Template talk:Wikipedia languages/Archive 7

Archive 1

←

Archive 5

Archive 6

Archive 7

Archive 8

ur.wiki

Urdu Wikipedia has passed 70,000 mark but not included in the list. Please update. --Tahir Mahmood (talk) 07:26, 17 June 2015 (UTC)

As noted in the template's documentation, "this is not a complete list of Wikipedias containing 50,000 or more articles; Wikipedias determined to consist primarily of stubs and placeholders are omitted." —David Levy 15:56, 18 June 2015 (UTC)

Danish Wikipedia has more than 200,000 articles

Please update the front page at the bottom. --94.189.38.110 (talk) 13:26, 18 June 2015 (UTC)

I'll copy this to Template talk:Wikipedia languages, where it belongs. Art LaPella (talk) 13:50, 18 June 2015 (UTC)

Done. —David Levy 15:56, 18 June 2015 (UTC)

The same goes for the Bulgarian Wikipedia.--Givern (talk) 06:32, 23 June 2015 (UTC)

Proposal: Capitalize first letter of all language names

They are currently written in sentence case of the language itself. This gives a weird mix of capitalizations like "Deutsch · español · français · italiano · Nederlands · polski · русский · svenska". It looks ugly together and many readers will probably think it's an error. Language names are capitalized in English and although the names are written in the respective languages, it seems more natural to capitalize them all in the English Wikipedia, especially when they are all capitalized in the language list to the left of the main page and other pages. The first letter is also capitalized at https://www.wikipedia.org and most foreign Wikipedias under "Languages" at {{Wikipedia languages}}. We capitalize the first letter in many similar bulleted lists in navboxes so there would be nothing odd about doing it here. The implementation would be a trivial addition of {{ucfirst:}} in a single place in {{Wikipedia languages/core}}. PrimeHunter (talk) 22:44, 5 July 2015 (UTC)
Note: I started this discussion after seeing Talk:Main Page#Wikipedia languages. PrimeHunter (talk) 22:46, 5 July 2015 (UTC)

• Agree with capitalizing.--Jim in Georgia Contribs Talk 23:00, 5 July 2015 (UTC)
• Oppose. Years ago, we decided to list the links in accordance with the individual languages' conventions (not those of English), reflecting the fact that the section was created mainly to serve said languages' readers. In the context of a list intrinsically comprising a "mix" of dissimilar languages, I see nothing "weird" or "ugly" about applying the same standard to the capitalization.
  Unlike "the language list to the left of the main page and other pages", the language names are not bulleted (only the four tiers are, and each does begin with an uppercase letter), nor is a style employed in the English language contextually relevant. Attempting to make the text look like English actually strikes me as counterproductive, given the inherent impossibility of succeeding. Many of the items don't even begin with Latin characters, so what uniformity would this change achieve? It would merely cater to (some) native readers of English via a superficial resemblance thereto and the illusion of consistency (because, after all, who cares about those funny squiggles?) within a template that exists primarily for other readers' benefit. I don't know what drives the various Wikipedias' formatting decisions, but that certainly isn't consistent with our longstanding approach. —David Levy 00:28, 6 July 2015 (UTC)

  As discussed below, I find PrimeHunter's argument persuasive. Accordingly, I'm now neutral (and might decide to support the proposal). —David Levy 13:50, 6 July 2015 (UTC)

  The first letter was upper case from 2001 [1] until 18 June 2012 where gerrit:7306 changed the behaviour of the used magic word (compare 18 June and 19 June). The switch to lower case was an automatic consequence of that and not the result of a decision about the main page language lists. Krinkle changed it to upper case in January 2013 [2] and you changed it to lower case 24 days later.[3] Has there been a discussion which decided to accept the automatic change to lower case? PrimeHunter (talk) 01:59, 6 July 2015 (UTC)

  I didn't mean to imply that the lowercase formatting was present from the beginning or adopted as the result of a decision about the section; I noted the behavior's addition to MediaWiki (and a couple of issues that arose as a result) in the discussion to which you linked in your proposal. My point, as stated above, is that it makes sense (in my opinion) to "[apply] the same standard to the capitalization" that we've long applied to the list's formatting in general. For most of the template's existence, we've adhered to the principle that it's intended to function primarily as a resource for readers of the languages listed and should be tailored accordingly (not for optimal accessibility in English, which we've consciously reduced for the sake of improving its accessibility in the other languages).
  I don't recall the idea of adjusting the template's capitalization arising before the MediaWiki improvement occurred, nor am I aware of any conversation on the matter of how to handle the MediaWiki update occurring before yesterday. The technical change certainly was discussed, and its impact on this section was covered, but I don't think that any sort of proposal or debate materialized. Much more attention was focused on the sidebar's language links, for which unforeseen side effects generated a good deal of concern. When it came to this template, conversely, any "discussion which decided to accept the automatic change to lower case" amounted to little more than the following exchange:
  

  "Hey, I just noticed that some of the language links at the bottom of the main page have become lowercase. What's that about?"
  "The software was just updated to use each language's native formatting of its name. If one appears in lowercase, that means that it's customary in that language. The rest are still capitalized."
  "Oh, okay. Thanks for the explanation."

  Krinkle's edit and mine, respectively, constituted the "B" and "R" in "BRD". Had someone pursued it, "D" would have been a discussion along the lines of this one, presumably. —David Levy 03:11, 6 July 2015 (UTC)

• Oppose A solution in search of a problem. David Levy explains why the current convention works, no impending need to change to another arbitrary convention. --Jayron32 02:58, 6 July 2015 (UTC)
• Support If you look at es.wiki or other languages where in the native language all languages names are lower case, even they capitalize every entry in their list of Wikipedias.[4][5] I think our approach is quite foolish. In bulleted navboxes throughout en.wiki, we use capital letters for every navbox entry (even when some are proper nouns and others are not). There is no reason not to use the same approach here. Calliopejen1 (talk) 05:42, 6 July 2015 (UTC)

  If you look at es.wiki or other languages where in the native language all languages names are lower case, even they capitalize every entry in their list of Wikipedias.
  Do you possess first-hand knowledge of why other Wikipedias use that format, or have you simply assumed that it reflects a particular convention of equal relevance here (despite the existence of significant differences between English and any of those languages)? Various Wikipedias do all sorts of things differently, on their main pages and elsewhere. You've cited a particular practice resembling one that you wish to implement at the English Wikipedia, thereby ignoring countless other differences between those Wikipedias and this one.
  For all we know, many Wikipedias may have simply copied the formatting used here (before we began basing the template on language codes, thereby enabling the subsequent change), as they often do (and we copy some of them too, of course). This might be a situation analogous to one in which a source cited in a Wikipedia article based its information on claims that originated in an earlier, unsourced revision of the very same Wikipedia article.
  Of course, I don't know that to be the case. It's sheer speculation, so I don't assert that it carries any weight. We should base our decision on the merits, not on what other Wikipedias do (for whatever reasons they do it).
  In bulleted navboxes throughout en.wiki, we use capital letters for every navbox entry (even when some are proper nouns and others are not).
  Firstly, there appears to be continuing confusion regarding the meaning of "bulleted". Again, only the template's four tiers (the text of which does begin in uppercase) are bulleted.
  Secondly, content from an English Wikipedia article differs fundamentally from that of this template (which, by its very nature, contains a combination of items that neither can nor should be presented in a homogeneous manner.) I don't understand the desire to manufacture pseudo-uniformity across an assortment of items inherently written in entirely different languages. It's like affixing blue ribbons to an apple and an orange in an attempt to pass them off as two of the same fruit.
  And again, many of the language names aren't written with the Latin alphabet, so we can't even pretend to harmonize them. Under our current format, each language name is presented in its native format. This maximizes their consistency. Under the proposed format, some language names would be altered purely because of their superficial similarity to others (as perceived by native readers of English, the users for whom the section is least useful). Meanwhile, a bunch of other language names will remain untouched because they differ more, which prevents us from even pretending to apply the same treatment to them. So what will we have accomplished? (So much for consistency.) —David Levy 06:40, 6 July 2015 (UTC)

