Talk:Wikimedia News/archive 3

Showing coming milestones?

An anon user has been removing the "next coming milestone" (4,000,000) from the Wiktionary table and adding more than the "next milestone" in both the Wikipedia table (the 4,000,000 and 5,000,000 rows) and the Wikinews table (the 80,000, 90,000 and 100,000 rows). I'm not sure the reason for the inconsistency, so I've asked them to explain here. In the meantime, does anyone else have an opinion on how we should handle the upcoming milestones in the tables ? In the past, I believe the next milestone (and its color) was mentioned in the wiki source in <!-- HTML comments --> only, but at some point they were made visible (I can't remember now whether I did that or someone else). Now we have some strange hybrid scheme being used that I don't understand. Hopefully we can get some clarification from the user... - dcljr (talk) 10:55, 4 June 2012 (UTC)

Apparently not. They keep reverting things back to the way they want to do it with no further explanation, despite repeated requests to discuss the issue here on this talk page. As described above, this leaves the tables in an inconsistent state, with some showing 1, 2 or even 3 empty "upcoming" milestone levels at the top and others not showing any (stopping at the last achieved milestone). I really don't know what they're thinking. In their latest edits, they even added the same colors to the Commons table that are used in the other tables. Problem is, the Commons table consists of a very different selection of milestones, so the colors don't "make sense" there the way they're intended to in the other tables (see the #Colors section above and en:Wikipedia talk:Milestone statistics/Archive#New table colors for explanations of the colors used in the first 7 tables — short version: 1* = blue, 2* = green, 5* = yellow, interpolate for levels in between). Anyway... I'm sorry, but these inconsistencies are simply not acceptable. I know it's not a Big Deal, but it's the principle of the thing: when someone disagrees with or questions your edits, you take it to a talk page and discuss things; you don't just keep resubmitting the same changes until the other user gives up. So, to User:82.227.172.214 I say: until you put forth some kind of argument (not necessarily even one I can agree with... just any argument) for why your way of doing things should be adopted, I'm going to keep reverting away your changes. Not because I'm just trying to "get my way", but because you haven't tried at all to work toward consensus on this and your changes don't make sense. As can be seen in section after section above, pretty much every major change I've made to the page I've asked for consensus about on this talk page. You're just trying to ram through your changes with no discussion of what you're thinking or why it's a good idea. You've said that we shouldn't show the 4,000,000 row in the Wiktionaries table until English reaches 3,900,000. Okay, that's one way of doing things... so why not apply the same idea to any other table on the page? Why do it this way only in the Wiktionaries table? Explain it to me. Please. - dcljr (talk) 03:00, 10 June 2012 (UTC)

Oh, well. The IP has now been blocked. - dcljr (talk) 08:47, 11 June 2012 (UTC)

Hello, an user dcljr has been added the "next coming milestone" (6,000,000 rows) from the Wikipedia table. I'm not sure the reason for the inconsistency, but this is the maximum milestone from (5,000,000 rows all wikis). In the meantime, does anyone else have an opinion on how we should handle the upcoming milestones in the tables remove the next milestone after (5,000,000 rows) ? I want edit to believe the final milestone (and its color) was mentioned in the wiki source in <!-- HTML comments --> only, but at some point they were made visible (I can't remember now whether I did that or someone else). Now we have some strange hybrid scheme being used that I don't understand. Hopefully we can get some clarification from the user dcljr... - 2a01:e35:2e3a:cd60:ed2b:6c0f:2468:f8c0 (talk) 21:34, 1 November 2015 (UTC)

If you look at all of the tables under "Projects by number of content pages", you will see that each one lists an empty row for the next milestone that hasn't been achieved yet (among those that we actually track, of course). It's been this way for years. I didn't just change to some strange "hybrid" system. I'm following the way it's been done here since at least 2012. (By the way, given the parallel wording of this IP user's comment and my own at the beginning of this section, I am inclined to believe that this is trolling. But on the off-chance this is a legitimate question, I am willing to discuss it.) - dcljr (talk) 04:37, 2 November 2015 (UTC)

Well, you must waiting to be at least 9,500,000 articles from English Wikipedia to create a next milestone at 10,000,000 rows articles. But, you cannot create the next milestone at 6,000,000 rows articles in Wikipedia because it's too much early to create. Or, waiting to English Wikipedia reached 5,900,000 articles to create a non articles milestone at 6,000,000 rows articles. - 2a01:e35:2e3a:cd60:ed2b:6c0f:2468:f8c0 (talk) 20:07, 3 November 2015 (UTC)

