Jump to content

Wikipedia:Village pump (WMF)

From Wikipedia, the free encyclopedia
 Policy Technical Proposals Idea lab WMF Miscellaneous 
The WMF section of the village pump is a community-managed page. Editors or Wikimedia Foundation staff may post and discuss information, proposals, feedback requests, or other matters of significance to both the community and the Foundation. It is intended to aid communication, understanding, and coordination between the community and the foundation, though Wikimedia Foundation currently does not consider this page to be a communication venue.

Threads may be automatically archived after 14 days of inactivity.

Behaviour on this page: This page is for engaging with and discussing the Wikimedia Foundation. Editors commenting here are required to act with appropriate decorum. While grievances, complaints, or criticism of the foundation are frequently posted here, you are expected to present them without being rude or hostile. Comments that are uncivil may be removed without warning. Personal attacks against other users, including employees of the Wikimedia Foundation, will be met with sanctions.

« Archives, 1, 2, 3, 4, 5, 6, 7

Wikimedia Foundation Bulletin: an experiment

[edit]

Hi all. We invite your feedback on a proposed way to improve communication from and about the Wikimedia Foundation. The Wikimedia Foundation Bulletin is an experiment to establish a more standardised format and cadence. It would include headlines and links from the Wikimedia Foundation's technical work; Foundation activities with communities and affiliates; as well as with other stakeholders like readers, donors, regulators, the media, and the general public.

A short overview of the concept itself is on Meta at m:Wikimedia Foundation Bulletin, with the first “trial” issue at m:Wikimedia_Foundation_Bulletin/2024/06-01 - also copied below. You can subscribe to the bulletin via talk page delivery on any Wikimedia wiki. Depending on the feedback received, we might start this as a regular Bulletin for the coming fiscal year (which starts July 1).

This is an experiment: we want to know what you think, what is missing, what is too much, and whether this is something that we should consider investing more time and effort into. Please post your feedback on the Bulletin talk page - on the concept itself, and suggestions on anything from the design to specific words used would also be helpful. You can also provide feedback in this thread; by email to askcac@wikimedia.org; or at the next Conversation with the Wikimedia Foundation’s Board of Trustees on 27 June at 18:00 UTC.

On behalf of the Wikimedia Foundation Community Affairs Committee, MPeel-WMF (talk) 18:38, 18 June 2024 (UTC)[reply]

I've put the second issue below as well, and have signed this page up to receive them automatically in the future. Thanks. Mike Peel (talk) 12:38, 3 July 2024 (UTC)[reply]

Bulletin June 2024

[edit]

MPeel-WMF (talk) 18:38, 18 June 2024 (UTC)[reply]

Wikimedia Foundation Bulletin June Issue 2

[edit]

Mike Peel (talk) 12:38, 3 July 2024 (UTC)[reply]

These are excellent. Please keep them up, we could find a better way to integrate links to such newsletters from a global news page on Meta as well. [perhaps alongside one-line links out to the latest newsletters on individual projects] – SJ + 17:59, 13 July 2024 (UTC)[reply]
thanks, Sj! just to clarify, are you talking about this page: Internal news media? --アンタナナ 10:06, 15 July 2024 (UTC)[reply]
I happened to view this with a different browser zoom level today, and FYI in case it helps, I find the one column version more readable than the two column version. –Novem Linguae (talk) 06:03, 14 July 2024 (UTC)[reply]
Thanks @Novem Linguae, very helpful - we've had the same feedback about the two column version on French Wikipedia too so for the next issue (or maybe the one after that) we'll make the switch to single column. MPaul (WMF) (talk) 09:30, 15 July 2024 (UTC)[reply]
@Novem Linguae just circling back to say we switched to a single column version. Thank you again for the feedback! MPaul (WMF) (talk) 13:50, 26 July 2024 (UTC)[reply]

POTY (Picture of the Year) competition needs help!

[edit]

POTY desperately needs new volunteers who can do the things required to run the competition. With the current state of the committee, it is likely that there will be no POTY this year, as the main member who ran scripts for the competition has burned-out from doing wikipedia tasks and isn't up for it. Others on the committee are also missing in action.

