MF-Warburg wrote: "Let's not forget that the creation of new wikis is stalled at the moment for technical reasons. My motivation to advance approvals is not very high due to it, since I dislike approving projects which are active at a time, but might not still be active later when the technical issues are resolved but are then created nevertheless."
Is that still not resolved? I thought creation of Neapolitan Wikisource showed we were moving past that. (Of course, I thought that about Western Armenian Wikipedia, too.)
Also: It's a very bad problem when the projects become inactive expressly because their communities see us as incapable of creating their wikis, so the communities throw up their hands and go home. It's not because they don't care about this, but because they feel that if they will never get anywhere, they might as well not waste more time. And it's our fault, not theirs. As Marcos Williamson says, we need to find a middle ground between protecting against hoax and empowering these communities. My original proposal here was intended to give us time to do due diligence, but not unlimited time. Time is fair; unlimited time is not. And to those who complained that I was setting arbitrary time limits, I would respond that I was setting time limits. Any time limit is arbitrary at one level, but there needs to be something compelling LangCom to act instead of to sit on its hands, which has happened a lot recently.
MF-W, once the technical issues are resolved, I plan to allow the communities closest to "ready", if they have gone inactive, to be considered active again as soon as they have a single month with three active users at 10+ edits. Making them wait another three months is just rubbing salt in the wound.
Steven
Sent from Outlook<http://aka.ms/weboutlook>
I'm happy to wait for your proposal, Amir. Still, if there is something that is going to require LangCom approval, there needs to be something in the proposal that compels LangCom to act within a reasonable time frame. Yes, we're volunteers, but so are the people who are creating the projects in Incubator (etc.), and they deserve not to have to wait a year.
I will also add that as long as I'm not so frustrated with the system that I can stay active, I do watch the active tests on Incubator. Yes, it's possible for me to be fooled by a hoax, as I don't usually read the languages involved. Still, as long as Incubator/Beta/Old Wikisource are being monitored, it needs to be a pretty robust hoax on the concerted part of a reasonable number of contributors in order to make it all the way to the point of approvability.
Finally, then, is the tradeoff between avoiding hoaxes and being responsive to new-project communities. The chances that such a community is a good-faith community that is being hurt by delay are probably much greater than the chances that such a community is a hoax community. So while I think we need a chance for our due diligence, we should not be allowed an arbitrarily long time to do that due diligence, either.
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: Wednesday, August 21, 2019 8:00 AM
To: langcom(a)lists.wikimedia.org <langcom(a)lists.wikimedia.org>
Subject: Langcom Digest, Vol 70, Issue 12
Message: 1
Date: Wed, 21 Aug 2019 08:52:07 +0300
From: "Amir E. Aharoni" <amir.aharoni(a)mail.huji.ac.il>
To: Wikimedia Foundation Language Committee
<langcom(a)lists.wikimedia.org>
Subject: Re: [Langcom] Proposed amendment to LPP (was: Final approval
for four projects)
Sorry, no.
No verification at all and simply assuming good faith after some arbitrary
time is asking for trouble. There already were cases of hoaxes in the past.
Not many cases, but they did happen, and we're talking about a whole site,
even one case is major trouble.
HOWEVER, I absolutely do recognize that we have unnecessary and harmful
bottlenecks in the approval process, and because of that I'm working on
another proposal that will ease up at least some of these bottlenecks.
I've just came back from Wikimania, and I'm immediately going for a
vacation until the end of August, but after that—expect surprises,
hopefully good ones.
--
Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי
Gerard wrote:
<start>
Hoi,
Define unreasonable. Amir gave an estimation when it is reasonable to
expect a result.
Thanks,
GerardM
<end>
For the current instance, based on where we are now, Amir's estimation is fine. As long as someone is actually trying to follow up on these projects, and gives a reasonable estimate as to how long it will take, I'm fine.
James's question, and mine, is more around the big picture. Please remember that three of the four projects we are talking about here (Guiane Creole WP, Saraiki WP, Tacawit Wiktionary) were identified in an email I wrote on 17 December 2018 (about eight months ago) as already being provisionally approved and awaiting language verification. The fourth (Mon WP) is perhaps a month newer. But 7-8 months and longer—Saraiki WP was provisionally approved in October 2018—is absolutely not reasonable by any standard. Quite frankly, I was desperate to do something to move these along, because being nice and playing by the rules was doing absolutely nothing. (Remember, too, that I wrote pleasant, polite reminders to the committee about these four projects on March 14 and June 6.) I'm sorry, Gerard, that you didn't like me doing what I did. But what I did is far less objectionable than requiring communities to wait this long for us to complete language verification.