I think I know what you're saying. OK, that's one way of doing it (although that doesn't quite match the HTML comments you added to the page), but the fact is, since 2012 all the tables here (for the "interlingual" projects, like Wikipedia, Wiktionary, etc.) have shown one empty row at the top — and we haven't been waiting until a wiki was "close" to the milestone to add the empty row. By the way, things are done differently at en:Wikipedia:Milestone statistics. In that table, no empty row appears at the top until the new milestone is actually achieved. Either way is fine, but we should be consistent with all similar tables on the same page (the tables for "multilingual" projects, like Commons and Wikispecies, are handled differently because they contain a very different selection of milestone levels). So are you actually looking for a consensus to change the way we handle the tables here? - dcljr (talk) 05:28, 4 November 2015 (UTC)

Ok, thank to contribute in English Wikipedia milestone. But, the original maximum rows of Wikipedia is 5,000,000 rows articles in Wikimedia News. You're understanding that it is not possible to add the next milestone at 6,000,000 rows articles in Wikipedia this year. So if you want add to the next milestone at 6,000,000 rows articles in Wikipedia or Wikitionnary, please wait for in 2017 to create a new rows. - 2a01:e35:2e3a:cd60:ed2b:6c0f:2468:f8c0 (talk) 21:47, 4 November 2015 (UTC)

That's a strange diff to link to, but OK, yeah, in Feb 2011 the Wikipedias table showed empty 4-million and 5-million levels. So why are you using the fact that it used to show two empty rows to justify now not showing any empty rows? I guess I just don't understand. - dcljr (talk) 02:39, 5 November 2015 (UTC)

No. I guess i just watch the original milestone but this rows is maximal at 5,000,000 articles in Wikipedia and 5,000,000 entries in Wiktionnary. I have a difference on February 2011 and November 2015. I say the Wikipedias table showed empty 4-million and 5-million levels in February 2011 and now on November 2015 the Wikipedias table showed empty 4-million and English Wikipedia at 5-million levels on Wikimedia News. I justify doesn't not create the next milestone but, all wikis the milestone is maximum at 5,000,000 rows. When you create a next milestone from exemple : (the French Wikisource reached 1,500,000 text units) you can create the next milestone at 2,000,000 to 5,000,000 rows text units. So you've got more than 4 milestones remaining to create. Second exemple : (the English Wiktionnary reached 5,000,000 entries) you can not create the next milestone. The milestone is maximum at 5,000,000 rows entries. From now, you must add the next milestone from Wikiquotes to Wikivoyages and do not create the next milestone at Wikipedia and Wiktionnary. I remove the 6,000,000 rows articles in Wikipedia because you can not add the next milestone. Are you understand ? - 2a01:e35:2e3a:cd60:ed2b:6c0f:2468:f8c0 (talk) 14:39, 6 November 2015 (UTC)

No. Can anyone else explain this? Perhaps if you posted your comment in your native language, that might help? You lost me at "all wikis the milestone is maximum at 5,000,000 rows." This seems to be related to the HTML-comment suggestion that "You cannot add the next milestone after 5,000,000 rows." But this doesn't make any sense for any tables other than the Wikipedias and Wiktionaries, because in the other projects no languages are anywhere close to the 5-million level. You say, "From now, you must add the next milestone from Wikiquotes to Wikivoyages and do not create the next milestone at Wikipedia and Wiktionnary." Why? How does the maximum milestone of Wikipedia and Wiktionary languages affect in any way the milestones of Wikiquote, …, Wikivoyage languages? You say, "(the French Wikisource reached 1,500,000 text units) you can create the next milestone at 2,000,000 to 5,000,000 rows text units. So you've got more than 4 milestones remaining to create." Are you saying all 4 levels should be created at the same time?? (I am now wondering whether this has anything to do with the HTML comment, "EDITORS: Please list only ONE milestone (the latest one) per language." If so, that was intended to prevent people from listing two different milestones in the same table for the same language, which has happened in the past; it has nothing to do with whether empty rows are shown at the top of the tables.) Anyway, I'm saying we either: (1) always have one empty row at the top of each table showing the next available milestone, or (2) have no empty rows at the top of any table at any time. Both of those approaches are internally consistent; we simply need to decide which way we want to do it (I have been taking approach #1). You seem to be suggesting something else. What? What is the rule you want us to follow? Is it that no empty rows be shown in a table until the largest wiki in that project is within some distance of the next milestone? If so, what distance? In your French Wikisource example, it was within 500,000 text units, but in your comment on 3 November 2015, you cited one case where it was within 500,000 articles and another where it was within 100,000 articles. This is (perhaps) why I'm having trouble figuring out what rule you want to follow. You don't seem to be using a single, consistent rule. - dcljr (talk) 02:19, 7 November 2015 (UTC)

