Wikimedia Forum

← Discussion pages

Wikimedia Forums

Archives →

The Wikimedia Forum is a central place for questions, announcements and other discussions about the Wikimedia Foundation and its projects. (For discussion about the Meta wiki, see Meta:Babel.)
This is not the place to make technical queries regarding the MediaWiki software; please ask such questions at the MediaWiki support desk; technical questions about Wikimedia wikis, however, can be placed on Tech page.

You can reply to a topic by clicking the "[edit]" link beside that section, or you can start a new discussion.

Wikimedia Meta-Wiki

• About Meta
• Discussion pages
• Request pages
• Policies and guidelines
• Information and statistics
• Categories
• Help pages
• Meta-Wiki discussion page

SpBot archives all sections tagged with {{Section resolved|1=~~~~}} and sections whose most recent comment is older than 30 days.

Any datasets / Wikipedia statistics to visualize?

I don't just view AI as a 'threat' to Wikimedia and don't think the optimal state would be for generative AI to just use Wikipedia instead of Wikimedia also using generative AI to their benefit.
One area of useful application is the creation of data visualizations. Below you can see some examples of what has become possible; I made these using genAI and matplotlib.

• 
  $depth^{*}={\frac {words}{articles}}\cdot {\frac {non{\underline {}}articles}{pages}}$ metric data at User:Tim Ocean
• 
  Community Wishlist chart showing fraction of wishes done done
• 
  Count of categories with open Category for Discussion (CfD) cases over time on Commons

Is there any other data you'd like to see visualized? Especially of interest are metawiki pages with tables of data that have not yet been visualized in a chart. Prototyperspective (talk) 17:46, 17 March 2026 (UTC)

@Prototyperspective: very interesting visualizations. Would it be possible to show changes over time for this depth? E.g. as a gif? For the ~50 largest wikis (largest by number of active editors)?

And hmm, could you develop a quality indicator like this:

Number of all citations used in all articles (<ref> tags or/and appropriate templates) / Number of words

×

Number of articles with any sources (<ref> tags or/and templates) / Number of articles without sources (0 <ref> tags and templates)

×

Number of featured + good articles / Number of stubs (< 100 words)

THX :) Hedger z Castleton (talk) 16:13, 21 March 2026 (UTC)

This question would need to be directed to Tim Ocean. I'd also like to mention that not all Wikipedias are included there (the largest and some apparently random ones) and it's not a good indicator either (may well be better than the original depth metric but eg German Wikipedia probably doesn't score worse than the Wikipedias above it – it just has a practice of some cleaner state of fewer ancillary pages wherever avoidable).

I'm not collecting the data – I just visualized the data so your idea is best put on some requests page; maybe Tim Ocean is able/interested in getting data for a metric like that set up(?) Prototyperspective (talk) 20:05, 21 March 2026 (UTC)

@Tim Ocean:, could you...? ^^ Hedger z Castleton (talk) 12:26, 30 March 2026 (UTC)

Hi. I would likely analyze all Wikis using my statistics and variants of it, but I need to learn how to scrap Web pages first, because there are hundreds of Wikis :D Or somebody would need to do it and provide a data set. Tim Ocean (talk) 15:16, 30 March 2026 (UTC)

*gladly Tim Ocean (talk) 15:19, 30 March 2026 (UTC)

• 
  pageviews with added 7-day moving average
• 
  spikes clipped
• 
• 
  ENWP AfC decline reasons
• 
  Uses of charts in English and German Wikipedia by year of latest data of the charts (circa how many outdated charts)

Here's some more. The Pageviews Analysis display was really bad to put it mildly, showing the value of each day without some moving average to be able to make sense of a chart, and often to be able to see it at all. The first image is just the 7-day moving average added. The other two had their spikes clipped (here is how it looks like with spikes). If you have data to visualize I suggest you make a matplotlib chart out of it so people can see and make sense of the data or send it my way and I'll probably make one. --Prototyperspective (talk) 17:34, 23 April 2026 (UTC)

Invitation to Global Call on Annual Planning

Community members are invited to a global call to discuss the latest trends affecting our movement, as part of the Wikimedia Foundation’s annual planning process. We will be sharing our global trends analysis, specifically focusing on Contributors and Readers. There will also be a demo and breakout discussions.

