1. As I indicated last week, I am rejecting the request for Wikipedia Simple German, encouraging anyone interested to consider trying to add simple content to dewiki.
2. I am going to change the request for Wikipedia Teochow from "on hold" to "reject" (per this email<https://lists.wikimedia.org/pipermail/langcom/2018-March/001999.html> to LangCom).
3. I am going to change the request for Wikipedia Kumaoni from "on hold" to "reject-as-stale". In January, I had marked this request (dating back to 2010) as "on hold" because no one had ever showed up to create a test project. (See this message<https://lists.wikimedia.org/pipermail/langcom/2018-January/001857.html>.) But since then we have started marking these as "reject-as-stale", and that is appropriate here.
Sent from Outlook<http://aka.ms/weboutlook>
In the meantime, I intend to continue to move other old requests through. Here are the first set of Wikipedia requests that have been open since 2011:
* Wikipedia Jawi (Djawi)<https://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_Jawi> (djw): Extinct Australian language. No test project was ever created. Rejecting.
* Wikipedia Bafia<https://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_Bafia> (ksf): Spoken by 60,- 120,000 Cameroonians. Test has about 20 pages and is moribund. But there's no reason this should not be eligible.
* Wikipedia Mundang<https://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_Mundang> (mua): Language of about 240,000 in Chad and Cameroon. One page in test dating to the time of the request, then nothing. Requester no longer active at WMF. Rejecting as stale.
* Wikipedia Simple German 4<https://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_Simple…>: The only thing that makes this request different is that it refers to a standard for simplified German being developed at the University of Hildesheim; see https://www.uni-hildesheim.de/leichtesprache/leichte-sprache-regeln/#c36425…. I'm going to wait a week for comments from the Committee, but I intend to reject this, with the following additional notations: (1) It is better to have such content live in German Wikipedia. (2) In any event, such content would have to incubate in German Wikipedia first; proposed "simple" projects are not permitted in Incubator. (3) If and only if this becomes a well-accepted standard, there is substantial content residing in German Wikipedia, and then German Wikipedia petitions the committee to spin the simple project off, we would consider it in the future.
* Wikipedia Kwasio<https://meta.wikimedia.org/wiki/Requests_for_new_languages/Wikipedia_Kwasio> (nmg): Another Cameroonian language, now with 25,-75,000 speakers. Two pages only, from time of request, then moribund. Requester no longer active at WMF. Rejecting as stale.
Steven
Sent from Outlook<http://aka.ms/weboutlook>
In response to MichaeL:
>> I. The language itself
>> The proponents of the project have convinced me that Montenegrin is comparable as a language standard to Serbian, Croatian or Bosnian.
>How have they convinced you? Where are the pages of paradigms?
Understand what they have and have not convinced me. They have convinced me that Montenegrin is as entitled to be called a language as Serbian or Croatian—no more, and no less. You have frequently gone back and compared this situation to the very fine work you did on Western Armenian, but I don't think it's at all comparable. In this case, I am merely saying that Montenegrin is one of the four language standards within the macrolanguage Serbo-Croatian. So: If you were asked, today—leaving aside history—whether an independent Serbian Wikipedia would be eligible under current rules, what would you say? If you would say no, then we're pretty much in the same position. If you would say yes, then I'd like to know why Serbian qualifies, but Montenegrin doesn't.
>> Conclusion
>> Several people have said to me that Montenegrin is more similar to other Serbo-Croatian varieties than US and UK English are to each other; would I insist on separate projects if they happened to have separate language codes? No, I wouldn't. But that's because on the whole, the various English-speaking communities around the world do manage to co-exist with each other quite well—and tend to blunt each other's excesses a bit, too. Sadly, that's not the case here.
>So you want us to enable their divisions?