Because, you say : "Is it that no empty rows be shown in a table until the largest wiki in that project is within some distance of the next milestone ? If so, what distance ?" Yes, the distance of milestone it's maximum at 5,000,000 rows but it is the rules all wikis. However, it has nothing to do with people who aim to prevent the next milestone. Accordingly because, Wikipedia and Wiktionnary in Wikimedia News you don't add the 6,000,000 rows because it is too long. Maximum 32 lines from 100 to 5,000,000 rows all wikis. But, you don't know. Or of course, maybe if English Wikipedia reaches 5,900,000 articles You add one milestone at 6,000,000 rows articles exceptionally. - 2a01:e35:2e3a:cd60:ed2b:6c0f:2468:f8c0 (talk) 14:50, 7 November 2015 (UTC)

Umm… How did you decide that having more than 32 rows in a table makes it "too long"? I don't see how this is a reasonable criterion to use, so I'm going to stop discussing this now. - dcljr (talk) 01:27, 8 November 2015 (UTC)

But ok, it's your decision. I leave you exceptionally a empty milestone at 6,000,000 rows articles in Wikipedia. And after, you needn't not add a empty milestone at 6,000,000 rows entries in Wiktionnary after the English Wiktionnary reached 5,000,000 rows entries. It's ok ? Well, i'm tired of this discussion. - 2a01:e35:2e3a:cd60:ed2b:6c0f:2468:f8c0 (talk) 19:29, 9 November 2015 (UTC)

Grammar

OK, this is really starting to bug me... Could fluent English speakers please give their opinions about how the news items are worded? The standard format is:

• X has reached Y.

Which is fine... but when additional context is needed, I've been using:

• X has reached Y, as Z happens.

Or sometimes:

• X has reached Y, after Z happens.

But those don't seem quite right to me (although they would be fine if "has reached" were replaced by "reaches" — unfortunately, we don't use "headline tense"). Wouldn't the phrase "has reached" call for something more like...?

• X has reached Y, now that Z has happened.

Unfortunately, that just sounds weird to me. Sometimes I try to avoid the problem by saying something like:

• X has reached (or fallen below) Y, having gained (or lost) Z.

But that doesn't work as well when I need to attribute a rise to bot activity or a fall to administrative action:

• X has reached (or fallen below) Y, having gained (or lost) Z due to Q.

Ugh. Any ideas how we could improve the wording when additional context is provided? Should we just switch to headline style?

• X reaches Y as/after Z happens.

Or use headline style only when there's additional context (keeping "has reached" for most cases)? Opinions? - dcljr (talk) 02:57, 13 April 2013 (UTC)

Back when I edited this page frequently, I used to use headline style for everything. It's simpler that way. But it seems past perfect tense has won out... – Minh Nguyễn (talk, contribs) 23:59, 8 February 2014 (UTC)

Archive for 2013

Let's move the records in 2013 to archive (or delete if the archive exists)? --Brateevsky (talk) 14:18, 19 January 2014 (UTC)

The archive already exists. See #Removing 2012 content for my thoughts on removing old content (from a year ago), and this diff for my current ideas regarding removing 2013 content. Since it's after Jan 15th, I've gone ahead and removed the first 3 months of 2013. - dcljr (talk) 14:34, 19 January 2014 (UTC)

OK. --Brateevsky (talk) 18:50, 21 January 2014 (UTC)

Sinhala Wikipedia backs to 500 Articles

• The Sinhala Wikipedia is removing 9500 articles, then back to 500 articles. — The preceding unsigned comment was added by 172.4.225.64 (talk) 7 February 2014 (UTC)
  • OK, when that happens, we can announce it. - dcljr (talk) 03:27, 7 February 2014 (UTC)

Burmese to 1,000,000

Burmese to 1,000,000 on May 3, 2014, with a lot of articles and stubs. — The preceding unsigned comment was added by 172.4.225.64 (talk) 3 May 2014 (UTC)

@Dcljr: why is an IP adding false info here? PiRSquared17 (talk) 23:44, 3 May 2014 (UTC)

Because they have nothing better to contribute to humanity, I would assume. - dcljr (talk) 01:28, 4 May 2014 (UTC)

crhwiki and mhrwiki

As noted on the page, IanraBot is creating a lot of practically useless, empty pages. I'm not sure what the purpose of these articles is. Are they really going to add content to all these year articles? PiRSquared17 (talk) 03:23, 9 May 2014 (UTC)