Check out the Discussion here [1]. •Shawnqual• 📚 • 💭 03:47, 6 July 2024 (UTC)[reply]

Consider posting to WP:VPT where our programmers hang out, and consider including in your post links to https://github.com/legoktm/poty-scripts and to https://poty-stuff.toolforge.org/ so that technical folks can easily examine the scripts. –Novem Linguae (talk) 17:57, 6 July 2024 (UTC)[reply]
I posted on VPT and did not get any replies! :-/ The section was just archived •Shawnqual• 📚 • 💭 07:29, 15 July 2024 (UTC)[reply]
@Shawnqual: Anyone figure out a solulu for this one yet? If not I may be able to pitch in. jp×g🗯️ 12:28, 22 July 2024 (UTC)[reply]
This might be the latest repo: https://gitlab.wikimedia.org/toolforge-repos/poty-stuffNovem Linguae (talk) 20:01, 22 July 2024 (UTC)[reply]
@JPxG: Nop. We're still stuck in a limbo here and what seems to be a dead end. Please help if you can! •Shawnqual• 📚 • 💭 23:53, 26 July 2024 (UTC)[reply]

 You are invited to join the discussion at Wikipedia:Village pump (miscellaneous) § U4C Special Election - Call for Candidates. –Novem Linguae (talk) 03:33, 10 July 2024 (UTC)[reply]

Living without the WMF?

[edit]

The political evolution of the US is worrying, given that the WMF is based there. What if in a few years the US government takes control of the WMF, seizes its assets, or becomes otherwise hostile? Can Wikipedia as we know it survive without being based in the US? Are there plans for decentralization or redundancy? Sylvain Ribault (talk) 19:08, 18 July 2024 (UTC)[reply]

Without speaking to the political situation, I will note that Wikipedia is backed up on server farms in other countries, and anyone can operate a clone of Wikipedia from anywhere, even if they could not use the name "Wikipedia". Donald Albury 20:27, 18 July 2024 (UTC)[reply]
This sort of doom-mongering is never helpful. * Pppery * it has begun... 20:32, 18 July 2024 (UTC)[reply]
Interesting question. The WMF itself is pretty much entirely centralised, but it's always been pretty good at making it easy to mirror or fork Wikipedia. Our license is of course also a big help. So in this kind of scenario, I imagine preserving the content would be no problem at all, but reassembling the community would be difficult, and rebuilding the kind of financial resources the WMF has (to host, maintain, and develop Mediawiki) would be very challenging indeed. – Joe (talk) 20:42, 18 July 2024 (UTC)[reply]
And to what extent could the existing Wikimedia chapters help? How dependent are they from the WMF? Sylvain Ribault (talk) 09:01, 19 July 2024 (UTC)[reply]
Whatever dystopian future awaits the US, that the WMF would be taken over by political hacks is 3 or 4 apocalypses removed from reality. Headbomb {t · c · p · b} 11:59, 19 July 2024 (UTC)[reply]
What specifically in regard to wikipedia do you find worrying about American politics? Horse Eye's Back (talk) 17:24, 19 July 2024 (UTC)[reply]
I think the main actual obstacle is that Wikipedia's existence depends on a fairly high degree of active maintenence of the MediaWiki software, and also -- most crucially -- that search engines give us a gigantic volume of incoming traffic. Incidentally, the forks that have existed have routinely had trouble with being absolutely slaughtered in Google rankings because their content is all considered by the algorithm to be "plagiarized" from Wikipedia. jp×g🗯️ 12:26, 22 July 2024 (UTC)[reply]

Sunday July 28 Strategic Wikimedia Affiliates Network meeting (Results of Movement Charter ratification)

[edit]
SWANs gathering for a conversation

Hello everyone!

The Strategic Wikimedia Affiliates Network (SWAN) is a developing forum for all Wikimedia movement affiliates and communities to share ideas about current developments in the Wikimedia Movement. It expands on the model of the All-Affiliates Brand Meeting (following the re-branding proposal by the WMF) to help lay some of the groundwork for further Wikimedia 2030 strategy process work.