In an ideal world, I would prefer not to. But the divisions already exist and the history already exists, both in the world at large and within our WMF microcosm. If you have a way to wipe out these divisions and have everyone work together in harmony on a single Serbo-Croatian project, then please tell me how to do that. I'm saying very clearly: I do not think it is possible. (Do you think the Montenegrins are upset now? Try to close the other three projects and merge them into Serbo-Croatian. Then you'll really hear screaming.) Maybe none of the four projects (Serbian, Croatian, Bosnian, Montenegrin) should really exist as independent projects. But three do, and we are not going to delete them. And if those three do, the fourth must also; it's no less deserving than the others.
Steven
Michael
------------------------------
Subject: Digest Footer
_______________________________________________
Langcom mailing list
Langcom(a)lists.wikimedia.org
https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.wik…
------------------------------
End of Langcom Digest, Vol 54, Issue 5
**************************************
Sent from Outlook<http://aka.ms/weboutlook>
I would like to retract the attached email as it seems like there are
actually no current request for a code for Hainanese that I was mixing
something else up with it
2018-01-30 TUE Phake Nick <c933103(a)gmail.com> wrote:
> One thing to consider:
> - In the previous code application, Hainanese was mentioned as a thing to
> consider about before splitting out Teochew. There are currently request
> for a code for Hainanese which will probably take some times to handle, and
> I would not expect request for Teochew to surface before that one get
> created
>
> 2018年1月29日 18:26 於 "Steven White" <Koala19890(a)hotmail.com> 寫道：
>
>> OK. I'm assuming that (a) the concept of closing stale requests as I've
>> proposed is generally acceptable, and (b) that at least in the cases other
>> than Teochew I can proceed.
>>
>>
>> With respect to Teochew, I'm going to mark it as "on hold/waiting",
>> pending a language code. But if we don't see a new request at SIL in a
>> year, then I'm going to close. Please let me know if that is acceptable.
>>
>>
>> There are, in fact, a couple of other requests from 2010 still open.
>> There are two requests on different Balochi projects, which I thought
>> should wait until Satdeep finished his investigations into that. There is a
>> request for "Southern Min in Hanji," which I intended to leave sitting
>> until we had a discussion of when different scripts need different projects
>> and when not. But apparently phabricator T165882
>> <https://phabricator.wikimedia.org/T165882> says that the community has
>> agreed to a namespace for Hanji, so this can be closed as resolved. There
>> is a request for Wiktionary Pitcairnese
>> <https://meta.wikimedia.org/wiki/Requests_for_new_languages/Wiktionary_Pitca…>
>> that can be closed as stale along the same lines as the others here. And
>> there is a request for Wikipedia Chinuk wawa that is supported by a few
>> pages in the Incubator, so I'm going to mark it eligible.
>>
>>
>> Steven
>>
>>
>> Sent from Outlook <http://aka.ms/weboutlook>
>>
>>
>> ------------------------------
>> *From:* Langcom <langcom-bounces(a)lists.wikimedia.org> on behalf of
>> langcom-request(a)lists.wikimedia.org <langcom-request(a)lists.wikimedia.org>
>> *Sent:* Friday, January 26, 2018 7:00 AM
>> *To:* langcom(a)lists.wikimedia.org
>> *Subject:* Langcom Digest, Vol 52, Issue 24
>>
>> Send Langcom mailing list submissions to
>> langcom(a)lists.wikimedia.org
>>
>> To subscribe or unsubscribe via the World Wide Web, visit
>>
>> https://nam01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.wik…
>> or, via email, send a message with subject or body 'help' to
>> langcom-request(a)lists.wikimedia.org
>>
>> You can reach the person managing the list at
>> langcom-owner(a)lists.wikimedia.org
>>
>> When replying, please edit your Subject line so it is more specific
>> than "Re: Contents of Langcom digest..."