Loss of Reflinks tool

It seems that Reflinks, previously hosted on Toolserver, is no longer functioning. Reflinks is a tool essential for our mission to provide citations and enhance verifiability. It simplifies the job of populating references, yet the developer feels that the WMF is forcing him to abandon the tool. It's exactly cases like these that makes me wonder where I should be asking questions about the whys and the wherefores. Not only that, I'm extremely disappointed and slightly angry that this has happened, and would like to appeal to the powers that be to reconsider. --Ohconfucius (talk) 07:48, 1 July 2014 (UTC)

This might not be the best place to complain about this. Certainly, the "powers that be" will not be reading this page. There seems to be a bitchfest in progress at w:Wikipedia:Village pump (technical)#Toolserver shut down. - dcljr (talk) 08:18, 2 July 2014 (UTC)

Nahuatl Wikipedia to 10,000

The Nahuatl Wikipedia has been reached to 10,000 articles, with 70 articles added. — The preceding unsigned comment was added by 172.4.225.64 (talk) 22 August 2014 (UTC)

Yes, but those articles were added by you and are not in Nahuatl but English. It is my opinion that we should not announce the milestone until the articles are in Nahuatl, which I have requested that an admin over there do (or delete, if they feel that's more appropriate). - dcljr (talk) 02:38, 22 August 2014 (UTC)

Pages have been deleted by a global admin. - dcljr (talk) 05:03, 23 August 2014 (UTC)

March 29th demotions

As I just announced, a great many (65) wikis are being demoted today to lower article-count milestones because their on-wiki article counts (which were just plain wrong due to a number of long-standing bugs in the MediaWiki software) have just been recalculated. Usually when a wiki gets demoted in one of the tables, the date the previous milestone was reached is used. Obviously, if the article counts were wrong, then the previous milestone's date is likely to have also been wrong. For this reason (and because there's just so many of them!), I am using "today's date" (29 March 2015) for all of the new, lower milestones in the tables. - dcljr (talk) 04:30, 30 March 2015 (UTC)

I see a few wikis were actually already in the "right" level, but with the "wrong" date (i.e., their increase to a higher level was not noted at the time they reached a higher milestone, and they just fell back down to their old milestone level again). I'm using the 29 March 2015 date for these, too, because, as I said before, the "old" date is likely not correct, anyway. (Actually, this might not be true for wikis that have been at lower levels for many years [say, 8 or more] — but this would lead to arbitrary, unreliable "judgment calls", so I'm not bothering. They all get the same date.) - dcljr (talk) 04:57, 30 March 2015 (UTC)

Actually, sometimes a wiki was not promoted to a higher level because it was noticed that the milestone was reached "illegitimately" (e.g., because of a known article-counting bug, such as the importing of articles increasing the article count by the total number of revisions instead of the actual number of imported pages — and this is why even the count on the date a wiki first opened for editing is not necessarily reliable). I noticed/realized this as I was updating the tables. I may be back later to replace the dates with more "accurate" ones, if I want to spend the time slogging through old Wikistats data (which are counted "from scratch", independently of the on-wiki stats). - dcljr (talk) 05:48, 30 March 2015 (UTC)

I think that using 29 March 2015 as a date for all these wikis sends a wrong message. We usually use the date when the wiki went up, not when the wiki went down. For example, we see at 5,000 that both Zazaki and Sanskrit Wikipedias have reached it on 29 March. However, in reality these wikis had completely different sizes as of 29 March: Zazaki Wikipedia was very close to 5,000, while Sanskrit one was close to 10,000. Thus I think 29 March is an appropriate date for wikis that have grown on this date, but for wikis that have decreased it is better to use the previous date: even if it is not quite correct, this is the best estimation of the date project first reached the milestone — NickK (talk) 08:39, 31 March 2015 (UTC)

You want to go through and figure out the "right" date for 65 wikis (actually a little less, since some wikis dropped out of their tables completely)? Be my guest. [g] I'll probably do this Some Day... just not now. - dcljr (talk) 06:26, 1 April 2015 (UTC)

The "right" date doesn't exist, IMHO. We don't have tools for article count time travel and anything might have happened in the meanwhile. With the monthly update, we'll have to look for events only in the past month. :)

Adding notes to past milestones can be useful, but given the size of this correction it's also useful to have a list of all changes together in a single section. (In fact this was asked at the Wikimedia Forum, it was a very useful job.) --Nemo 07:23, 5 April 2015 (UTC)

I wonder if it is correct to include closed wikis in your massive list under March 29. For example, Sindhi Wikinews, which dropped below a milestone, has been closed for some time, so I don't think it is right to include it in the list. Closed wikis have no prospect of increasing their article counts, for obvious reasons, and should probably be excluded from this page altogether (except perhaps to report on their closure). This, that and the other (talk) 05:48, 5 April 2015 (UTC)