  I'm Danish and find "Dansk" in an upper case list easier to locate than "dansk" in a mixed list. The mix gives a confusing impression that the list may consist of segments starting at each capital. And after seeing some capitalizations I don't know whether to look for dansk or Dansk. I suspect most people across languages would expect their language to be upper case in a language selection list. The Google search español français italiano português (four languages we write lower case) shows it is much more common to use upper case, also for horizontal lists. Maybe bulleted list is the wrong term for a horizontal list with terms separated by bullets/interpuncts/whatever, but we do generally use upper case on such lists. The main page itself has many and all except the languages use upper case. While each language is intended for people who know that language, this is still the English Wikipedia and most readers of the language lists will know few or none of the languages. And while there is logic in both systems, the logic in mixed capitalization is hard to guess. Lots of readers will think we goofed up on our main page and fair or not, it can easily affect their impression of us. When I see a text with poor capitalization, it negatively affects my view of the author. PrimeHunter (talk) 12:06, 6 July 2015 (UTC)

  The dots separating the languages are, indeed, interpuncts. A bullet is a different mark, typically used at the beginning of a text line (not between items appearing thereon).
  While anecdotal, your explanation of why you find "Dansk" in an uppercase list easier to locate than "dansk" in a mixed list certainly gives me pause for thought. If most native readers of the languages listed (or those that use the Latin alphabet, really) share this view, I'd be inclined to reverse my position. Perhaps we could solicit input from such individuals. —David Levy 12:38, 6 July 2015 (UTC)

  Since this is the English Wikipedia and not our common entry point https://www.wikipedia.org, input from native English readers who sometimes visit other languages would also be relevant. PrimeHunter (talk) 13:13, 6 July 2015 (UTC)

  Indeed, all good-faith input is desirable. Historically, we've prioritized the template's utility among native readers of the languages listed (to the extent that we relegated their English names to the hover text, thereby reducing clutter and freeing up space), but accessibility to native English readers certainly isn't unimportant. —David Levy 13:50, 6 July 2015 (UTC)

• Support I stand by my original point from 2013 (see Revision #530893992). The rules about language name capitalisation apply to contexts where the name is used as a word. For example "View page in English" would be something like "Leggi in italiano" in Italian. However that doesn't change the fact that a title starts with a capital letter (sentences, list items, buttons, tabs, etc.). That is the case in Italian as well. This template doesn't produce a natural language sentence that happens to include some language names. Instead, these are clearly presented as independent entities for linking purposes. Each link is essentially a button. Title casing is therefore appropriate and expected. --Krinkle (talk) 11:13, 6 July 2015 (UTC)
• Support this change. Krinkle is absolutely correct; the horizontal list is akin to a bulleted list (just visually laid out slightly differently) and should be formatted the same way. — This, that and the other (talk) 13:36, 6 July 2015 (UTC)
• Note: Given the clear consensus to restore the capitalization, I've self-reverted its removal. The discussion can continue, of course, but a shift in the prevailing view seems highly unlikely. —David Levy 14:56, 7 July 2015 (UTC)

Vietnamese Wikipedia

Hi, please update this template because vietnamese Wikipedia has already over 1m articles. Na Tra (talk) 10:43, 29 July 2015 (UTC)

Done. —David Levy 15:49, 29 July 2015 (UTC)

turkish wikipedia now have 250.446 article please update this

turkish wikipedia now have 250.446 article please update this.--Koroğlu (talk) 17:59, 13 August 2015 (UTC)

Done. —David Levy 23:29, 13 August 2015 (UTC)

Urdu Wikipedia

Urdu Wikipedia is not included but some other wikipedias which have rather less articles, edits, active users and depth are there. Is that fair? Another reason is that most of Urdu speakers (in Pakistan and India) even don't know that a wikipedia in Urdu language exists that's why they mostly use English wikipedia, by adding Urdu Wikipedia's link many of these people will know about Urdu Wikipedia and it might be helpful for Urdu Wikipedia that will possibly get more readers and volunteers. Thanks. — Preceding unsigned comment added by 119.160.118.133 (talk) 09:12, 15 August 2015 (UTC)

We stopped relying on the "depth" calculations, which often are artificiality inflated (not necessarily on purpose) via edits performed by multiple bots. To see what I mean, view the edit history of this placeholder article, the first random page that I viewed. Unfortunately, it appears that much of the Urdu Wikipedia is in a similar state.

Your point regarding the attention that a link would provide is valid, but it applies to numerous Wikipedias (many of which match or exceed the Urdu Wikipedia's quality, even if the absence of mostly-empty pages has resulted in a lower article count). If we're to maintain any sort of qualitative criteria, it isn't feasible to include them.

In the past, there's been some discussion of creating a new main page section to showcase a a rotating selection of sister sites (other languages' Wikipedias and/or other Wikimedia Foundation projects in English). If this occurs, perhaps we could cover Wikipedias omitted from our regular list. —David Levy 14:48, 15 August 2015 (UTC)

You are somewhat right but i think it is better to have some South Asian languages on en wikipedia's main page after all most of people there are anglophones, like Spanish wikipedia have links of those languages whose speakers speak Spanish as second language even quality of those is inadequate. — Preceding unsigned comment added by 119.160.116.208 (talk) 04:57, 16 August 2015 (UTC)

Request to differentiate Traditional and Simplified Chinese pages

Hey, Wikipedia newbie here.

So, I'm a Chinese language student in China (PRC) and have been studying for 2 years here.

Unfortunately, when I go on a lot of Wikipedia pages in Chinese, the pages are almost always a co-mingled messy conglomeration of Traditional and Simplified characters. Sometimes this is not a problem as 60% or so of the character sets are 100% the same like in the case of some basic pronouns and most numbers. Sometimes this is only a small problem like in the case of 国/國 and 说/說, which mean "country" and "say," respectively. But there are MANY other cases where I have no idea what it is trying to say. To show you how different Simplified and Traditional can be, here is the not-as-uncommon-as-you-might-think characters for "dragon": 龙/龍--there's just no way to guess something like that, and there are a lot of cases like that.

To find page after page after page with co-mingled alphabets like that is kind of ridiculous, and at the very least appears extremely unprofessional in Chinese. Though most of the the savvy Chinese internet users who regularly use Wikipedia almost certainly have a strong understanding of both Simplified and Traditional, I'm sure there are a lot of people like me who only know one character set and thus are really put off from using Wikipedia in Chinese because of the annoyance and difficulty.

Many websites that cater to both Mainland and non-Mainland users tend to have an option to change every webpage to Traditional (繁体) or Simplified (简体) without difficulty. Thus, I propose that we split the Chinese pages into two sets--one in Traditional and one in Simplified, but since there is one-to-one correspondence between characters, every time one page in one character set is altered, the opposite respective page in its own character set receives that same change.

China (PRC) only blocks some parts of Wikipedia but not all of Wikipedia--in fact I'm not using my VPN right now, which demonstrates my point. China is the largest country in the world and speaks one of the most common languages in the world. To not cater to this group of potential users would be a silly mistake if it can be easily fixed.

The one problem for me is that I have no programming background so I have no clue how easy this would be in practice. So I throw this subject to you. Thanks for reading! :-) — Preceding unsigned comment added by Sophyphreak (talk • contribs) 21:53, 17 November 2015 (UTC)

This is the talk page for the template that produces the "other Wikipedias" footer seen on a number of pages. It is not for general discussions of Wikipedia. For the *English* Wikipedia, such a question should probably go on Wikipedia:Help desk or on Wikipedia:Village pump. But those are only for the English Wikipedia, the other languages projects are largely independent, and set their own rules for content and such, and the policies here (en-wp) don't apply. So you should probably ask at the Chinese Wikipedia ([6]) help desk, which is (I think!) at [7]. Note there are also existing Wikipedia's for classical Chinese ([8]), Cantonese ([9]) and Wu Chinese ([10]), which presumably have their own help desks (although these seem rather less active places than the main Chinese Wikipedia). Rwessel (talk) 22:50, 17 November 2015 (UTC)

@Sophyphreak: Extremely late, but did you try the dropdown options that you get if you click the tab "不转换" (which maybe could be more clearly labelled) at the top of each page? It then gives you the choice which variant you want to view articles. (And I'm pretty sure this option was already present before 2015.) In Monobook skin, you get all the tabs without the dropdown and can just click the one you want. Double sharp (talk) 08:51, 15 January 2020 (UTC)

Languages sorted by views

I know it's not our usual criteria but I sorted Wikipedia languages by views per hour in November 2015 based on https://stats.wikimedia.org/EN/Sitemap.htm. Languages with more than 1000 views are listed. This includes all languages currently in {{Wikipedia languages}}. The last column marks them.