>>
>>
>> Today's Topics:
>>
>> 1. Re: Final group of projects with requests lingering since
>> 2010 (Phake Nick)
>> 2. Re: Final group of projects with requests lingering since
>> 2010 (MF-Warburg)
>>
>>
>> ----------------------------------------------------------------------
>>
>> Message: 1
>> Date: Thu, 25 Jan 2018 09:53:21 +0800
>> From: Phake Nick <c933103(a)gmail.com>
>> To: Wikimedia Foundation Language Committee
>> <langcom(a)lists.wikimedia.org>
>> Subject: Re: [Langcom] Final group of projects with requests lingering
>> since 2010
>> Message-ID:
>> <
>> CAGHjPP+tUooqAWcJdrKA+nYNZY3Qi+MZnrJquY0ywOVYamSKfA(a)mail.gmail.com>
>> Content-Type: text/plain; charset="utf-8"
>>
>> 2018年1月25日 03:49 於 "MF-Warburg" <mfwarburg(a)googlemail.com> 寫道：
>> >
>> > Well, but it's equally true (and written) that "If there is no valid ISO
>> 639 code, you must obtain one. The Wikimedia Foundation does not seek to
>> develop new linguistic entities".
>>
>> My understanding on the description of "does not seek to develop new
>> linguistic entities" is that WMF does not seek to develop new language and
>> thus it would like confirmation from ISO standard regulation body, instead
>> of the code itself.
>>
>> > We do absolutely not want to invent our own codes, because that gets
>> really messy, especially when at some point a language does get a real
>> code.
>>
>> Why not tentatively use e.g. ISO639-6 code as a working code in incubator
>> or for the project before it could get a 639-1/2/3 code? after it get a
>> code in ISO 639-1/2/3 then it should be possible to move things over.
>> Although all the code change requests have been piled up for years in
>> phabricator but that should hopefully be sorted out one day.
>>
I. The language itself
The proponents of the project have convinced me that Montenegrin is comparable as a language standard to Serbian, Croatian or Bosnian. That, by itself, does not justify the creation of a Montenegrin Wikipedia. Surely it's mutually intelligible with other varieties, so to that extent you could argue that Montenegrin speakers could contribute elsewhere (at least if forced, but see points below). That having been said, if we were starting over now—if we had no projects in Serbo-Croatian at all, or if only the macrolanguage project currently existed—it would be very hard to justify treating any of the four differently from each other.
If that were the current situation, I'd probably agree with you not to create Montenegrin Wikipedia ... or Serbian, Croatian or Bosnian. But if you insisted on creating the other three, I would require you to create Montenegrin, too.
II. Current facts on the ground
The proponents of the project have convinced me that, at best, it is difficult for Montenegrins to contribute constructively to the other projects. This is true from the point of view of both language standards and content. There are many examples both of NPOV violations on subjects related to the politics of the region and on the use of Montenegrin linguistic varieties being rejected on the other projects. Based on the usual standards of project autonomy, it is very difficult for us to force these other communities to give equal access to the Montenegrin community. (And to some extent, it's probably reasonable for the Serbian, Croatian and Bosnian projects to prefer their own linguistic standards, even if the NPOV issue itself is still a problem on those projects.)
As far as the macrolanguage project itself, I suppose we could hope to reserve that for the use of Montenegrin. But we can't really enforce that position on that community, either. And shwiki is such a mess now that the Montenegrin community would have an easier time starting over than in fixing it.
The other result of all this is that a lot of Montenegrins simply don't care to participate; they simply don't want to bother fighting. And that goes toward violating WMF's goal to give everyone access, as per the next point.
III. Rule 3: "Sufficiently unique" vs. "free and unbiased access"
The long-time position being articulated by members of the committee relies on Point 3 of the "Requisites for eligibility": "The language must be sufficiently unique that it could not coexist on a more general wiki." It seems to me, though that the rest of the point is being ignored: "The committee does not consider political differences, since the Wikimedia Foundation's goal is to give every single person free, unbiased access to the sum of all human knowledge, rather than information from the viewpoint of individual political communities."