I can see why noting the ones that are closed would be a good idea. Excluding them from consideration altogether? I don't agree. If a wiki changes milestone levels, I think it should be reported here (except in cases of vandalism). And when it changes because the previous level was just plain wrong, well, all the more reason to note it. Nemo: regarding "tools for article count time travel", that's pretty much what Wikistats is. And no one was suggesting that we not keep the March 29th announcement all together or that we go back and change past announcements, NickK was just questioning the dates I used in the tables. - dcljr (talk) 12:08, 5 April 2015 (UTC)

Tagalog Wiktionary

I've just "relisted" the Tagalog Wiktionary at 15K in the Wiktionaries table with today's date. Usually I let a milestone stand if a wiki quickly drops below its most recently achieved milestone, as long as it looks like "normal activity" that caused the drop (i.e., not the deletion of the very articles that caused the milestone in the first place). The Tagalog Wiktionary grew more than 10 times in size over a 4-day period in late November 2011, topping out at just over 16K, and then 5 days afterwards lost 20% of its entries in a 24-hour period to end up at just below 13K. It's been below 15K ever since. Given the vast swings in entry count over such a short period of time and the long time its spent below 15K since, I think it's reasonable to discount the earlier milestone and go with today as the date it legitimately crossed the 15K mark. - dcljr (talk) 01:02, 31 May 2015 (UTC)

Removing old content

As has been done in the last few years, I have added HTML comments suggesting that we keep the 2015 announcements around for a while, since I really dislike coming to a "news" page like this one and seeing almost no content because everything has been archived. (The 2015 content is already at Wikimedia News/2015.) As I did last year, I have suggested removing three months of old announcements at a time, twice a month starting Jan 15th (thus, the Jan–Mar 2015 announcements will be kept until around Jan 15th, Apr–Jun until around Jan 31st, Jul–Sep until around Feb 15th, and Oct–Dec 2015 until around Feb 29th). Objections? - dcljr (talk) 04:21, 2 January 2016 (UTC)

I agree! --Antonio1952 (talk) 14:59, 3 January 2016 (UTC)

Formatnum

Currently large numbers in announcements (i.e., at least 1,000) are wrapped in a {{formatnum:}} parser function. I believe I was the one who started this convention (at least, on a large scale). At the time, I was under the mistaken impression that this marked up the number in some special way that users could take advantage of to display numbers differently by setting some preference or using CSS. At some point (I can't remember when) I discovered I was wrong about this. As you can verify by looking at the generated HTML for this comment, numbers processed by "formatnum" (like the "1,000" in the first sentence of this comment) are not marked up in any special way, they just have thousands separators (on this wiki, commas) inserted in them, and thus they turn out exactly the same as numbers containing "manually inserted" commas (like the "1,000" in the sentence you are now reading). So, unless someone objects citing a really good reason, I'm going to stop using "formatnum" in new announcements, and will replace their usage by commas in all previous announcements (at some future date, when I have time). This should lessen server load while having absolutely no affect on the final HTML. (BTW, the only legitimate objection I can think of is that removing "formatnum" will prevent "copying and pasting" the wiki source from this wiki to some other wiki where the thousands separator is different. But I don't know if anyone actually does that, and I'm not sure we need to worry about keeping that easy to do.) - dcljr (talk) 04:57, 26 January 2016 (UTC)

OK, having received no objections, I will implement this plan (removing all instances of {{formatnum:}}) in my next update (i.e., later today). - dcljr (talk) 19:26, 29 January 2016 (UTC)

The actual effect of formatnum is documented at mw:Help:Magic_words. Nemo 21:55, 29 January 2016 (UTC)

Right. (Sorry, readers, for not linking to relevant documentation.) - dcljr (talk) 03:52, 30 January 2016 (UTC)

Page inutile

Cette page de chiffres est sans intérêt, même ridicule en deuxième position du menu gauche. Ce n'est pas ce que j'attends d'une page appelée Actualités ou Wikimedia News.
Inversement, la page Home donne les Wikimedia News ou les Wikimedia Meta-Wiki News, beaucoup plus intéressantes, mais qui ne doivent pas tenir lieu de page d'accueil. TigH (talk) 09:26, 4 March 2016 (UTC)