At this meeting we will focus on the results of the Movement Charter ratification. We will also discuss the aftermath of the Board of Trustees' decision to veto the Movement Charter, including their recent proposals. We will also cover updates about upcoming Wikimania 2024.

This month, we are meeting on Sunday, July 28, and you are all invited to RSVP here.

UTC meeting times are and

Nadzik (talk) 17:35, 20 July 2024 (UTC)[reply]

[edit]

One of our most important tools, Earwig's Copyvio Detector, depends on access to Google. According to the tool's creator and operator, The Earwig, the WMF has kindly been paying for this Google access. Unfortunately, we've been hampered by a strict limit on the number of searches allowed per day. The Earwig mentioned that there might be a way to work out a special arrangement with Google to increase the cap. Would someone at the WMF be able to pursue this?

In case it helps, this is a vital tool to a number of English Wikipedia processes, and it would surprise me to learn that the sister projects aren't using it as well. We use the tool routinely as part of our new page patrol, articles for creation, contributor copyright investigations, did you know, good article, and featured article processes. Historically, the WMF has taken a special interest in supporting volunteer work that focuses on our legal responsibilities, of which compliance with copyright law is an obvious example. Firefangledfeathers (talk / contribs) 01:34, 31 July 2024 (UTC)[reply]

My understanding is that Google has a hard daily limit of 10,000 API accesses per day for absolutely everyone across the board, without exception. User:Novem Linguae/Essays/Copyvio detectors#Earwig copyvio detector. My impression is that an exception wasn't possible because Google doesn't provide an exception to anyone. Earwig would know best though.
Was this post made because Earwig said "please ask WMF on my behalf to negotiate with Google", or is this more of general question? –Novem Linguae (talk) 04:15, 31 July 2024 (UTC)[reply]
cc Chlod. –Novem Linguae (talk) 04:18, 31 July 2024 (UTC)[reply]
Earwig didn't ask me to do anything on his behalf. He mentioned that "the WMF pays for it, but Google's API terms limit our usage without some kind of special arrangement that I have been unable to get." This was at a discussion at his user talk. I wasn't sure who might be able to negotiate a special arrangement, and I'm not sure it's a possibility, but this was the best place I could think to ask. Firefangledfeathers (talk / contribs) 04:26, 31 July 2024 (UTC)[reply]
A bit odd that Earwig isn't doing the advocating himself, but on the linked user talk page, it does sound like he's asking for some help with this. Would The Earwig be willing to share his contacts at WMF that have helped with this in the past? Sounds like WMF pays for the tool, so there's some accounting/finance/grants contact that knows a little about it. And we also have partnerships people like NPerry (WMF) that I believe has worked with Google before. –Novem Linguae (talk) 04:43, 31 July 2024 (UTC)[reply]
From what I've heard, the WMF contact that Earwig had has since left the Foundation and wouldn't be able to help in this case. You are correct that WMF pays for the tool. I had mentioned this at the Hackathon with staff and it seems there's some resistance in getting the cost of extra tokens funded, although I'm unsure of exactly how the WMF's budgeting process works, so no clue on the impact it has in this situation (considering we don't have a Google liaison to begin with). Chlod (say hi!) 05:20, 31 July 2024 (UTC)[reply]
Even the name of a former WMF employee contact would be helpful. Let's get all this documented so we can start figuring out what WMF departments/teams have assisted in the past. –Novem Linguae (talk) 05:26, 31 July 2024 (UTC)[reply]

Hi all. Firefangledfeathers, thanks for starting the discussion. Novem, sorry if this thread came up a bit strangely. In truth, I struggle a little with motivation these days, so I really appreciate others' help getting the ball rolling. I am still here, though—it comes in waves. (It's also good to involve the community in the tool so the institutional knowledge isn't stuck with me.)