The position that "the committee does not consider political differences" is a fine one when we are starting off on a level playing field. But under the circumstances, it is my view that it is not viable to ignore political differences in this case. After all, the current situation is not one where "the viewpoint of individual political communities" is fully equal. Where we are now, in fact, is that every single "individual political community"—except the Montenegrin community—has its viewpoint already entrenched in the system. If we do not consider political differences in this case, we are, in fact, entrenching the viewpoint of some individual political communities at the expense of others. And that expressly violates the remit of the Language Committee.
I suppose that instead of creating Montenegrin Wikipedia, we could try to get the other projects to give equal access to the Montenegrin community. Good luck enforcing that, though.
I will argue in point V below that it is more politically neutral to allow Montenegrin than to reject it.
IV. Committee position on macrolanguages
The committee's current position allows projects in macrolanguages sometimes, but expresses a clear bias in favor of having projects in individual component languages rather than in macrolanguages. It is clear that this position is not absolute. Still, ruling against Montenegrin goes against that trend, rather than for the trend.
V. Language codes, LoC/SIL and LangCom's neutrality
Surely, the main reason we rely on SIL's decisions around language codes is that they are the official standard-keeper, and we are not. But as part and parcel of that, by relying on SIL's decisions, we are putting the burden of sorting out linguistic considerations from political ones on SIL, not on ourselves. Now, we are all aware that sometimes, at the borders, we might prefer to see things differently from the way SIL does. That is why there is now a procedure in place for situations where language codes don't exist. But unquestionably the existence or non-existence of a language code represents a strong default position on how LangCom should act. Indeed, we normally require a supermajority to allow projects that don't have ISO 639–3 codes.
In the past, part of the argument against Montenegrin has been "SIL [Ethnologue] describes it as just another name for Serbo-Croatian". Fine. Then, it was a politically neutral decision to reject Montenegrin, and would have been a politically "motivated" position to accept it. Now, the situation is reversed. Now, it is a politically neutral decision to accept Montenegrin, and a politically "motivated" position to reject it.
I am fully aware that many of you believe that Montenegrin's winning of a code was a political, rather than a purely linguistic, victory. There are academics who don't agree with that, but suppose that it is true. Let that be SIL's problem (or the Library of Congress's), not ours. When we choose to disagree with SIL, I think we have to justify that.
Finally, let me add that the Montenegrin community managed to get action not only at SIL, but actually at LoC first, getting the first change to ISO 639–2 in about five years. Again, maybe that was a political victory. But personally I don't think we ought to putting ourselves in a position where we are second-guessing all these experts.
VI. The Incubator test
The rules for allowing a test on Incubator are less stringent than the rules for approving a project. Accordingly, there has been a test project on Incubator since December. At this point, it is probably the highest-quality project we have in Incubator now, including the ones just being approved. There are about 65 editors (33 with over ten edits each) and 1,200 main space pages in the project. Pretty much none of them are the 1–2 sentence pages we often see on Incubator projects. Of the ten pages I just checked, nine had references, and the other was a list page. Solely on the basis of whether the community is working to create a serious encyclopedia project consistent with WMF's goals, I'd say that this community is very deserving of recognition.
VII. Appearance of neutrality and fairness
Say what you will about the rules, a situation where Montenegrin doesn't get a code appears profoundly unfair. Superficially, this situation is not much different from supporters of Ancient Greek complaining that Latin has a project, but they don't, because the rules changed at a certain point. But the intense political rivalry in the Balkans makes this a much less trivial case; after all, supporters of Ancient Greek don't try to interfere with the use of Latin on Latin Wikipedia. This case is simply one that I don't think we can justify by falling back on the rules. I'd far rather "bend the rules" in the direction of fairness—particularly because I don't even think this would be bending the rules. I think the rules can easily be interpreted to allow Montenegrin, rather than to reject it.