The above comment seems to be saying that this page (Wikimedia News) is useless, has a misleading title, and is not worthy of being the second link in the sidebar. The title was chosen many years ago; apparently the original plan was to have a mix of different types of announcements, but the actual contents pretty quickly narrowed down to almost exclusively article-count milestones. I'm not sure whether there's a better name that would be short enough to fit in the sidebar. As for the position of the link in the sidebar: where should it be? (Speaking of bad page titles in the sidebar, "Babel" and "Babylon" have always seemed to me to be a pair of hopelessly confusing titles.) Google's translation of the second sentence was pretty rough, but I think it's saying the "news" content transcluded into the Main Page from {{Main Page/WM News}} (see also Goings-on) is more like what users would expect Wikimedia News to be. OK, sure. What are you suggesting we do about that? - dcljr (talk) 08:49, 5 March 2016 (UTC)

Thank you, it's right.

In my mind, I remove "Wikimedia News" from the sidebar. Easy ! After, at another place, whith another name, if you want only and if you find them...

About the second item, it's too much... The Main Page needs a rewrinting : it's an accumulation of links, in confusion, with news. (compare with Grants:IdeaLab/Inspire (more recent) as first example of what I mean). I can't say more, neither do something.

Thanks. TigH (talk) 17:59, 5 March 2016 (UTC)

Cree Wikipedia

@Dcljr: Since March 14 the Cree Wikipedia has always more than 100 articles. --Antonio1952 (talk) 19:21, 2 April 2016 (UTC)

But every time it has passed 100 (7 times in the past year), the most recently created articles have always been in English and/or just nonsense. This includes the time it reached 100 on the 14th (see pages deleted on March 16th). Every time I see a wiki passing a milestone I look at the newest pages to see if it would have reached the miletsone counting only "genuine" looking articles in the language of the wiki. When I did that on the 16th (didn't actually check on the 15th), I determined that it wouldn't. Did you have a different experience? - dcljr (talk) 01:03, 3 April 2016 (UTC)

Since 14 March, h 12.00, the cr.wiki has always more than 100 articles despite the cancellations of articles (see the history of this table). At this moment, it has 108 articles and the latest articles were created one on March 22, one on March 28 and six on April 4 (see this or this page). On March 22, cr.wiki had already 101 articles! --Antonio1952 (talk) 20:22, 4 April 2016 (UTC)

But how many of those articles were legitimate and how many were complete crap and were later deleted? You can't tell that just by looking at the Wikipedias/Table page. You have to consider things more carefully. Look: No main-namespace pages were deleted between 04:05 (UTC) on 13 March 2016 and 15:44 (UTC) on 16 March 2016, right? The wiki passed 100 sometime between 00:00 (UTC) on 14 March 2016 (99 "articles") and 12:00 (UTC) on 14 March 2016 (103 "articles"). Now, I checked the wiki at 00:34:03 on 2016-03-16 and its article count was 104 (consistent with this info). But I looked at the most recently created articles at that time and saw that at least 5 of them were complete crap, created (almost certainly) for the sole purpose of pushing the wiki past 100. This is why I didn't report the "100" milestone at that time. Later that day (15:44 on 2016-03-16), 16 pages were deleted from the main namespace. Now as a regular user, I can't see now what those pages contained, but I know 10 of them were "number" pages, like this (see the contents shown for 5 June 2015 — that's the kind of pages this anon user creates: nonsense pages with links so the pages count toward the article count). Apparently, not all of the pages had links because after the 16 pages were deleted, the wiki only went down to 100 (as of 00:00 on 17 March 2016). So obviously the wiki passed 100 sometime before then right? Well, no, because this deletion shows a nonsense page that was created at 04:05 on 13 March 2016 and deleted at 14:14 on 29 March 2016, so that page shouldn't count toward the March 16 or 17th article count, either. (And this was only one of 15 useless main-namespace pages deleted around that time. How many were present on March 14th and how many counted as articles? I don't know.) So, anyway, this shows that the wiki did not, in fact, have 100 legitimate articles on March 14th. Suffice it to say, determining whether a wiki "legitimately" passed a milestone is only (barely) possible very soon after the supposed milestone happened, and not days or weeks later. So… what date shall we use for Cree passing 100? I checked NewPages at c. 00:50 on 2 April 2016‎, and found that the supposed "100th article" at that time was created on March 22nd. Right now that same article (I think) is showing up as the 101st article. That's because things change as articles are added and deleted. Unless you want to carefully track when specific articles were created and deleted over a month ago, I don't know if you can come up with a better date. I just know that the wiki did not actually have 100 legitimate articles on March 14th. So what other date would you suggest? - dcljr (talk) 05:20, 5 April 2016 (UTC)

In my opinion, the correct date is March 14.