Code	Language	Views/h	Articles
en	English	6,067,067	5,022,395	(X)
de	German	876,241	1,881,065	X
ru	Russian	871,999	1,271,240	X
es	Spanish	726,624	1,175,371	X
ja	Japanese	717,454	993,516	X
fr	French	603,875	1,700,734	X
it	Italian	303,440	1,238,144	X
zh	Chinese	286,205	849,936	X
pl	Polish	241,524	1,146,140	X
pt	Portuguese	226,592	895,796	X
nl	Dutch	120,952	1,844,444	X
tr	Turkish	95,447	256,816	X
ar	Arabic	76,080	393,073	X
sv	Swedish	69,682	2,158,119	X
cs	Czech	66,518	338,592	X
id	Indonesian	58,972	371,991	X
fa	Persian	47,607	475,210	X
ko	Korean	45,253	333,780	X
uk	Ukrainian	44,973	603,799	X
fi	Finnish	41,935	383,582	X
hu	Hungarian	36,892	375,807	X
vi	Vietnamese	34,747	1,141,303	X
he	Hebrew	31,024	181,336	X
th	Thai	27,895	98,896	X
no	Norwegian	25,635	425,490	X
ro	Romanian	23,845	353,084	X
da	Danish	22,520	211,716	X
el	Greek	19,718	112,141	X
bg	Bulgarian	16,911	210,401	X
sr	Serbian	16,286	328,968	X
ca	Catalan	15,779	486,702	X
hr	Croatian	13,563	146,471	X
kk	Kazakh	13,081	218,472
sk	Slovak	12,121	207,376	X
lt	Lithuanian	8,719	174,944	X
simple	Simple English	8,581	116,115	X
et	Estonian	8,177	139,925	X
sl	Slovene	5,752	148,447	X
ka	Georgian	5,536	101,406
sh	Serbo-Croatian	5,454	429,133	X
hi	Hindi	5,204	102,808
tl	Tagalog	4,896	64,438
az	Azeri	4,724	101,999
ms	Malay	4,200	279,784	X
lv	Latvian	4,141	64,319	X
hy	Armenian	3,527	191,108
bs	Bosnian	2,862	65,866	X
sq	Albanian	2,732	56,841
zh-yue	Cantonese	2,342	41,248
mk	Macedonian	2,100	84,532
eu	Basque	2,068	216,901	X
ta	Tamil	2,052	83,352
gl	Galician	1,847	126,194	X
eo	Esperanto	1,774	222,366	X
mr	Marathi	1,730	43,569
arz	Egyptian Arabic	1,719	14,527
bn	Bengali	1,700	39,437
ceb	Cebuano	1,575	1,435,540
be	Belarusian	1,354	109,683
ur	Urdu	1,336	94,219
nn	Nynorsk	1,315	124,430	X
te	Telugu	1,210	63,827
ml	Malayalam	1,179	41,361
is	Icelandic	1,173	40,138
af	Afrikaans	1,162	37,800
mn	Mongolian	1,068	17,125

Should we change something to help readers find what they may be looking for? For example, Kazakh is not included but has more articles and 10 times as many viewers as Nynorsk. But Norwegians are more likely than Kazakhs to view the English Wikipedia. Note the main Norwegian variant no is not under discussion. There is a related discussion with a shorter table at Template talk:Wikipedia languages/Archive 6#Further discussion. PrimeHunter (talk) 00:22, 16 December 2015 (UTC)

It's a very gameable criteria (people can easily write a bot to constantly spam page lookups on a specific Wikipedia), but it might be useful to detect outliers if there's cause to think the numbers aren't being gamed. I think adding Kazakh (as an outlier in a weirdly high number of hits) might be reasonable despite the quality problems, and same with removing Nynorsk - the reason Nynorsk is as high on article count as it is is probably related to the ease of taking Bokmal articles and sticking them in Nynorsk, even if few readers care. Everything else can probably stay, though. My two cents. SnowFire (talk) 21:31, 17 January 2016 (UTC)

Japanese WP

The Japanese Wikipedia version now exceeds 1,000,000 articles. Congratulations --Thomas Wozniak (talk) 11:43, 20 January 2016 (UTC)

The Japanese WP is now under 1M again. Seems like they just deleted a thousand or so articles. --Portalian 15:04, 25 January 2016 (UTC)

Protected edit request on 25 April 2016

---

Change:

| This Wikipedia is written in [[English language|English]]. Started in {{Start date|2001}}, it currently contains [[Special:Statistics|{{NUMBEROFARTICLES}}]] articles. }}
Many other Wikipedias are available; some of the largest are listed below.

To:

| This Wikipedia is written in [[English language|English]]. Started in {{Start date|2001}}, it currently contains [[Special:Statistics|{{NUMBEROFARTICLES}}]] articles. }}
Hundreds of other Wikipedias are available; the largest are listed below:

---

Rationale: There are 272 other Wikipedias at this time. "Many" is not descriptive, while "Hundreds" is more accurate. "some of the" is a weasel phrase. "the" is more direct. Thank you. Ping me back. Cheers! {{u|Checkingfax}} {Talk} 04:46, 26 April 2016 (UTC)

At least the last, "the largest are listed below", would simply be incorrect. Something of a judgment of useful content is made, and Wikis like Cebuano (3rd biggest by number of articles) and Waray-Waray (#9), which consist mostly of bot-created stubs are omitted from the list presented here. IMO (and I've lost that argument before), the Swedish and Dutch Wikis should be removed for the same reason - the counter argument is that both of those have a core of a few hundred thousand actual articles (despite the vast majority of the ~3 and 1.8 million articles being bot-created stubs).

I also have a personal dislike for using "hundreds" to mean anything less than three hundred, but that might just be me. Rwessel (talk) 05:52, 26 April 2016 (UTC)

I tend to agree that the current message is better than the proposed message. I have disabled the request due to lack of consensus. Perhaps "Nearly 300 other Wikipedias are available" would work? — Martin (MSGJ · talk) 08:27, 26 April 2016 (UTC)

Armenian in the main list

I am not a wikipedia editor, neither have any statistics other than article count. Armenian wikipedia has almost 200k articles counted now, I suppose they'll hit the number soon. You can watch my IP and make legitimate assumptions (like my opinion might be biased, alright), but maybe Armenian wikipedia could be displayed in the main languages list, to show some love to the editors and contributors.46.241.223.86 (talk) 12:13, 15 May 2016 (UTC)

Euskara passes 250,000

Euskara (Basque) has passed 250,000 articles.[11] PrimeHunter (talk) 21:06, 24 June 2016 (UTC)

 Done — xaosflux ^Talk 17:24, 25 June 2016 (UTC)

کتولی

زبان کتولی ادامه زبان طبری در منطقه علی آباد کتول است. — Preceding unsigned comment added by S87208 (talk • contribs) 08:23, 28 November 2016 (UTC)

Please make a 500 thousand category in the https://en.wikipedia.org/wiki/Main_Page#Wikipedia_languages

Hi,

I noticed that the quantity-based categorization of languages on the bottom of the main page contains a one million category, a 250 thousand category, and a 50 thousand category.
Since some of the languages categorized as having more than 250 thousand pages have now more than 500 thousand, including Persian, Arabic, and Portuguese (nearing one million), and considering that there is a wide gap between 250 thousand and one million, it would be appropriate to add a 500 thousand category for listing these languages. This would encourage language diversity and add to Wikipedia's common knowledge and neutrality (As an Iranian, it would make me proud too :).
Thanks in advance.
Love,
Alireza1357 (talk) 08:29, 5 April 2017 (UTC)

Update number of Wikipedias above 1 million articles

Hi, the Meta list has 13 Wikipedias above 1,000,000 articles, but this template only has 10. Also, the number of Wikipedias above 50,000 and below 250,000 is actually 46, but this template only lists 16. The number between 250,000 and 1,000,000 looks fine. Just thought that should be updated. Icebob99 (talk) 23:49, 15 March 2017 (UTC)

@Icebob99: It says "some of the largest are listed below". Here "some" is a subtle hint that not all the largest are listed. Some are deliberately omitted due to low quality determined by random sampling. The talk archives mention testing 50 random articles. Some languages mainly consist of bot-generated stubs. I once tested 50 random page histories in Cebuano without finding a single human edit. Nearly the whole wiki is made by a bot by a Swedish editor (I think his wife speaks the language). The four million Cebuano articles get on average around two yearly page views so very few people probably miss the language from our main page. I think https://www.wikipedia.org uses page views to choose the languages around the Wikipedia globe. PrimeHunter (talk) 21:26, 8 April 2017 (UTC)