Conclusion
Several people have said to me that Montenegrin is more similar to other Serbo-Croatian varieties than US and UK English are to each other; would I insist on separate projects if they happened to have separate language codes? No, I wouldn't. But that's because on the whole, the various English-speaking communities around the world do manage to co-exist with each other quite well—and tend to blunt each other's excesses a bit, too. Sadly, that's not the case here. It is difficult, though not quite impossible, to justify Montenegrin Wikipedia solely on the grounds of linguistic uniqueness. But based on every other criterion we are supposed to evaluate, if we were starting over now, we'd either have only one Serbo-Croatian Wikipedia, or we would have separate projects for Serbian, Croatian, Bosnian and Montenegrin. Since we can't stuff the other three back in the bottle, the right thing to do now is to accept Montenegrin Wikipedia.
Respectfully,
Steven
Sent from Outlook<http://aka.ms/weboutlook>
Forwarded from on-wiki. It was posted there on Friday 2 March at about 18:30 UTC.
“Dear LangCom members,
I am writing to you with intention to clarify some things regarding the Montenegrin Wikipedia project. I am aware that there is a lack of support for this project to be allowed, and the main argument is that it is a “variety” of Serbian language, claimed even by some members here. I’d like to remind you that not a single serious institution claims this to be true, they all list it as a member of Serbo-Croatian macro language, not a part of Serbian (an important distinction).
Next, some question the use of two added letters all while claiming that they can simply be replaced with sj, zj. As per Montenegrin orthography, this is not true, I can explain everything in detail on the discussion page for anyone interested in this, please, be free to ping me there. And on the same page plenty of examples have been given to show how these letters are in fact used. Also, alongside a few other users, I have shown some clear differences between Ijekavian Serbian and Montenegrin (let me remind you that ~90% of articles in Serbian Wikipedia are written in Ekavian standard, not Ijekavian). The most important feature of Montenegrin is jekavian iotation, not present in any other SC language, and this means that it is impossible to follow Montenegrin standard if we were to write Serbian (or Bosnian, Croatian) Wikipedia.
And let’s not forget the reason that most members here wish to ignore, but sadly, due to historical and political reasons, cannot. Serious NPOV violations are present on Serbian Wikipedia on literally every article regarding Montenegro (and lots of other political, historical topics related to other countries and people but it is not relevant to Montenegrin Wikipedia right now). I have listed quite a few examples of that in discussion page and even some of reverting changes that were made to regain NPOV, ping me for all of that as well if you cannot find it on that cluttered page. Politics do in fact play a big part in all of this, and explain why Montenegrin community is, besides all the reasons listed above, also uninterested to contribute to such projects, and there is a clear intention to write neutral articles retaining NPOV among our community, not to create propaganda-style articles that can sadly be found on all of these other projects.
Some propose using Serbo-Croatian Wikipedia. The very existence of it is harshly criticized amongst most of Serbian, Croatian community, as they find it unnecessary and are often feeling angry at SC community merely copying their articles, as claimed by some users. I won’t argue about necessity of SC Wikipedia, but I felt that this is a good introduction for my main reason why Serbo-Croatian Wikipedia is not a good solution for Montenegrin either.
Very large number of articles have parts directly copied from Serbian, Croatian, Bosnian Wikipedias, resulting in a huge mess on articles, with a single sentence being written half in Croatian, half in Serbian, which is utterly confusing for the reader, switching between dialects, word forms, grammar rules etc. In the discussion page I’ve shown how different a Montenegrin Wikipedia article would be from an article in Serbo-Croatian to Mr Amir Aharoni, please, be free to check that as well, and I can do that for any article you want me to, so you have a side by side comparison, just like the one I made there. A very large number of Croatian words is unheard of and unknown in Montenegrin, to the point where you simply cannot understand whatever you are reading about (scientific topics to the point of unintelligiblity!). And Croatian is vastly dominant in Serbo-Croatian Wikipedia (alongside those Ekavian Serbian parts in the middle of a single sentence). You can agree that this is not a good ground for developing a serious project written in Montenegrin language.