This is a great opportunity to meet Foundation leadership – including our new CEO, Bernadette Meehan – as she listens, learns, and engages with communities around the puzzle globe.

Join us for this workshop to share your thoughts on how people access and share knowledge, and help shape the Wikimedia Foundation’s Annual Plan.

Date: Thursday, May 7, 2026

Time: 5:00 PM UTC – Check your local time here

RSVP: Register at the event page to get the link to the meeting. Eureka-WMF (talk) 19:09, 24 April 2026 (UTC)

Wikinews

where do people go now? BigKrow (talk) 23:12, 24 April 2026 (UTC)

There is a dedicated discussion page: Migration of Wikinews and future hosts. AtUkr (talk) 12:01, 25 April 2026 (UTC)

Request for comment (global AI policy)

A request for comment is currently being held to decide on a global AI policy. Thank you! MediaWiki message delivery (talk) 00:57, 26 April 2026 (UTC)

Notify editors of a new-language proposal in languages which they speak

At Requests for new languages, proposals often languish with little input, I imagine largely because nobody knows about them. User categories for language proficiency exist on 243 Wikipedia projects (and presumably more on other projects), imo a global bot should be created that notifies users when there is a proposal in a language they speak. Obv spam is a concern which can hopefully be addressed. I don't know if this is the right place to discuss, made a wish as well. Thoughts? Kowal2701 (talk) 14:27, 30 April 2026 (UTC)

RNLs do not need a lot of input, especially not if the language is already well-established as a language used by Wikimedia users. So that would not be very helpful. --MF-W 21:41, 1 May 2026 (UTC)

Hi, I'm not sure I understand what you mean, AFAICT an incubated project needs an active group of editors for it to be launched. Using Xiang Wikipedia as an example, this proposal would notify editors who have en:Category:User hsn (incl. its sub-templates) on their user page (of which there are 4 on enwiki alone) that there is an incubated Xiang-language wiki that they can join and help start Kowal2701 (talk) 00:44, 2 May 2026 (UTC)

If a test-project has active speakers, it should be up to them to potentially contact other speakers of the language. Certainly not something that can just be mass-messaged by anyone who sees that a test-project exists. Also, there are many reasons why people are not already contributing. -- MF-W 19:40, 11 May 2026 (UTC)

Certainly not something that can just be mass-messaged by anyone who sees that a test-project exists., why on earth not? If it gets more input into the test-wikis and doesn’t cause disruption, I fail to see how that isn’t desirable. Do people not want there to be more projects? Is RNL supposed to be a quagmire that’s incredibly difficult for volunteers to surpass, wherein they get next to no support? Kowal2701 (talk) 20:14, 11 May 2026 (UTC)

It's not really friendly to spam people about a project that you know nothing about. It would be more polite to coordinate the messaging with any active participants of a test-project. In addition to potentially avoiding social problems (like contacting someone who opposes the test-project), this would also make it possible to get translations for the message into that language. WhatamIdoing (talk) 03:09, 12 May 2026 (UTC)

I'd personally oppose operating such a bot just for promoting RFL supporters, as it will only cause unnecessary harrasements. But if and when there are important subdomain changes (create, close, and/or delete), I certainly support bot notifying. ~2026-23305-80 (talk) 03:03, 2 May 2026 (UTC)

Just to clarify, this isn't to get more input into the actual requests, but to get more users actively editing in the incubator so that the projects can be successfully launched as new wikis. Sorry, I should've linked to Incubator in my first sentence instead of RFL Kowal2701 (talk) 18:21, 2 May 2026 (UTC)

Just post a message to the Village pumps of the language in question. Posting a message to all of the active members of those projects/websites is almost never going to be justifiable spam. It will also definitely get you into deep trouble with Europeans, myself included. The message being about Incubator instead of RFL changes absolutely nothing. Snævar (talk) 18:40, 10 May 2026 (UTC)

I'm not suggesting that we message all members of any project, just members who have a language category on their userpage that corresponds to a current proposal at RFL. Eg. if there was a proposal for a Northern Sami Wikipedia, I'd notify active editors in no:Kategori:Bruker se (which is about 10 editors), as well as those in that category on other Wikipedias Kowal2701 (talk) 18:51, 10 May 2026 (UTC)