@PrimeHunter: thanks for the clarification... kinda makes sense that way. Icebob99 (talk) 23:01, 9 April 2017 (UTC)

Strong instead of b edit request

Sorry for the very minor change, but <strong> instead of <b> is appropriate for the "complete list" link, for semantic importance relative to the lists. Change is in this /sandbox edit (diff). No visual change in most browsers, so probably uncontroversial. Matt Fitzpatrick (talk) 23:23, 25 April 2017 (UTC)

 Done — xaosflux ^Talk 02:16, 9 May 2017 (UTC)

BENGALI WIKIPEDIA

bengali wikipedia achieve 5,000 article , on 30 April 2017 . It should be added.TNX. -- md masum (talk) Wikipedia:Milestones

You mean 50,000 articles. It does look like it should be added. PrimeHunter (talk) 22:16, 10 May 2017 (UTC)

Might be the wrong place

but can this article:https://fr.m.wikipedia.org/wiki/Islam_(parti_politique_belge) be translates from Dutch to English? Knightofjustice123 (talk) 03:09, 3 July 2017 (UTC)

Adding Tamil interwiki link

Since Tamil wikipedia have surpassed 100000 articles, I propose that it should be added to the template.--Prash (talk) 14:35, 28 August 2017 (UTC)

Remove spurious bold (''')

On the view source page for Template:Wikipedia languages, clicking on "Submit an edit request" spuriously links to Wikipedia:Main Page/Errors, rather than here. Please fix this error, or explain why it is this way. And now to the main request:

In Template:Wikipedia languages, there is one (1) unpaired bold markup ('''), causing a Missing end tag lint error. Please remove it. —Anomalocaris (talk) 01:11, 10 April 2018 (UTC)

Done.[12] "Submit an edit request" links to Wikipedia:Main Page/Errors due to code for cascading protection of templates transcluded on Main Page. It's the same for the other templates there. The idea is that main page problems should be fixed quickly and many administrators may watch Wikipedia:Main Page/Errors. PrimeHunter (talk) 01:44, 10 April 2018 (UTC)

PrimeHunter: Thanks! —Anomalocaris (talk) 20:12, 10 April 2018 (UTC)

Request for Template Updating

Since Chinese has surpassed 1000000 articles (which is a very big milestone), please update this template. Sadiosdo (talk) 08:55, 19 April 2018 (UTC)

Please update the template to reflect that Chinese (中文) Wikipedia now has over 1,000,000 articles. Double sharp (talk) 01:51, 24 April 2018 (UTC)

Was done by Rmhermen. — xaosflux ^Talk 12:49, 24 April 2018 (UTC)

Started in 2001

"Started in 2001" has nothing to do with languages. I propose changing This Wikipedia is written in English. Started in 2001, it currently contains 5,646,153 articles. to This Wikipedia is written in English and currently contains 5,646,153 articles. wumbolo ^^^ 18:32, 9 May 2018 (UTC)

Request for Template Updating

Malayalam wikipedia (ml.wikipedia.org) currently has more than 50,000 articles. Please update this template. Adithyak1997 (talk) 18:07, 17 June 2018 (UTC)

 Done Ruslik_Zero 20:30, 18 June 2018 (UTC)

Cebuano wiki

The Cebuano language Wikipedia currently has the second most articles (5,382,840). Shouldn't it be added to the template in the 1 million+ category? Stoiccowls (talk) 19:30, 13 June 2018 (UTC)

No, the quality is too poor. Quoting my post at Template talk:Wikipedia languages/Archive 7#Update number of Wikipedias above 1 million articles: It says "some of the largest are listed below". Here "some" is a subtle hint that not all the largest are listed. Some are deliberately omitted due to low quality determined by random sampling. The talk archives mention testing 50 random articles. Some languages mainly consist of bot-generated stubs. I once tested 50 random page histories in Cebuano without finding a single human edit. Nearly the whole wiki is made by a bot by a Swedish editor (I think his wife speaks the language). The four million Cebuano articles get on average around two yearly page views so very few people probably miss the language from our main page. PrimeHunter (talk) 20:53, 13 June 2018 (UTC)

That sounds reasonable enough but there should be a clearly stated standard that is evenly applied. I did the same random sample test on the Swedish Wikipedia and 45/50 were created by bots, including 42 by the same bot writing the Cebuano Wikipedia (Lsjbot). Only 8 articles had any human edits and only 4 had more than 1 human edit. If we're discounting bot-written stubs it doesn't seem like Swedish should come close to qualifying for the 1 million+ category. Stoiccowls (talk) 14:13, 14 June 2018 (UTC)

If we do list a langauge then we always use the total number of articles including bot-generated stubs. Giving a false count would be odd. Swedish has an inflated article count but without the bots they would have enough quality content to qualify for the list as "More than 50,000 articles". I think it's OK to include Swedish and use their actual count. It's isn't always practical to give a clearly stated standard. Editors constantly make judgments about all sorts of things. PrimeHunter (talk) 23:49, 14 June 2018 (UTC)

A clear standard is not incompatible with a subjective one. Most parts of the main page have guidelines that don't get buried in talk archives; these guidelines leave room for admins to use judgment but they provide some framework and motivation for those judgments. It's not clear from this template or talk page that any standards exist and you're not doing a great job of explaining your logic either. On one hand you agree that bot-written articles should count the same as other valid article, but on the other hand you disqualify entire wikis from appearing because they have "too many" of these articles. From what I can tell the Cebuano wiki has over five million articles that are accurate, properly cited, and notable. Nearly all are stubs but almost none would qualify for rapid deletion. It doesn't appear that it was created in bad faith or in an attempt to game the system. What encyclopedic goal is served by excluding it from the main page? Stoiccowls (talk) 14:00, 18 June 2018 (UTC)

I think the main purpose of the language list is guiding readers to a wiki they may be interested in. Almost nobody reads the Cebuano Wikipedia, and I think nearly all of the articles are of near-zero interest to people who can actually read the language. It's spoken in the Philippines, called Pilipinas in Cebuano. The word "Pilipinas" occurs in less than 1% of the articles. That's what happens when a bot creates millions of stubs about whatever is in some databases unrelated to the country and language, e.g. insect species from other continents. For comparison, "United States" occurs in 21% of articles in the English Wikipedia. The featured languages at https://www.wikipedia.org/ are determined by page views of the wiki. If you want a clear standard then that might be an idea. PrimeHunter (talk) 22:11, 19 June 2018 (UTC)

United States is a large important country but we have readers and editors from around the World. Let me give the example of my own country Denmark, called "Danmark" in Danish. We only have 5.8 million people but "Danmark" occurs in 35% of articles in the Danish Wikipedia, because it's written by humans who know the language and write about what is of interest to Danes. PrimeHunter (talk) 22:20, 19 June 2018 (UTC)

While I don't agree with these standards, I'd respect them if they were based on consensus and being applied to each applicant. After reading through the talk archives, this is clearly not the case. 5/5 of the 50K+ Wikipedias that I've checked are between 85-95% stubs per a 50 article random sample, several were bloated with bot content, and at least two of them represent languages that functionally don't have native speakers. None of them would pass your implied test of having 50K+ articles that exceed the typical quality of an Lsjbot stub. These issues were only addressed before an approval twice, in both cases because the submitter self-reported them. Pageviews, which you are now citing as a standard, were only suggested as a criteria once and the proposal was promptly tabled because pageviews are extremely game-able. This scheme of searching for the native country appears to be something you just created now and suffers the same game-ability issue. As an outsider, it looks the rules are being invented and applied arbitrarily and that multiple well-sourced attempts to point this out have failed to produce any improvement. I'd recommend the active admins here work with the community to create some guidelines and procedures, even if they're highly subjective, because otherwise you can expect this conversation to keep happening every time the last one gets archived. Stoiccowls (talk) 20:24, 20 June 2018 (UTC)

There will always be people who complain about something no matter how standardized the rules are, if those rules just happen to not give the result they want. Discussions at the English Wikipedia are full of examples of that. But this talk page actually has low activity. The discussion you suggest might require more time than we spend in years, and I doubt it would reduce future debates. I actually think it would increase them because more people would become aware of this page and wish to participate in discussions. I think page views should matter but respect that others disagree. The Danish Wikipedia has 23 times more page views than Cebuano,[13] and 512 times more views per article. But we already list Danish and not Cebuano so I have nothing to complain about there. PrimeHunter (talk) 22:01, 20 June 2018 (UTC)

I wasn't suggesting you write a policy to save time, I was suggesting you do so to avoid being repeatedly and rightly accused of having an arbitrary, uneven, unclear process that fails its stated quality control objectives. I'm surprised that your goal is to avoid interest and discussion of this template, that doesn't seem very Wikipedian. It's true that people will always complain about everything you do but that's not a justification for not trying to do the right thing. If this template is actually worth keeping on the main page, it's worth administering properly.

