012 (UTC)

Whoops, I spoke too soon. I didn't realize that many pages at Wikidata were going to contain interwiki links. The on-wiki article count (i.e., "content pages" at d:Special:Statistics) is now just above 1 million, but the highest-numbered "Q" page (consecutively numbered page in the main namespace) is something over 1,525,000 — although the "total pages" count is 1,470,000. For the time being, I'm personally not announcing article-count milestones at Wikidata, but I'm not removing them when other people add them (as long as they're correct). - dcljr (talk) 17:40, 23 December 2012 (UTC)
Following this exchange at Wikidata, I've decided that the on-wiki stats of that wiki are "good enough" to go with for "article count" announcements, so I've added 2 million on Jan 4th to the Wikidata table (based on saved records). Still not sure how I feel about bold in the Wikidata announcements, so I've left it off of the one I just posted. I'm sure someone will come by later and want to add it... - dcljr (talk) 02:00, 18 January 2013 (UTC)
You know, now I'm starting to lean towards including bold on Wikidata announcements, because if it's going to have a table on the page, like Commons and Wikispecies, it should probably be treated the same with respect to having bold milestones. Opinions? - dcljr (talk) 23:47, 15 February 2013 (UTC)
Agreed on both bold and having a table. James F. (talk) 18:14, 18 February 2013 (UTC)

Abkhazian below 1K again

I've decided to "take back" the promotion of the Abkhazian Wikipedia I made back on October 7th, because 3 days later an administrative action removed a large number of "illegitimate" articles (apparently pages written in the wrong language), some/most of which I believe were the pages that made the wiki go over 1K in the first place. (Granted, I don't have access to the page histories of the deleted articles, but ab:Special:NewPages doesn't show any page-creation activity during that time period, so I assume that means the rise to 1K was due to the pages that were deleted on Oct 10th.) Anyway, I've put the wiki back to where it was in the "Wikipedias" table before the promotion, with the same date as before. - dcljr (talk) 01:04, 21 November 2012 (UTC)

some wikipedias have fallen below milestones

the aromanian wikipedia has fallen below 60,000 articles on december 15, 2012. can anyone fix it, please?Vincentangeles005 (talk) 13:02, 15 December 2012 (UTC)

also for zazaki wikipedia which also went below 10,000 articles sometime between december 10-15.Vincentangeles005 (talk) 13:02, 15 December 2012 (UTC)
As I've mentioned here before, I (for one) usually don't demote wikis when they drop below recent milestones unless it's the result of a particularly major administrative action (e.g., deletion of thousands of articles, especially ones that just caused the milestone in the first place). This is what has just happened to the Aromanian Wikipedia (a loss of almost 60% in article count in one day), except that I can't tell whether it was mainly the most recent additions that were just deleted. The wiki is now just above 20K, a level that it reached on 22 May 2010, thanks to massive bot activity. Since I can't really say the 20K level was any more "legitimate" than the 60K level (or any in between), I've decided to demote Aromanian to the 20K level in the Wikipedias table with today's date (that is, the 15th) instead of the date it first reached the milestone. As for the Zazaki Wikipedia, that has dropped to just below 10K, a level that it quickly rose to, again thanks to bot activity, back in June 2011 (doubling in size in a week). It then grew to 15K three months later, where it stayed until mid-Nov 2012, at which point it started to lose articles steadily, thanks to an admin's massive cleanup effort. Looks like the cleanup is still ongoing. I'm willing to wait a bit to see where it bottoms out. - dcljr (talk) 02:24, 16 December 2012 (UTC)
As suggested would happen by an admin over at Zazaki Wikipedia, that wiki has finally dropped below 5K, thanks to a long-term cleanup effort (which appears to be alllmost finished). Hopefully sometime soon it will (legitimately) rise back up to 5K. When that happens, I plan to relist it in the table at that level (from its current 15K level). - dcljr (talk) 04:16, 19 February 2013 (UTC)
Well, it's been just over 4K for a while now, so I decided to treat it the same as the Aromanian Wikipedia (which I just demoted) and demote it to the 2K level in the table with the date it first reached that milestone, AFAICT. - dcljr (talk) 01:11, 15 March 2013 (UTC)

what is the http 991 problem?

http 991 status is inaccessible to some projects. is it a real problem? Vincentangeles005 (talk) 12:21, 22 December 2012 (UTC)

What is an HTTP 991 status? - dcljr (talk) 17:43, 23 December 2012 (UTC)
it is 200 but unserializing api output failed, which affected most wikimedia projects. can i create new pages and edit them if there is a 991 status? Vincentangeles005 (talk) 02:33, 26 December 2012 (UTC)
I don't know. You might have better luck asking at mw:Talk:API or w:Wikipedia:Village pump (technical). - dcljr (talk) 01:58, 2 January 2013 (UTC)

Removing 2012 content

Because it bothers me when I see a page like this containing very little information, I've inserted HTML-comments into the page source requesting that editors not remove the older 2012 content until certain dates in the coming months. Last year I tried to do something similar, but ended up ignoring most of my own date choices because they seemed to leave the older info on the page for too long. This year, I've decided to request that people leave the January–March 2012 sections on the page until at least 1 Feb 2013, the April–June 2012 sections until at least 1 March 2013, and July–December 2012 sections until at least 1 April 2012. Any objections? - dcljr (talk) 01:52, 2 January 2013 (UTC)

Commons milestones

Are we going to do every multiple of 1 million for Commons' files? That's gonna be a lot of entries going forward... especially seeing as how the wiki's growth rate is still increasing (i.e., accelerating growth). Should we put off more Commons announcements until it hits 20 million, then start doing, say, every 5 million after that? - dcljr (talk) 03:15, 2 February 2013 (UTC)