On 14 March between 0.00 and 12.00, the cr.wiki passes from 99 to 103 articles; on 16 March reaches 105 articles; after the massive deletion of March 16 (last deletion at 15.44), it has 100 articles (17/3 h 0.00); in the following days it grows up to 114 articles; after the massive deletion of March 29 (last deletion at 14.19), it has 102 articles (30/3 h 0.00).

Then, in my opinion, it is indifferent whether an article, created before March 14, has been deleted and replaced with another; the important thing is that the articles have been always in a number equal to or greater than 100: from March 14 cr.wiki has always had 100 or more articles. --Antonio1952 (talk) 20:52, 6 April 2016 (UTC)

P.S.: Note the article you have cited was not created on March 13 and deleted on March 29 but was deleted on March 13, created later and deleted again on March 29!

P.P.S: To help us, you can contact the administrator of cr.wiki: Amqui.

"not created on March 13" Oops, yes you are correct. My bad. My point still stands that when a wiki passes a milestone we should not count crap articles that were just created simply to make the wiki pass the milestone. You can't tell that just by looking at the article count. As I said, I checked the recently created articles at the time and there were enough crap articles recently created to not make the milestone legit. But if you simply must have the 14th milestone, very well. It's not worth continuing to argue over. - dcljr (talk) 03:04, 7 April 2016 (UTC)

And now cr: is back under 100. - dcljr (talk) 00:13, 5 November 2016 (UTC)

Sic transit gloria mundi! --Antonio1952 (talk) 20:52, 5 November 2016 (UTC)

en.Wikt has entries in one third of the world's languages, over 2330

Sometime between March (when en.Wiktionary had entries in 1900 languages) and April (we now have entries in 2330+ languages; big diff), en.Wiktionary expanded its coverage to the point that we now have entries in almost exactly one-third of the world's languages, using a conventional estimate that there are 7000 languages spoken in the world.

This includes a small number of constructed and reconstructed languages (see here). On the other hand, I've been steadily adding referenced entries in various natural languages, meaning that the number of those we include is probably higher now than it was at the time of the dump, and it might even be possible to figure out when exactly we reach(ed) the threshold of covering one third of the world's natural languages. -sche (talk) 20:08, 15 April 2016 (UTC)

Wikipedias table updated

I have just updated both the Wikipedias table here and the milestones table at en:Wikipedia:Milestone statistics with some past milestones that never got into the tables at the time, either by mistake or "by design" (meaning the milestones looked bogus). As I've mentioned before, I try to only announce milestones that look "legitimate" and ignore those that are due to vandalism (including articles in the wrong language and patent nonsense), because in such cases the wiki almost always drops below the milestone shortly afterwards when the bogus articles are deleted. Sometimes, though, this drop below the milestone doesn't happen because enough legitimate articles have been created (or enough older articles have been linked, making them count as articles) in the meantime. This means some legit milestones will be missed, on occasion. So, every once in a while I check both tables against the "live" stats to see if any changes need to be made. This is what I just did. The dates I've used reflect either announcements made here (sometimes milestones are announced but the tables are not updated) or the numbers recorded by the Perl script I use to track article counts (which would match, more or less, the dates one could get from the page histories of List of Wikipedias/Table or w:Template:NUMBEROF [when it's being updated]). - dcljr (talk) 10:21, 25 August 2016 (UTC)

Don't forget the extremely rapid growth of the Cebuano Wikipedia as well. Note the difference between bot-generated articles and bogus articles. Johnny Au (talk) 02:32, 19 February 2017 (UTC)

No empty top row in all tables

Despite all the drama about this issue just over a year ago, I've decided I don't like having an empty row showing at the top of every "interlanguage-project" table (Wikipedia, …, Wikivoyage). I propose hiding the empty rows in HTML comments (for all the tables).

At the risk of being pedantic, here's an example of what I'm proposing:

Table of Wikipedias now	Table of Wikipedias under this proposal
Wikipedias by article-count milestone Milestone Languages (dates milestones reached, in chronological order) 6,000,000 5,000,000 English (1 November 2015) 4,000,000	Wikipedias by article-count milestone Milestone Languages (dates milestones reached, in chronological order) 5,000,000 English (1 November 2015) 4,000,000
…	…

Comments? Objections? - dcljr (talk) 04:56, 9 January 2017 (UTC)

OK, I'm going to do it… - dcljr (talk) 02:02, 22 January 2017 (UTC)

Done. - dcljr (talk) 00:52, 25 January 2017 (UTC)