On a side note, the issue with game-able criteria is that they stop working when you start applying them. If you start thresholding page views today, someone will write bots to make billions of hits on the "bad" pages tomorrow and then you'll be backtracking about how gee, page views aren't meaningful after all. That happened with depth years ago; a depth of 5 requirement screened out the "problem" Wikipedias, but then one of them modified their bots to do tiny edits until it passed the threshold easily and the requirement had to be scrapped. This is why criteria should be discussed before admins start applying them. Stoiccowls (talk) 15:44, 22 June 2018 (UTC)

Esperanto Wikipedia

The Esperanto Wikipedia recently surpassed 250,000 articles. Can we update the template, please? —LLarson (said & done) 14:23, 22 September 2018 (UTC)

 Done Fish+Karate 12:22, 24 September 2018 (UTC)

Request for Template Updating

Armenian wikipedia (hy.wikipedia.org) currently has more than 250.000 articles. Please update this template.--GeoO (talk) 14:40, 8 January 2019 (UTC)

Needs an update

Bulgarian Wikipedia currently has more than 250 000 articles. The template shows less. — Preceding unsigned comment added by 94.156.160.114 (talk) 16:56, 3 March 2019 (UTC)

Done.[14] PrimeHunter (talk) 18:26, 3 March 2019 (UTC)

How many lines, how many entries?

As the various wikipedias grow, the number of wikipedias eligible for entry in our current 3 lines (More than 1,000,000 articles, More than 250,000 articles, More than 50,000 articles) will continue to grow. Right now for the latter two definitions, we have more wikipedias than we can fit on one line on an average screen resolution. Perhaps we should decide the minimum and maximum wikipedias per line and then modify the criteria accordingly (at appropriate milestones). What do others think? Greenshed (talk) 15:55, 22 February 2019 (UTC)

The criteria are already adjusted sometimes. I don't think we should set a fixed minimum and maximum count of languages. There are other factors to consider like how round the thresholds are and the length of language names. "English (simple form)" and "ไทย" are currently in the same group but use very different space. PrimeHunter (talk) 22:53, 22 February 2019 (UTC)

Agree that there are other factors to consider (as hinted at in my original post). Take your point that the length of language names does vary. The bit that all seems rather vague is when the criteria are adjusted what is the basis for said adjustment? Greenshed (talk) 23:14, 25 February 2019 (UTC)

It's based on a specific discussion here or an administrator boldly doing it. Changes are years apart. The current thresholds are from 2015.[15] PrimeHunter (talk) 23:26, 25 February 2019 (UTC)

I would offer that we should have between 5 and 20 entries per line and adjust the thresholds to allow for round numbers. Greenshed (talk) 20:59, 13 March 2019 (UTC)

Macedonian Wikipedia

The Macedonian Wikipedia now has more than 100,000 articles, which is two times the minimum threshold for inclusion. Could any of the admins add it to the list? Thanks.--Kiril Simeonovski (talk) 06:53, 2 May 2019 (UTC)

It was rejected at Template talk:Wikipedia languages/Archive 6#Macedonian Wikipedia and Template talk:Wikipedia languages/Archive 5#Macedonian Wikipedia. I haven't made a new evaluation. PrimeHunter (talk) 09:53, 2 May 2019 (UTC)

@PrimeHunter: Those two discussions were made more than five years ago when the project had 40,000 and 73,000 articles. Importantly, there was no evaluation in the second discussion and it was rejected outright because of the first one that pre-dated it for more than three years.--Kiril Simeonovski (talk) 11:11, 2 May 2019 (UTC)

Yea, this definitely warrants a new evaluation. --Local hero ^talk 16:35, 10 May 2019 (UTC)

Edit request 17 June 2019

Moved from Template talk:Wikipedia languages/core

 – * Pppery * _{it has begun...} 14:13, 19 June 2019 (UTC)

As no.wikipedia is in fact in Bokmål Norwegian, please add parameter |no=Norsk Bokmål, thx. --Liuxinyu970226 (talk) 22:59, 16 June 2019 (UTC)

 On hold since I have no idea if nowiki accepts both Norwegian_language#Bokmål_and_Nynorsk and if this differentiation should be manually configured here on the English Wikipedia. Perhaps nowiki user: @Haros: could provide some more information? — xaosflux ^Talk 13:34, 19 June 2019 (UTC)

Do the change. I could indeed provide more information, but just not to invite the civil war here as well, I will just say that norsk bokmål was the description on the interwiki links for many years. Haros (talk) 14:46, 19 June 2019 (UTC)

@Haros: thank you for the feedback. (Haros is a nowiki 'crat that I invited here to verfiy). — xaosflux ^Talk 23:19, 20 June 2019 (UTC)

Side note, the last thing we want is a "foreign" language war here - if this is challenged it should be reverted and likely referred to metawiki. — xaosflux ^Talk 23:19, 20 June 2019 (UTC)

Edit request relating to listing of the Simple English Wikipedia

It currently says "English (simple form)" on this template. I think it should be changed to Simple English as that what it is referred to as everywhere else. Thanks, Vermont (talk) 03:29, 12 August 2019 (UTC)

@Vermont: I'm generally supportive of changing that label, but think there may be a reason not to here: This line is alphabetical and that would put that english dialect down in to the "S"'s. Would English (Simple English) work for you? — xaosflux ^Talk 13:32, 12 August 2019 (UTC)

@Xaosflux: Yep, that seems good too. I just found "simple form" to be odd. Thanks, Vermont (talk) 13:33, 12 August 2019 (UTC)

 Doing... - reviewing and validating. — xaosflux ^Talk 13:34, 12 August 2019 (UTC)

 Done OK @Vermont: this is live now for logged in editors, main page cache may take a little bit to update that template for logged out users. — xaosflux ^Talk 13:37, 12 August 2019 (UTC)

Thanks Xaosflux! Best regards, Vermont (talk) 13:46, 12 August 2019 (UTC)

Hebrew Wikipedia

The 250,000 article in the Hebrew Wikipedia was written a day or two ago; can someone move עברית to the list of Wikipedias with more than 250,000 articles? SHAMZAN talk 06:00, 13 September 2019 (UTC)

Done. PrimeHunter (talk) 16:14, 13 September 2019 (UTC)

Thx. SHAMZAN talk 17:17, 15 September 2019 (UTC)

Danish Wikpedia

Danish (dansk) Wikpedia has now more than 250.000 articles. The template shows less. Could any of the admins move the danish Wikipedia? --Glenn (talk) 20:45, 3 June 2019 (UTC)

Would like to remind everyone of this again, as it seems no one has answered. —Biscuit-in-Chief :-) ^{(Talk – Contribs)} 22:11, 9 June 2019 (UTC)

Yes I also just noticed this and it seems like its still not been updated.Johanhilge (talk) 12:48, 18 July 2019 (UTC)

Edit request 18 July 2019

Please move * {{Wikipedia languages/core|da}} to the second category (250,000+). This was first requested on June 3, above, and since no one's opposed it since then I assume it can safely be requested. —Nizolan ^{(talk · c.)} 22:34, 18 July 2019 (UTC)

 Done --kingboyk (talk) 00:31, 19 July 2019 (UTC)

on my phone on the Wikipedia.org homepage, it is still in 100,000 DemonDays64 (talk) 18:27, 28 September 2019 (UTC)

oh I’m dumb XD there isn’t a category between 100K and 1M on mobile :P DemonDays64 (talk) 18:28, 28 September 2019 (UTC)

Edit request: Asturian Wikipedia

The Wikipedia version in Asturian just reached 100,000 articles, could somebody with permissions add it to the list? Thanks. YoaR (talk) 10:25, 28 January 2020 (UTC)

{{edit protected}} added. * Pppery * _{it has begun...} 01:53, 5 April 2020 (UTC)

Done we don't have a category for >100,000 but I've added to >50,000 — Martin (MSGJ · talk) 19:53, 12 April 2020 (UTC)

Macedonian Wikipedia (mk)

Hello admins, hello @MSGJ:. I am a bureaucrat from mk.wiki and I would like to ask our Wikipedia to be added to this list of Wikipedias. As of my writing, we are currently at 106.817 articles, and, on average, our edition is of sufficient ratio of article size (i.e. depth) per number to be added to the list of other languages. All of the admins and our community in genral are in anticipation of this addition and will be very happy to see it here.