I think 1m -> 20m, then every 5m -> 100m. After that point we can worry some more. James F. (talk) 18:13, 18 February 2013 (UTC)
My measurement stick would be the length it needed to achieve another milestone. If it's less than a month (right now it's still 59 days), then we should go for 5 million interval. But James' idea is also doable. Bennylin 16:54, 21 February 2013 (UTC)

cbk-zam demoted again

I've taken the liberty of demoting the Zamboanga Chavacano Wikipedia in the "Wikipedias" table again because the bot-created stubs that caused the wiki to blow up to 10,000 two weeks ago have been deleted by cross-wiki admins/stewards. It's now back to the 1,000 level with the date it first reached that milestone. - dcljr (talk) 01:59, 1 March 2013 (UTC)

Aromanian Wikipedia to 10K in table

The Aromanian Wikipedia has dropped from 25,127 articles 24 hours ago to about 10,500 as I post this (and it seems very likely that it will continue shrinking). Originally I used today's date (i.e., 13 March 2013) when demoting it to the 10K level in the "Wikipedias" table, mostly because I couldn't find the date it first reached 10K in my records. But a check of the wiki's table at Wikistats.org and the page history of m:List of Wikipedias/Table verifies that it blew up from c. 5K to over 20K in less than 2 weeks back in mid-May 2010, and actually passed 10K on (most likely) 11 May 2010, so that's the date I am now using in the table. FYI. - dcljr (talk) 01:52, 14 March 2013 (UTC)

I knew I would have to demote it again today... The Aromanian Wikipedia has now dropped below 5K, so I've done more research and found that it passed 2K back around 22–25 December 2007. So that's where it stands now (2K in table, with date range 22–25 December 2007). I wonder if this aggressive deletion of bot-created micro stubs will spread to other wikis.... - dcljr (talk) 00:59, 15 March 2013 (UTC)
I do hope other admins have the same courage :) Bennylin 08:19, 3 April 2013 (UTC)
Finally demoted Zazaki the same way. - dcljr (talk) 01:12, 15 March 2013 (UTC)
And now the Aromanian Wikipedia has dropped below 2,000 articles as a result of continuing administrative deletions of "no-content" articles, so I've demoted it again to the 1,000 level in the Wikipedias table with the date it first reached that level, AFAICT (somewhere between 10 Nov and 16 Dec 2007). - dcljr (talk) 05:41, 2 April 2013 (UTC)

Nepali Wiktionary to 100

It seems that when I updated the article counts for the Wiktionaries back on 15 Oct 2012, I neglected to remove the Nepali Wiktionary from the 2K level (it should have been removed from the table completely since it dropped below 100 when its on-wiki stats were fixed on 10 May 2012). Well, today it reached 100 again, so I've demoted it to the 100 level with today's date. This is a legitimate increase to 100, so the date is today (well, yesterday) and not the date it first reached 100, years ago. - dcljr (talk) 00:22, 8 April 2013 (UTC)

Aromanian and Russian Buryat to 10K

The Aromanian and Russian Buryat to 10K, after 2 or 3 weeks falling down to 1K (made a mistake of 1K articles) with a lots of stubs to add articles, more to come to 20K cooming soon. —The preceding unsigned comment was added by 38.97.5.251 (talk) 23 April 2013‎

I don't know who you are (you forgot to sign), but please stop disinforming people by adding false facts to the Wikimedia News main page. For the second time, a wrong fact about Aromanian wiki has been submitted. If you are not happy because of the fact that the Aromanian Wikipedia has lost so many articles, go over there and create as many articles as you can, but with more text content and information on them. Jagwar grrr... 20:12, 23 April 2013 (UTC)
Actually, neither the Aromanian nor the Russian Buryat Wikipedia even have 10,000 total pages, forget articles. I've reverted all of the anon's recent edits. - dcljr (talk) 02:35, 24 April 2013 (UTC)

German Wikivoyage

In the last 28 hours, the German Wikivoyage dropped from 15,230 to 12,772 content pages with no apparent corresponding deletion activity. Based on a discussion at voy:de:Wikivoyage:Lounge#Seitenstatistik that (I think) I kind of understand, it seems that the on-wiki stats were wrong and have just been fixed (although I can't find a bug report about it, so I'm not sure how it was accomplished). According to the deletion log, there have been fewer than 400 page deletions since the separate Wikimedia wiki was created on 15 January 2013, so it looks like it's never actually been above 15K since that date — but since I don't know for sure, I've moved the wiki to 10K in the table with the date 16 May 2013. - dcljr (talk) 04:04, 17 May 2013 (UTC)

The explanation. - dcljr (talk) 22:36, 17 May 2013 (UTC)

Aromanian to 1,000,000

Aromanian to 1,000,000 on June 25, 2013, with a lot of articles and stubs. —The preceding unsigned comment was added by 38.97.5.251 (talk) 26 June 2013‎

Nope. - dcljr (talk) 00:01, 27 June 2013 (UTC)

Wikivoyage articles

I'm not sure why the "content pages" for Wikivoyage haven't always been referred to as "articles", like Wikipedia and Wikinews. Clearly (IMHO), what Wikivoyagers are writing are "articles" about what you need to know when visiting different places. So I've gone ahead and made the change everywhere on the page that needed it (none of the archived pages needed changing, since the project is so new). - dcljr (talk) 01:22, 28 September 2013 (UTC)


Archive 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.