To that end, I am proposing the following amendment to the provision about language verification. I am open to some adjustments, but allowing projects to sit this long and wait for us is just not acceptable. Where this amendment is to be added is in the Language Committee's Handbook, Final Approval<https://meta.wikimedia.org/wiki/Language_committee/Handbook_(committee)#Fin…>, item #2. Subitem #3 is to be followed by new subitem #4:
4. The Language Committee has 30 days from the time a project is provisionally approved—meaning: approved, except for language verification—to identify and contact an expert for the language verification. If no expert is contacted within 30 days, then on the assumption of good faith, the project will be finally approved. If an expert is contacted within 30 days, the Language Committee has an additional 60 days to obtain the final language verification. If no language verification (or failure of language verification) is received by then, on the assumption of good faith, the project will be finally approved. Overall, any project for which the Language Committee has failed to get language verification (or failure of language verification) within 90 days will be approved on the assumption of good faith.
I think this gives us plenty of time to do what we need to do, without requiring communities to wait on us for months without comment. This would apply for all projects receiving provisional approval from this point on. But in parallel, given that Amir started working on these four projects around August 15, I would also propose that if we have not finalized language verification by October 15, these four projects also be finally approved.
Steven
Sent from Outlook<http://aka.ms/weboutlook>
As long as Amir is actually trying to contact experts, I am ok to wait. The fact that no one on the committee was bothering to work on language verification was my principal problem.
And, Gerard, when you start doing your job on this committee, I will answer you.
Steven White
koala19890(a)hotmail.com
Sent from my iPad
Now that things are starting to move again, I intend to approve four projects: Guiane Creole WP, Saraiki WP, Mon WP and Tacawit (Shawiya) Wiktionary. No, none of these has been verified. But this is not the fault of the people who created the projects, this is the fault of the Language Committee, which did not do its job. Accordingly, I will approve and send to phabricator for creation all four of these projects by 17:00 UTC on Monday, 19 August. I will not do that in a particular case only if ALL of the following happens:
* Someone objects in a particular case that they have reason to be concerned about the validity of the language.
* That person also specifically commits to contacting an expert immediately.
* Then, within 48 hours, that person has contacted the expert, and identified the expert to the mail list.
* Then, the expert has 7 days to respond.
Otherwise, I need to fall back on "Assume Good Faith". And look, if one of these turns out to be another Siberian, we'll be embarrassed, and we'll delete the project. But I've been watching all of these projects for the last two years, and I don't have any reason to believe there is a problem.
If anyone tries to object in any way other than the specific way I have outlined above, I intend to ignore that person. Sorry, but at this point, the Committee only has the right to intervene if it intends to become active again and do its job.
Steven
Sent from Outlook<http://aka.ms/weboutlook>
Hi,
In Wikidata there was a proposal to add special fallback pseudo-languages
to make it possible to have labels that work by writing system rather than
by language:
https://www.wikidata.org/wiki/Wikidata:Project_chat/Archive/2019/06#Multila…
The proposal was to use language codes such as mul-latn, etc.
It kind of makes sense to me, but I'd like to ask the language code
standards experts in the committee: is such usage standard?
--
Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי
http://aharoni.wordpress.com
“We're living in pieces,
I want to live in peace.” – T. Moore
Hi,
I suggest approving the N'Ko Wikipedia at nqo.wikipedia.org .
The Incubator[1] has been fairly active, the translation of all the most
used messages is complete[2] and the localization activity is going on[3].
Dr. Coleman Donaldson, a linguistic anthropologist who investigates
language, literacy and society in Manding-speaking West Africa,[4] and
author of several papers on N'Ko language and culture says that the texts
in the Incubator are indeed written in N'Ko (see the forwarded email below).
Any objections?
[1] https://incubator.wikimedia.org/wiki/Wp/nqo
[2]
https://translatewiki.net/wiki/Special:Translate?filter=%21translated&actio…
[3] https://translatewiki.net/wiki/Portal:Nqo
[4] https://www.colemandonaldson.com/
--
Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי
http://aharoni.wordpress.com
“We're living in pieces,
I want to live in peace.” – T. Moore
---------- Forwarded message ---------
מאת: Coleman Donaldson <colemandonaldson(a)gmail.com>
Date: יום ב׳, 5 באוג׳ 2019 ב-10:39
Subject: Re: N'Ko incubator
To: Amir E. Aharoni <amir.aharoni(a)mail.huji.ac.il>
Morning!
I just had a look and a few random pages and indeed, they were all written
in "proper N'ko".
Cheers,
Coleman
On Sun, Aug 4, 2019 at 12:24 PM Amir E. Aharoni <
amir.aharoni(a)mail.huji.ac.il> wrote:
> Hi,
>
> How are you?