If possible, we would like to be added both here, and to the languages sidebar on the left-hand side. Much appreciated and many thanks! B. Jankuloski (talk) 07:11, 24 July 2020 (UTC)

 Done @Bjankuloski06: this has been added. — xaosflux ^Talk 13:54, 5 August 2020 (UTC)

"Wikipedias determined to consist primarily of stubs and placeholders are omitted."

This appears to be the statement that the exclusion of Cebuano (and a few others) from the list transcluded to the main page is based on. Is this determined subjectively or objectively and is there a particular column in the tables used at meta:List_of_Wikipedias which is key? (I'm guessing depth, but I don't know what the break point is)Naraht (talk) 01:33, 8 April 2020 (UTC)

Some editors check 50 random articles. See Template talk:Wikipedia languages/Archive 7#Update number of Wikipedias above 1 million articles. PrimeHunter (talk) 01:39, 13 April 2020 (UTC)

Thanx.Naraht (talk) 02:36, 13 April 2020 (UTC)

@PrimeHunter: if you are more familiar with the criteria than I am, would you consider writing an FAQ for this page to describe them? It seems to be some combination of number of articles, page views, quality and depth. It would be good to get some kind of guideline written up. — Martin (MSGJ · talk) 14:15, 18 November 2020 (UTC)

@MSGJ: I'm not doing it. I'm inactive as editor (don't know future), and inexperienced here. PrimeHunter (talk) 10:12, 19 November 2020 (UTC)

Absolutely fine. It was just a suggestion and we are just volunteers! — Martin (MSGJ · talk) 11:27, 19 November 2020 (UTC)

Uzbek Wikipedia

Hi, English Wikipedia administrators!

I am administrator & bureaucrat from Uzbek Wikipedia and I would like to ask our Wikipedia to be added to this list of Wikipedias. We are currently at 137,000 articles, to be added to the list of other languages. All of the administrators and our community in genral are in anticipation of this addition and will be very happy to see it here.

If possible, we would like to be added both here, and to the languages sidebar on the left-hand side. Cheers! Malikxan ^talk 16:59, 7 November 2020 (UTC)

I have added {{editprotected}} to these two requests so that they may receive attention. Double sharp (talk) 12:13, 18 November 2020 (UTC)

 Administrator note According to meta:List of Wikipedias, Uzbek Wikipedia has a depth of 43, which is higher than e.g. Asturian (depth 9) which is included in the template. Awaiting comments from others who understand how this depth value should be interpreted. — Martin (MSGJ · talk) 14:27, 18 November 2020 (UTC)

The meaning of depth is The "Depth" column (defined as [Edits/Articles] × [Non-Articles/Articles] × [1 − Stub-ratio] ) is a rough indicator of a Wikipedia’s quality, showing how frequently its articles are updated. It does not refer to academic quality. A higher number is generally better I think, based on that. (And indeed, I'd say so per just the 1m+ table, which lists bot-favorite war, ceb, vi hanging out toward the bottom when sorted this way.) --Izno (talk) 15:59, 19 November 2020 (UTC)

@Izno: help me understand this. The 1 - Articles/Total looks like it should be the same as NonArticles/Total. How does stub ratio get calculated? — Martin (MSGJ · talk) 12:16, 23 November 2020 (UTC)

@MSGJ: I presume that NonArticles is all non-article space pages, and the ratio rather describes the difference between stubs and articles. Whoever added the below formula was incorrect about this term, as I assume stub ratio is Stubs/Articles, though that is not particularly described on that page. --Izno (talk) 18:41, 24 November 2020 (UTC)

${\text{Depth}}={\frac {\text{Edits}}{\text{Articles}}}\cdot {\frac {\text{NonArticles}}{\text{Articles}}}\cdot \left(1-{\frac {\text{Articles}}{\text{Total}}}\right)$

@MSGJ: we probably should pick a standard depth cut-off, as sites that have gone heavy with bot articles like cebwiki stick out, however we do have svwiki with depth <10, so its a bit tricky. — xaosflux ^Talk 16:29, 19 November 2020 (UTC)

If we accept this formula for "depth" as being a good metric for the quality of a Wikipedia, then setting a cut-off value which preserves roughly the same number of links that we currently have would be a transparent and fairer method to use. — Martin (MSGJ · talk) 19:27, 21 November 2020 (UTC)

Svwiki is full of fairly unhelpful stubs, compared to dewiki for instance. I would set the cutoff higher than 10. (t · c) buidhe 08:34, 22 November 2020 (UTC)

I am ever-amused that we do treat the bot-created wikis so poorly, considering Rambot's effect on ours, which is most-certainly in the long tail of articles we do not get around to maintaining now.

That aside, even those wikis with many bot articles sometimes (such as on sv) do have some significant work behind them elsewise. I think it would be good to consider how the current list got the way it is so we can see if we do indeed need a systematic approach to this. --Izno (talk) 18:41, 24 November 2020 (UTC)

Redesign

There are a number of things that could be improved about the design of this template for the main page:

• By listing the number of articles on en-WP, it duplicates the top of the main page. Also, the "started in 2001" is not about languages, so is misplaced.
• There is poor visual hierarchy. It'd be better to make the individual language links a little smaller/less emphasized than the tier headings.
• "English (Simple English)" is awkward. It'd be better to have just "Simple English".

Could we do a bit of redesigning? {{u|Sdkb}} ^talk 01:22, 21 September 2020 (UTC)

There is a discussion about the wording of "English (Simple English)" here, just above on this talk page. Jmchutchinson (talk) 05:13, 21 September 2020 (UTC)

@Jmchutchinson: Thanks for pointing that out. My suggestion would be to keep the sortkey at "E" but change the label as above. {{u|Sdkb}} ^talk 05:25, 21 September 2020 (UTC)

support RIT RAJARSHI (talk) 22:36, 21 September 2020 (UTC)

Make the table colourful. I would personally prefer blackish green for languages, red for headings and yellowish white background colour. RIT RAJARSHI (talk) 22:49, 21 September 2020 (UTC)

Protected edit request on 23 November 2020

Following up (part 1) from the above, please change simple=English (Simple English) to simple=Simple English. The sorting shouldn't change the sorting but will make the wording less clunky. {{u|Sdkb}} ^talk 22:12, 23 November 2020 (UTC)

 Done — Martin (MSGJ · talk) 22:21, 23 November 2020 (UTC)

Sandbox

How does this sandbox look for the remaining edits?

This Wikipedia is written in English. Many other Wikipedias are available; some of the largest are listed below.

Complete list of Wikipedias

1,000,000+ articles

• العربية
• Deutsch
• Español
• Français
• Italiano
• Nederlands
• 日本語
• Polski
• Português
• Русский
• Svenska
• Українська
• Tiếng Việt
• 中文

250,000+ articles

• Bahasa Indonesia
• Bahasa Melayu
• 閩南語
• Български
• Català
• Čeština
• Dansk
• Esperanto
• Euskara
• فارسی‎
• עברית
• 한국어
• Magyar
• Norsk bokmål
• Română
• Srpski
• Srpskohrvatski
• Suomi
• Türkçe

50,000+ articles

• Asturianu
• Bosanski
• Eesti
• Ελληνικά
• Simple English
• Galego
• Hrvatski
• Latviešu
• Lietuvių
• മലയാളം
• Македонски
• Norsk nynorsk
• Slovenčina
• Slovenščina
• ไทย

Here is what it looks like in context. I'm sure my coding is messier than it should be for something going on the main page, so any CSS whizzes feel free to fix it up. Kudos to @Jonesey95, Nihiltres, and Xaosflux: for help at VPT. {{u|Sdkb}} ^talk 21:02, 24 November 2020 (UTC)

It looks like an improvement to me. I swapped it in Preview in a couple of places where it is transcluded, and I checked it in Mobile view, and it looked fine. – Jonesey95 (talk) 21:10, 24 November 2020 (UTC)

@Sdkb: not good, same issue I brought up at VPT, for example look at it in monobook, and for an extreme example look at it in cologneblue. — xaosflux ^Talk 21:33, 24 November 2020 (UTC)

Xaosflux, yeah, that's def an issue. Any idea how to resolve it? {{u|Sdkb}} ^talk 21:35, 24 November 2020 (UTC)