Finally, I am aware that these other projects are grandfathered, but it doesn’t change the fact that they do exist, and Montenegrin community sees this as very unfair (to say the least) and really feels that based an all of the things listed above, and this fact that it’s the only SC variety without it’s own Wikipedia, it does deserve creation of it’s own project.
Sincerely, Luka (Wiki username: Lujki).”
Sent from Outlook<http://aka.ms/weboutlook>
I am about to send two messages concerning Montenegrin Wikipedia.
The first is a letter that was posted on-wiki from a member of that community in support of his project. I know the RFL page on the project is now a massive wall of writing, and my sense is that not all of you have dug into it. I believe this letter summarizes his community's arguments pretty well; if it motivates you to read some of the rest of the massive wall, that's great. But it gives you sense of their point of view.
The second is my promised argument in favor of the project. Before I send this argument, though, I want to make something very clear to everyone. I'm not from the Balkans in any manner whatsoever, Jewish or gentile, to the best of my knowledge. One branch of my wife's family comes from Hungary, and that's about as close as we get. So please understand that I have no personal interest in the outcome. I am making this argument because I believe that at this point it reflects the appropriate decision within the rules of WMF and the Language Committee. Thank you for hearing me out.
Steven
Sent from Outlook<http://aka.ms/weboutlook>
The page "State of Wikiversities<https://beta.wikiversity.org/wiki/States_of_Wikiversities>" gives an excellent overview of the current situation. To summarize:
There are 53 languages whose root categories at Beta Wikiversity (e.g., Category:YUE) have at least one page in them.
* Of those 53, Chinese<https://beta.wikiversity.org/wiki/Category:ZH> has nearly 800 pages in its root category, and five other tests have from about 375 down to about 260 pages. No other test has even 100 in its root category. (Caveat: I haven't tried to see if there are pages in subcategories but not in the language's root category.) Based on the list at Meta<https://meta.wikimedia.org/wiki/Wikiversity>, I will assume for now that fewer than 100 pages (and possibly even fewer than 200 pages) is considered prima facie "too small".
* Exactly four tests show some activity nearly every month, and a fifth test is worth a mention.
* Chinese: As I wrote recently, if one counted the substantial contributions by IP editors, there would be easily be sufficient activity to make this Wikiversity test approvable. This test had long stretches between 2012 and 2015 when it always met the activity requirement outright. I do not know why the project was not approved then, but it is certainly possible that the lack of approval at that point eventually discouraged some contributors. Given that this Wikiversity is also much bigger than any other test here—it would be around #10 in size of all Wikiversities if approved—LangCom should consider whether or not this should be approved now, even if the activity requirement is "not quite" met.
* Ukranian<https://beta.wikiversity.org/wiki/Category:UK> (305 pages) and Hebrew<https://beta.wikiversity.org/wiki/Category:HE> (87 pages) have some activity every single month. Neither has ever had three consecutive months of three registered users with ten edits each, though Hebrew had a stretch of 19 months during 18 of which it had at least two registered users with ten edits each.
* Estonian<https://beta.wikiversity.org/wiki/Category:ES> (262 pages) has some activity most months. It had stretches in 2009-2010 when it met the current eligibility requirement, but it's not really close at present.
* Lithuanian<https://beta.wikiversity.org/wiki/Category:ZH> (326 pages) met the eligibility requirement during nearly all of 2015, but has been completely dormant since. Again, I do wonder if the lack of attention at a certain point eventually discouraged some contributors.
* The other two "large enough" tests (Bulgarian and Persian) have a certain amount of intermittent activity, but nothing regular in the last couple of years.
As for the proposal to merge Beta into Incubator, my official position is to be neutral on this, and ultimately I am. But, frankly, from a strictly pragmatic perspective, I don't need the extra load in Incubator right now. And I don't need a bunch of former Beta administrators telling me I'm not handling their projects correctly. So either don't do it, or at least don't be in so much of a hurry to do it. It's easy enough to keep track of what's going on over there.
Steven
Sent from Outlook<http://aka.ms/weboutlook>