Singular headings for tables

OK, next on my wishlist is replacing the plural section headings for the tables (Wikipedias, Wiktionaries, etc.) with singular headings (Wikipedia, Wiktionary, etc.). The main reason for this is the fact that Wikinews is very awkward to pluralize (Wikinewses? Wikinews'?), so that hasn't been done (unless, of course, you think the plural of Wikinews is Wikinews). Note that this change need not result in any broken incoming links, since we can use anchored <span> tags containing the plural versions. Comments? Objections? - dcljr (talk) 01:53, 25 January 2017 (UTC)

OK, I'm going to do it… - dcljr (talk) 02:20, 18 February 2017 (UTC)

Done. - dcljr (talk) 03:26, 28 February 2017 (UTC)

In case anyone wants to verify that links to both old and new section headings all work as expected:

• Wikimedia News#Wikipedias → Wikimedia News#Wikipedia
• Wikimedia News#Wiktionaries → Wikimedia News#Wiktionary
• Wikimedia News#Wikiquotes → Wikimedia News#Wikiquote
• Wikimedia News#Wikibooks (section was never called "Wikibook")
• Wikimedia News#Wikisources → Wikimedia News#Wikisource
• Wikimedia News#Wikinews (section was never called "Wikinewses" or anything else)
• Wikimedia News#Wikiversities → Wikimedia News#Wikiversity
• Wikimedia News#Wikivoyages → Wikimedia News#Wikivoyage

They do. - dcljr (talk) 03:40, 28 February 2017 (UTC)

Tamil Wikipedia to 100,000

The Tamil Wikipedia has been reached to 100,000 articles, with 51 articles added. —The preceding unsigned comment was added by 2600:387:A:15:0:0:0:97 (talk • contribs) 23 August 2017‎ (UTC)

Already reported on the page (May 8). - dcljr (talk) 02:45, 24 August 2017 (UTC)

Mingrelian Wikipedia to 10,000

The Mingrelian Wikipedia has been reached to 10,000 articles, with 237 articles added. —The preceding unsigned comment was added by 2600:387:A:15:0:0:0:97 (talk • contribs) 23 August 2017‎ (UTC)

Already reported on the page (May 1). - dcljr (talk) 02:45, 24 August 2017 (UTC)

February 2018 updates

Thanks for the February list. I've read some commentary at phabricator:T59788#3980299 but I prefer to keep here my comments on specific cases. It's "normal" that Wikibooks counts were off, since we don't update them regularly. More interesting are the cases of uk.wiktionary, zh.wikiquote and vec.wikipedia: I see a few hundreds of deletions and imports over there, but nothing especially massive. What was the specific cause? --Nemo 14:25, 17 February 2018 (UTC)

Do you mean ur.wiktionary and vec.wikisource? Because uk.wiktionary and vec.wikipedia aren't in the list at Wikimedia News#February 2018. (As for an answer to your question, I don't have one. Beyond reporting the numbers, I haven't investigated any particular cases.) - dcljr (talk) 01:28, 18 February 2018 (UTC)

Abuse filter admins

Seb az86556 has twice changed the wording of the news item about the creation of new "Abuse filter" admins (17 December 2018), both times leaving the impression that the new admins were all given the same English name, "Abuse filter". This is clearly not true, as one can see by checking a few arbitrarily chosen wikis. For example, in the Azerbaijani Wikipedia the admin was called "Dəyişiklik süzgəci", whereas in the Nepali Wiktionary it was called "दुर्व्यवहार फिल्टर". These are the equivalent of "Abuse filter" in the local language of each wiki. So, Seb_az86556, please explain to me how it is "simply not true" that "Almost all Wikimedia content wikis have gotten a new administrator called (the local language equivalent of) 'Abuse filter'". - dcljr (talk) 19:37, 30 December 2018 (UTC)

Cherokee: "Abuse filter"

Gaelic: "Abuse filter"

Maori: "Abuse filter"

Navajo: "Abuse filter"

Somali: "Abuse filter"

Sotho: "Abuse filter"

Tibetan: "Abuse filter"

Uzbek: "Abuse filter"

Yoruba: "Abuse filter"

Zulu: "Abuse filter"

...

Aymara: "Filtro antiabusos"

Guarani: "Filtro antiabusos"

Nahuatl: "Filtro antiabusos"

...

q.e.d.

Seb az86556 (talk) 02:08, 31 December 2018 (UTC)

What you have discovered is that not all languages have translations for that string, in which case they use the English default or that of the fallback language. I still think your version gives the mistaken impression that there should be an admin literally named "Abuse filter" on almost all wikis, which is definitely not true. So, to be absolutely, precisely correct, I have changed the text again. Happy? - dcljr (talk) 03:18, 31 December 2018 (UTC)