Why are you hard-coding a background color at all? — xaosflux ^Talk 21:37, 24 November 2020 (UTC)

I think it's so that the horizontal rule doesn't go through the "More than X" text. It'd be better to find some way to get the actual line to stop behind the text rather than just covering it up. {{u|Sdkb}} ^talk 21:49, 24 November 2020 (UTC)

Also, in timeless and minerva the text is not vertically centered with that line, it is using the line as a floor. — xaosflux ^Talk 21:36, 24 November 2020 (UTC)

Yeah, I'm not happy with the prototypical model at all on reflection; it makes way too many assumptions about the environment in which it's rendered. I'd rather an old-school presentational table for backwards compatibility, with ARIA details added (in particular, note the role="heading" aria-level="3" on the text):

1,000,000+ articles

I'm cheating a bit on the handling of widths (the "line" cells are each set to have 50% width, which is obviously impossible) and that should get looked at by someone who remembers the details of table-cell sizing better than I do. At worst, we could hard-code a few preset line widths in CSS according to the size of the viewport; it wouldn't make much of an aesthetic difference. {{Nihiltres |talk |edits}} 23:01, 24 November 2020 (UTC)

Nihiltres I interrupt this conversation to inform you that your self-closed div tags were causing Self-closed tags lint errors, and I fixed them by changing them to ordinary div tags followed by </div>. This has no effect on the display. —Anomalocaris (talk) 20:42, 25 November 2020 (UTC)

Nihiltres, looks fantastic! I think I've implemented it right—the only remaining issue is that, in mobile view on large screens, the whole horizontal rule is shifted to the left rather than having an even amount of white space to the left and right of the rule as with desktop. Any idea how to fix that? {{u|Sdkb}} ^talk 02:12, 25 November 2020 (UTC)

Right, I rarely use that mobile view so I forgot that it applies margin-left: 0 !important; margin-right: 0 !important; to all tables (among other things), making styles that centre blocks generally not work (because the usual way to centre things involves setting both horizontal margins to auto). Yuck. I'll outline a few options to solve the problem:

1. Add !important to the end of the inline CSS margin rule on the main table to tell the default rule to fuck off, or similarly add TemplateStyles CSS to override the default. The catch is, it makes the styles hard to override for those who like to personally restyle the Main Page.
2. Set the width to 100% instead of 90%; it's not a big deal aesthetically and sidesteps the problem. Optionally, set the width to 100% only on Minerva using TemplateStyles (which is just a touch hacky, but should be fine).
3. Set the width to 100%, but add (empty) table cells with 5% width each at start and end. This is very stable, but using structure to impose style is gross. On the other hand, this style is already using a presentational table.

I don't have too much preference between these options, but I almost want (1) out of spite for that annoying feature of mobile. {{Nihiltres |talk |edits}} 05:26, 25 November 2020 (UTC)

I think the 90% looks better than 100%, so I'm not a fan of option 2. I went with option 1 for the sake of clean code, but if someone else prefers option 3, feel free to change the sandbox to that.

Some other small changes to note are that I moved "Complete list of Wikipedias" from the bottom to the upper right to save a line, since otherwise this redesign would cause the section to take up a fair bit more vertical space than the present version; and I changed "More than X articles:" to "X+ articles", which I think is cleaner.

The result looks ready to implement to me. I'll make an edit request soon if there are no further concerns. Thanks everyone for the CSS help! {{u|Sdkb}} ^talk 06:19, 25 November 2020 (UTC)

Protected edit request on 27 November 2020

Per the above, please implement the sandbox version. {{u|Sdkb}} ^talk 22:34, 27 November 2020 (UTC)

 Not done While I agree that the proposed version of this template looks nicer than the current version, the current template includes a few things that the proposed replacement does not include, and there doesn't seem to be any discussion above about these details. Specifically:

• The current template has some code that prevents the first few sentences from being included on Special:Statistics. The proposed replacement has removed this code. Was this intentional? Is there consensus for this omission?
• The current template includes language about the year that Wikipedia was started, and the current article count. The proposed replacement has removed this language. Is there consensus for this change?

‑Scottywong| [chat] || 09:20, 29 November 2020 (UTC)

Scottywong, thanks for taking a look. Both of those things follow from my first bullet point at the very top: By listing the number of articles on en-WP, it duplicates the top of the main page. Also, the "started in 2001" is not about languages, so is misplaced. Since I removed these bits from the template overall, there's no longer need to suppress them at the statistics page. (I wouldn't be opposed to putting the founding date somewhere else on the main page if we could find a spot where it fits, but it's not really a loss – it can be easily found by anyone who wants to read about WP's history, and there's no real benefit to promoting it beyond that.) {{u|Sdkb}} ^talk 18:58, 29 November 2020 (UTC)

I understand that reasoning, and I don't necessarily disagree with it, but the founding date and article count appears to be the status quo on the main page since as far back as March 2006. Given that this is a relatively large change to the main page, I wonder if it deserves a bit more discussion before implementing it. Does anyone else agree, or am I overreacting? ‑Scottywong| [prattle] || 22:09, 29 November 2020 (UTC)

Scottywong, that's perfectly fair. I'll drop a {{Please see}} at Talk:Main page to give people a chance to comment, and then reactivate the request after a few days if there appears to be consensus. {{u|Sdkb}} ^talk 22:27, 29 November 2020 (UTC)

Perfect, thanks. I'm happy to implement the new design if there are no significant objections after a few days, or any other interested admin may implement it as well. Feel free to either ping me and/or reactivate the request. ‑Scottywong| [talk] || 01:19, 30 November 2020 (UTC)

I have no issue with these removals for those reasons. --Izno (talk) 15:56, 1 December 2020 (UTC)

Good stuffand glad to see the link to the complete list has been placed at the bottom. My only gripe, which needs to be addressed for WP:ACCESSIBILITY is that the structure of the information is not reflected in the page's code. The information is a single list, with three items each containing a count of articles and a list of language items; but the implementation is three pairs of unconnected divs, the first of each pair being the count, the second separated tables, each followed by a div containing a list of languages. It would be better if list structures were used throughout; so:

<ul>
 <li>More than 1,000,000 articles:
  <ul>
   <li>العربية</li>
   <li>Deutsch</li>
   <li>Español</li>
   etc
  </ul>
 </li>
 <li>More than 250,000 articles:
  <ul>
   <li>Bahasa Indonesia</li>
   <li>Bahasa Melayu</li>
   <li>Bân-lâm-gú</li>
   etc
  </ul>
 </li>
 etc
</ul>
Full list link

Use CSS to ensure the top-level lis behave as divs, complete with your current styling. This would address WP:ACCESSIBILITY requirements of presenting lists of information as real lists for those who use assistive technology. Good job otherwise. Apologies for my quick-and-dirty wiki mark-up. Bazza (talk) 11:22, 30 November 2020 (UTC)

Bazza 7, are you looking at the sandbox or at the current version? {{u|Sdkb}} ^talk 16:36, 30 November 2020 (UTC)

@Sdkb: I was looking at [16] as linked from Main Page Talk, but missed the used of three tables for markup which is also less than ideal (tables are for tabulated data, not layout). I've amended my comments. The link to the full list seems more inituitive after the list of languages, as it is at the present. My comment and suggestion above about lists within a list being a preferable option still stands. Bazza (talk) 17:29, 30 November 2020 (UTC)

Bazza_7, moving the complete list to the upper right was so that the new design would not take up significantly more vertical space than the old design. Without knowing how many people actually use these links, I'm hesitant to allocate that space, but if there's strong desire for it to be at the bottom we can move it back there.

Regarding the tables, there's some discussion above in the #Sandbox section; it appears to be the best method available.