BTW, I am working on more effectively managing automated/excessive tool usage and will soon require OAuth to run searches (see this active thread on my talk). Right now the tool really doesn't have any usage guards or a way to limit individual users' activity, which isn't good when our resources are so limited. It's possible doing that will free up our resources a lot if a substantial fraction of our current usage is coming from malicious crawlers, despite Chlod and I's attempts at blocking them (the tool has been running for over a decade and it's never been this bad, though I have a theory what this is about). Even so, finding a way to increase our search quota will enable us to support some requested features that are current nonstarters, even if the tool's entire current quota could be devoted to it, like checking all new pages.

My main point of contact with the WMF in the past was Kaldari. The last time we spoke about the tool was 2020; since then, the situation has been unclear. (MusikAnimal, do you remember if we've spoken about this?) Last year Runab WMF and DTankersley (WMF) reached out to me to discuss the tool in the context of WMF efforts "to find ways to reduce single points of failure for tools that require a third party API", but after an initial conversation I haven't heard back aside from being told that Deb was moved to another project, so I'm not sure what happened with those efforts.

Frequently we've discussed adding an alternate search backend aside from Google. While Google is really the gold standard for breadth of search coverage, as far as I'm aware—and this is really what the copyvio detector needs, not necessarily quality/intelligence; people have suggested services like DuckDuckGo, but they're really unsuitable because they just republish raw results from Bing with some additional flair that is basically useless for us—something like Bing itself might work as an (automatic) fallback if we exhaust our Google credits for the day. I believe Bing has roughly equivalent pricing/usage limits as Google, but it's been a while since I've looked into it. And we/the WMF would need to establish a relationship with Bing for that to work; I don't know if that's a better idea than attempting to negotiate our Google limits. There are also other options like Yandex (which the tool did use one dark time in the past before the Google relationship and after Yahoo ended their free service... it wasn't great, at least for English results, but it's something that could be looked into for some other language projects, perhaps). Finally, there was a discussion on my talk earlier this year with Samwalton9 (WMF) about adding The Wikipedia Library as another search backend, and I did correspond briefly with someone at EBSCO about this, but again, I haven't heard from either of them about this in several months. — The Earwig (talk) 06:51, 31 July 2024 (UTC)[reply]

So the Google API proxy and the Google account it runs on are wholly part of Community Tech's budget. Kaldari was the contact in the past when they were my manager on CommTech. So the good news is Community Tech is still here, and we are actively maintaining this proxy (I just migrated it to a newer Debian a few days ago). The part that hasn't changed is our quota from Google, and sadly I doubt it will change. We are already paying hefty fines for the quota we have now, but I believe it is also correct that 10K is a strict limit from Google. I can see from the graphs in the API console that we almost always hit that limit within the first 12 hours of each day.
I am working on more effectively managing automated/excessive tool usage and will soon require OAuth to run searches … – that is most certainly the best immediate recourse for addressing this problem. From my years of shielding XTools from web crawlers, I can say with confidence that putting up a login wall by itself should make a big difference. I also think mitigating excessive and automated use is something that would probably be required before we could consider dishing out more money to Google. However again, I don't think such negotiations would get us anywhere anyway :(
As a general note, such "negotiations" are typically done these days via the Partnerships team. I went though them recently when we solidified our partnership with Turnitin. Speaking of which… do others find the "Use Turnitin" option of Copyvios at all useful? Because that's using the old Turnitin account (the new one can't be used outside CopyPatrol), and the last I checked there were still a few million credits left. MusikAnimal talk 20:52, 1 August 2024 (UTC)[reply]
Thanks for those details. So I can improve my notes at User:Novem Linguae/Essays/Copyvio detectors#Earwig copyvio detector, do you know why/how Google API Proxy ended up separate from the main tool? And does "paying hefty fines" mean that there is some sort of sliding scale of pricing and that getting near the cap gets more expensive? My notes currently state that Google API credits cost us $50/day. –Novem Linguae (talk) 00:52, 2 August 2024 (UTC)[reply]

Wikimedia Foundation Bulletin July Issue 2

[edit]
Subscribe or unsubscribe · Help translate

Previous editions of this bulletin are on Meta. Let askcac@wikimedia.org know if you have any feedback or suggestions for improvement!


MediaWiki message delivery 21:48, 1 August 2024 (UTC)[reply]