That's 10 editors, but if you leave out the ones who are beginners and the ones who haven't edited for a long time, it's only five. There are individuals who grumble about any messages being "spam", but they are a minority. WhatamIdoing (talk) 20:29, 10 May 2026 (UTC)

I've tried to use those categories before, and I found that they were mostly useless due to the high proportion of inactive editors listed there. For example, in w:en:Category:User hsn, five editors are listed and only one (1) has made an edit there in the last two years. Even zhwiki only has a few dozen in w:zh:Category:Hsn 使用者. That said, if you wanted to notify active users, then you can do that already. Just copy/paste the usernames from the cats into a list (leaving out the inactive ones), and leave a note on their User_talk: pages. If the list is unusually long, then you can use MassMessage. WhatamIdoing (talk) 02:46, 6 May 2026 (UTC)

@WhatamIdoing, yeah but it's messaging the editors on other language wikis (like frwiki) as well, and doing this for every proposal would take ages. I think I could request a script that collected names of active/semi-active editors who had a specific language category? Kowal2701 (talk) 19:51, 6 May 2026 (UTC)

Yes, a script could create such a list.

From Meta-Wiki, a single list of names can send messages to editors at any other wiki. See, e.g., MassMessage/Lists/Active users whose name starts with asterisk and Global message delivery/Targets/Tech ambassadors for the two different list-formatting styles. WhatamIdoing (talk) 01:19, 7 May 2026 (UTC)

And, of course, one the script (or query? You might ask at w:en:WP:RAQ) exists for one language, it's easy enough to change it to make a list for another language. WhatamIdoing (talk) 01:20, 7 May 2026 (UTC)

requesting global banner to advertise 2vrayn

99 million have no connections ... Due to internet blackout in iran farsi wikipedia has died i request a global banner to advertise v2rayng domain fronting free MIMT software developed by some user so we can all bypass internet censorshhip https://github.com/patterniha/MITM-DomainFronting Baratiiman (talk) 18:45, 8 May 2026 (UTC)

It's pretty unlikely that this would happen, but why would we advertise this one solution? ―Justin (koavf)❤T☮C☺M☯ 21:17, 8 May 2026 (UTC)

Better detecting bots and replacing our CAPTCHA

Hello from the Product Safety and Integrity team! After a successful trial of hCaptcha for bot detection on English, French, Japanese, and four other Wikipedias, we will be rolling out hCaptcha to all projects over the course of the next few weeks.

Specifically, this will include using hCaptcha for account creation, and for edits by newer users. This replaces the use of the traditional "type in the word" CAPTCHA that is currently required of many users for these flows. hCaptcha is set up to only challenge likely-suspicious activity, so very few humans will be interrupted at all. In our trial, we estimate that only about ~2% of human users were actually challenged.

For background - as part of our focus on securing the wikis and detecting bad-faith activity, we are building stronger protections against bots carrying out activities that are generally intended for humans, such as creating accounts and editing. As the web has evolved and computers have gotten smarter, our old CAPTCHA has become both challenging for humans and easy for computers.

We recognized this and began with a trial of hCaptcha, third-party bot detection service, to replace our old CAPTCHA. Since then, eight Wikipedias, including English, French, and Japanese, have been using hCaptcha to protect account creation and certain kinds of edits. hCaptcha is a company that specializes in bot detection, with experience protecting very large online websites while prioritizing user privacy in its design. We have implemented technical safeguards to reduce the sensitivity of the information we send to hCaptcha.

It also gives us "likely bot" signals for account creation and covered edits, regardless of whether the session was challenged. Those signals inform what we show CheckUsers and stewards, who use them to find and remove bad-faith activity that was likely done by bots and which may not have been found any other way.

You can read more about this project at our recently updated project page. We welcome comments here or at that project's talk page. Subscribe to the Wikimedia Foundation Bulletin to follow the most important news from the Foundation. Thanks! SGrabarczuk (WMF) (talk) 13:24, 12 May 2026 (UTC)

Wikinews/Wikiversity?

Should Wikinews merge with Wikiversity? BigKrow (talk) 22:18, 13 May 2026 (UTC)