Regarding lists within a list, would you be able to edit the sandbox to make it the way you want? Changes under the surface that help for accessibility but don't affect the normal appearance are certainly good. {{u|Sdkb}} ^talk 18:34, 30 November 2020 (UTC)

I have the same concerns as Bazza. I am going to take a shot at ameliorating them in Template:Wikipedia languages/sandbox2 while trying to retain the new styling. --Izno (talk) 21:46, 30 November 2020 (UTC)

Done. I question the validity and necessity of the HTML IDs, but other than that, please take a look. Flex-box should degrade gracefully for users who are using ancient browsers that we even support; most of the browsers that don't support flex-box aren't served any more because of the HTTPS restrictions (c.f. caniuse versus this list of browsers, which leaves only IE 9, 10, and Firefox 27 as not providing full support; even IE 11 rendering is respectable surprisingly). As for roles, role are only necessary where the role implies something different than the default (so role=presentation on the prettybars is unnecessary). --Izno (talk) 23:15, 30 November 2020 (UTC)

Izno, looks good; I checked on mobile and different skins and don't see any issues. Thanks for making the sandbox! {{u|Sdkb}} ^talk 01:06, 1 December 2020 (UTC)

Sdkb, as you've noticed and since responded, what I was saying is that I don't like how the complete list looks with the span floating after the rest of the paragraph. It doesn't look too great on some resolutions. --Izno (talk) 03:11, 1 December 2020 (UTC)

• 
  Look 1
• 
  Look 2

@Izno: I uploaded screenshots of what it could look like at narrow screen resolutions to make it easier to tell what we're talking about. If I'm understanding correctly, you prefer Look 1 and I prefer Look 2, although I agree neither are fully optimal. I think the best solution might be to take Look 2 but center-justify the complete list link at mobile resolutions. Alternatively, we could start with Look 1 but prevent text from appearing underneath the complete list link, i.e. make "some of the largest are listed" go onto a new line. I don't like quite as much as I would a modified Look 2, but it'd be better than an unmodified Look 1. {{u|Sdkb}} ^talk 03:29, 1 December 2020 (UTC)

I don't much like any of the options I've seen so far including the link as floating, to be honest. ;) --Izno (talk) 03:53, 1 December 2020 (UTC)

I'd vote for Look 2 if those are the only two choices available. Otherwise, another option to explore is to get rid of the floating "Complete list of Wikipedias" link altogether, and incorporate that link into the prose instead (for all resolutions, not just narrow ones). For instance, here are a few ideas:

• "This Wikipedia is written in English. Many other Wikipedias are available; some of the largest are listed below. A complete list of Wikipedias is also available."
• "This Wikipedia is written in English. Many other Wikipedias are available; some of the largest are listed below."

‑Scottywong| [confess] || 07:44, 1 December 2020 (UTC)

That second shorter version is my preference. Bazza (talk) 09:29, 1 December 2020 (UTC)

I would have looked at Sdkb's request today (UTC+0), but Izno has done the hard work already: thank you. Now the structure is in place, people can take as much time as they need to tweak the presentation. Bazza (talk) 09:36, 1 December 2020 (UTC)

I was tending this direction last night and now have adjusted the sandbox in what I'd like to see. I think it's important to make this link really obvious as being the list of everything, so I'm not such a fan of the 'short' version. I've also moved the nowraplinks and plainlinks to the list from the top class because it's pretty lengthy text to be wrapping in nowraplinks. --Izno (talk) 15:52, 1 December 2020 (UTC)

Complete list of Wikipedias

• I think Moving this link up in to the top line looks bad as-is, and looks worse in different ways at different resolutions. Main reason is that this differently-formatted text looks almost like the normal written text preceding it, but just different enough to be wrong - including how it aligns. I suggest that if we want this to be on the top, we just incorporate it in to the Many other Wikipedias are available; some of the largest are listed below. sentence. — xaosflux ^Talk 16:05, 1 December 2020 (UTC)

  Okay, so circling back on this, the "complete list" line is the only remaining thing to iron out. Xaosflux's suggestion works alright enough for me; I put that at Template:Wikipedia languages/sandbox2 and I'll add the edit request. {{u|Sdkb}} ^talk 03:14, 14 December 2020 (UTC)

 Done Izno (talk) 19:44, 30 December 2020 (UTC)

Analysis of using depth

A simple analysis follows of the changes entailed by setting a minimum depth of 10 — Martin (MSGJ · talk) 21:40, 23 November 2020 (UTC)

Articles	Number of links	To add	To remove
1,000,000+	14 (same)	Portugese	Swedish
250,000 - 999,999	19 (same)	Armenian	Min Nan
50,000 - 249,999	42 (27 more)	Bengali, Hindi, Urdu, Teugu, Marathi, Tagalog, Uzbek, Welsh, Afrikaans, Albanian and 18 other languages	Asturian

As the last category is now very much larger, I might recommend changing it to 100,000+, i.e.

Articles	Number of links	To add	To remove
100,000 - 249,999	25 (10 more)	Hindi, Urdu, Uzbek, Welsh, Georgian, Tamil, Azerbaijani, Belarusian, Volapuk, Latin, Kazakh, Cantonese, Tajik	Asturian, Malayalam, Bosnian

— Martin (MSGJ · talk) 21:52, 23 November 2020 (UTC)

That sounds good to me. I just don't see tons of super minor languages being linked from the main page of English Wikipedia as super important. Listing all the languages is what the main wikipedia.org portal is for, and for the tiny number of stray Malayalam speakers who end up here, there's a link to the full language list.

Regarding the broader question of determining languages to list, is there a reason we use article/edit count rather than total pageview count? We should be tailoring the page based on where readers want to go, and pageviews seems like a better indicator of that that'd allow us to sidestep the bot inflation issue. {{u|Sdkb}} ^talk 22:53, 23 November 2020 (UTC)

Sounds interesting. I'd certainly likr to see the stats on that and see how they compare with the other metrics. — Martin (MSGJ · talk) 23:12, 23 November 2020 (UTC)

Yes, we should always be serving our readers. Perhaps factoring in number of speakers of a language would tweak the focussed links? Stephen 23:57, 23 November 2020 (UTC)

I think there would be very little correlation between the number of speakers of a language and the quality of that language's Wikipedia, which I think is what we're trying to measure here? I liked the idea of the number of views because I think a high quality Wikipedia is likely to be viewed more. — Martin (MSGJ · talk) 17:02, 24 November 2020 (UTC)

That's not always true. There are high-quality Wikipedias that couldn't amass page views just because the number of speakers is low.--Kiril Simeonovski (talk) 15:04, 3 December 2020 (UTC)

If there's a Wikipedia that's high-quality but in a language essentially no one reads, I don't think we ought to link to it. {{u|Sdkb}} ^talk 03:21, 14 December 2020 (UTC)

Well, if a high-quality Wikipedia is not read enough for any reason, then we can help it receive more page views by linking to it. After all, the main page should link to high-quality content.--Kiril Simeonovski (talk) 14:52, 30 December 2020 (UTC)

@MSGJ: It's my honor to inform you that, Bengali Wikipedia (bnwiki) has also reached the milestone of 1,00,000 articles on 25 December. I think, we should emphasize on promoting good-quality Wikipedias from enwiki main page. Sorting by project view is not a good way to search out good-quality wikipedias with less speakers, and moreover, we also “should not neglect the major languages”. For instance, we have Hindi, Bengali and Urdu as the major languages in the Indian subcontinent but enwiki shows Malaylam. But these South Asian languages have more articles, more depth and more native speakers than Malaylam. They are also doing a great job. They also need to recognisions and promotions. I strongly expect enwiki main page to promote those “good-quality”, “deep enough” wikipedias from “each major part of the world” with “considerable number of native speakers”. The present criterion includes article number and depth (I am not sure whether there is anything else or not), I propose to include number of native speakers and articles/native speakers ratio, FA ratio, and GA ratio to the criterion. I think they would be explicit enough to determine which wikipedias are to be kept and which not. Thank you! :) Meghmollar2017 • Talk • 15:10, 9 February 2021 (UTC)

Georgian Wikipedia

Where is Georgian Wikipedia, there are more than 144,000 articles in kawiki--ჯეო4WIKI (talk) 20:52, 15 November 2020 (UTC)

I have added {{editprotected}} to these two requests so that they may receive attention. Double sharp (talk) 12:13, 18 November 2020 (UTC)

Comment: depth=38 — Martin (MSGJ · talk) 14:28, 18 November 2020 (UTC)

@MSGJ: But then, Dutch Wikipedia has a lower depth of 17 and then we also included it? --Liuxinyu970226 (talk) 00:09, 12 February 2021 (UTC)

Srpski => Српски

'Srpski' should be 'Српски' or 'Српски / Srpski' --Milan.j (talk) 11:39, 3 March 2021 (UTC)

Cebuano should be added to the main page

At the bottom of the homepage, where all the languages are listed, Cebuano is not listed in the category: 1,000,000+ articles, despite it having 5,000,000+ articles. It should be added on that list. — Preceding unsigned comment added by Jizzygizzyfoshizzyyy (talk • contribs)

Wikipedias infested with bot-created articles do not count. Aside from size, depth is the more important criterion. Howard the Duck (talk) 21:17, 28 August 2021 (UTC)