>
> We spoke about the N'Ko Wikipedia incubator once.
>
> I am thinking about approving it for a full domain, which will be at
> https://nqo.wikipedia.org .
>
> For this we need a confirmation from somebody who knows the language that
> this is indeed written in it. Can you please check that the pages in the
> list at
> https://incubator.wikimedia.org/wiki/Special:PrefixIndex?prefix=Wp%2Fnqo&na…
> look like they are written in proper N'Ko?
>
> Thanks!
>
> --
> Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי
> http://aharoni.wordpress.com
> “We're living in pieces,
> I want to live in peace.” – T. Moore
>
--
Coleman Donaldson
+49.1525.1587689 (Germany)
+1.267-269-6243 (US)
Amir,
Thank you for responding. If I had to summarize things in one sentence, it would be that on the whole the members of this committee mostly seem not to be involved in any kind of real way.
This is particularly true on-wiki. In my view, on-wiki is still supposed to be the core of what this is all about. And other than you, MF-Warburg and occasionally Satdeep, I can hardly remember a member of the Language Committee responding—for example at Talk:Language committee on Meta. I’m supposed to be a clerk, not necessarily even responding substantively. I do, and that’s OK to a point, but can’t anyone else ever be there?
Still, I’d venture my biggest frustration is with the fact that people on the committee are not even doing minimal work based on what gets posted to this mail list. For example: There are four projects that were tentatively approved in October-December 2018, pending language verification. I pointed this out again in an email on March 14. And I pointed it out again in an email on June 6. And nothing has happened; there have barely even been grunted responses from the Committee. If there is one piece of business I really have not wanted to get involved in, it’s discussing language eligibility with experts. Can’t anyone act on these? Really? How long do we let a project sit in this status without verification? At some point, these people deserve an answer, and if we cannot get verification—and we assume good faith—that answer should be yes.
(Related to both of the above, I am going to write the committee, probably tomorrow, about two other requests: one to close Bulgarian Wikinews, and the other the ongoing discussion about this Literary Chinese Wikisource request. I have recommendations for both, but could people really look on-wiki at the discussions, and not just depend on my summaries? Please? And then engage in a real discussion? These are not simple questions.)
I’m also frustrated with something that the committee can’t entirely control: the technical block at Phabricator preventing creation of new wikis. Now part of the frustration stems from the fact that I’ve used that as an excuse not to bring new wikis here for tentative approval. (The reality of why I haven’t done that, of course, is at least as much about the fact I have such a hard time securing language verification from the Committee.) But part of it is real: how long do people have to wait before getting their projects approved?
I’ll add: Phabricator seems to be moving on recoding some of the projects that use non-standard language codes, or at least adding the standard codes as aliases. That’s fine as far as it goes, and I’ll admit that this is probably easier than fixing the real problem blocking new wiki creation. Still, can’t we somehow try to get the developers to put a higher priority on new wiki creation? There are a few people who care a lot about the code issue, but in nearly all cases, that hasn’t been a really important problem. The fact that we can’t create new wikis is a really important problem.
And I’ll conclude by noting that people are starting to ask, “Why can’t we get verification now, notwithstanding the technical problems?” The only honest answer I can give is that LangCom doesn’t have its act together to do verification. If people want me to give that answer online, then I will. Better: get these verification requests moving, so that I can start pushing other projects along.
Thank you for listening.
Steven
Sent from Outlook<http://aka.ms/weboutlook>
(Amir wrote:)
Hi,
There's a phenomenon on wikis: Sometimes, when a person becomes an expert
at doing some kind of routine work, everyone else gets used to it and
decides that they don't have to do it, because the expert will take care of
it. No-one even bothers to learn the expert's modus operandi.
Unfortunately, the possibility that the expert will burn doesn't occur to
anyone. I am guilty of this myself.
Steven, I appreciate your work immensely. As a clerk, you reinvigorated the
Language committee's work and improved our processes.
Please let me where can I help, and I'll try my best.
--
Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי
https://nam02.safelinks.protection.outlook.com/?url=http%3A%2F%2Faharoni.wo…
“We're living in pieces,
I want to live in peace.” – T. Moore
בתאריך יום ב׳, 5 באוג׳ 2019 ב-1:13 מאת Steven White <
I need help. I'm doing all the work right now, and it's not fun any more. If I don't see some real commitments from some of you to get back involved ON-WIKI in about the next week, I intend to resign as clerk and as sysop on Incubator, and I'll leave the mess behind for the rest of you.
Steven White
koala19890(a)hotmail.com
Sent from Outlook<https://aka.ms/sdimjr> for iOS