deadline is 2025-09-28 23:59 UTC (currently 2025-09-18 02:33:18)

Calendar: current deadline is highlighted, and current UTC date is 2025-09-18 02:33:18.
August 2025
Mon Tue Wed Thu Fri Sat Sun
28 29 30 31 01 02 03
04 05 06 07 08 09 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31
September 2025
Mon Tue Wed Thu Fri Sat Sun
01 02 03 04 05 06 07
08 09 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30 01 02 03 04 05
October 2025
Mon Tue Wed Thu Fri Sat Sun
29 30 01 02 03 04 05
06 07 08 09 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30 31 01 02
The Signpost currently has 5745 articles, 717 issues, and 14090 pages (4608 talk and 9482 non-talk).
Current issue: Volume 21, Issue 11 (2025-09-09) · Purge
issue page · archive page · single-page edition · single-page talk (create)
Previous issue: 2025-08-09 · issue page · archive page · single-page edition · single-page talk


21:11 In the media

[edit]

While updating issue 21:10 with "AI slop" critiques, I stumbled across some AI slop: "Wikipedia Page Creation Expert: How Sangode Oloruntoba Samson (Samson Expert) [new line, smaller font] Helping Influencers and Startups Build Digital Authority". I hope this isn't what we have to look forward to. ☆ Bri (talk) 22:12, 7 August 2025 (UTC)[reply]

@Bri: I've added that story to ITM, leaving out the links (so as not to give the guy an advert). Yes, it does look like a horrible way to write an encyclopedia, but unfortuneately will will have to deal with this. Smallbones(smalltalk) 11:44, 15 August 2025 (UTC)[reply]
See Wikipedia:Wikipedia_Signpost/Next_issue/In_the_media#All_hail_Grnrchst_and_The_Signpost!
@Grnrchst: Congrats on the Ars Technica write up of your article [1] from the last issue! I've started a top section write up in ITM. It's always nice when the mainstream media picks up a Signpost story! My take on this is that it is very complimentary to Grnrchst and very well-written (by Nate Anderson, the Deputy Editor, who is not the Nathan Anderson I had a story about). He had the easy job, you had to do and explain all the research. A 2 line (or whatever) quote from you would be nice for ITM. Smallbones(smalltalk) 11:44, 15 August 2025 (UTC)[reply]
@Smallbones: Thanks for the ping. What would you like the quote to be about? The Ars Technica article, the story itself or something else? --Grnrchst (talk) 11:57, 15 August 2025 (UTC)[reply]
@Grnrchst:. About the Ars Technica article, or "the thrill of victory and the agony of defeat", or "Who is that guy?" or What your mother thought of the article? Or thank Bri for his copy editing. Think of this as the closest you'll get to making an Oscar acceptance speech! Don't worry - we can edit it! Smallbones(smalltalk) 18:08, 15 August 2025 (UTC)[reply]
I'm not sure I could fit "the thrill of victory and the agony of defeat" into a wee quote, that'd be more of an essay for me. I'll give it a think over. --Grnrchst (talk) 22:04, 15 August 2025 (UTC)[reply]

Congressional investigation looks like its done, other than inserting links to previous Signpost coverage, new articles as they come out, and basic clean up. There have been 2 edit conflicts which are hard to resolve because they are so similar. So I'll leave off editing and writing more in ITM until tomorrow. There are other or related stories I'm still working on. Smallbones(smalltalk) 14:48, 29 August 2025 (UTC)[reply]

21:11 publication date

[edit]

Seeing that the deadline template hadn't yet been updated, I have just tentatively set the next issue's publication date to August 24 (fourth Sunday of this month). @JPxG and Bri: let us know in case you expect to be unavailable that weekend and prefer a different time.

There will be a RR in this issue, and I should also be able to help with other parts (sadly I didn't have enough time in recent weeks to contribute to the last issue, except doing the socials).

PS: Ceterum censeo that updating the template with the default date for the upcoming issue should be part of the script (e.g., per the preference we had arrived at here for the 1st and 3rd Sunday of the month, setting it to whichever of these dates comes next, unless that would leave less than, say, two weeks between the issues, in which case it would become the Sunday after that). JPxG, please mention in case you were still going to work on something like that - otherwise I may throw together some test code for this soon.

Regards, HaeB (talk) 06:17, 15 August 2025 (UTC)[reply]

I am set to be some multiple dozens of miles from a usable Internet connection or computing device on the 24th, so I would recommend that either somebody else publish on that date or that it be postponed until the following week (e.g. the first or second). jp×g🗯️ 05:47, 21 August 2025 (UTC)[reply]
Thanks - I have moved it accordingly by slightly over a week to Sept 1 00:00 UTC (also considering that we don't have a lot of material yet, even for the sections that tend to materialize earlier). This way, the crunch time should still fall on a Sunday for most in the team. Regards, HaeB (talk) 06:12, 21 August 2025 (UTC)[reply]
Presently, I have no internet access (and am posting from my friend's shit, thanks Cody) and do not expect to have any until the 3rd. jp×g🗯️ 03:36, 30 August 2025 (UTC)[reply]
Thanks for the heads-up. I see the deadline hasn't been updated yet accordingly, so I'm going to do that myself shortly unless Bri indicates that he is planning to take over publication by the current deadline tomorrow. Regards, HaeB (talk) 01:35, 31 August 2025 (UTC)[reply]
I do have internet access now, but there are also seven pending hours of driving and unknown pending hours of sleeping after that... I look forward to either reading the new issue or helping put it out, whichever I wake up to :^) jp×g🗯️ 17:16, 2 September 2025 (UTC)[reply]

I could publish today, though it looks a little thin as I look at the prepared material. @Adam Cuerden, Bluerasberry, Headbomb, Igordebraga, Smallbones, and Soni: Do you want to move ahead with what we have at this point or wait? ☆ Bri (talk) 14:31, 31 August 2025 (UTC)[reply]

No opinion really. If you want some extra content, it should be pretty easy to find an essay to publish via the signpost. I've suggested WP:1Q in the past, but it's a pretty light essay and I wrote it so I'll defer on the decision. Headbomb {t · c · p · b} 14:36, 31 August 2025 (UTC)[reply]
I'm still working on a Disinfo report. I'm "comfortable" with publishing on the 3rd, though I should be ready tonight. Smallbones(smalltalk) 14:42, 31 August 2025 (UTC)[reply]
In that case, not publishing today. Headbomb, I took up your suggestion and created Wikipedia:Wikipedia Signpost/Next issue/Essay. ☆ Bri (talk) 14:48, 31 August 2025 (UTC)[reply]
Think the Traffic Report is ready, specially as I would update the most edited part if the Signpost wasn't done yesterday, but the source didn't do so. What's left to get published? igordebraga 14:28, 2 September 2025 (UTC)[reply]
I had actually already gone ahead and changed the template to September 3 (as a haphazard, optimistic guess based on the until the 3rd above). But that clearly didn't happen.
So yes, time to consider invoking the usual EICAWOL protocol: Bri, would you still be available to carry out publication tomorrow, assuming other team members help out with tying up the remaining loose ends at Wikipedia:Wikipedia_Signpost/Newsroom#Article status (in particular marking sections as ready for copyedit or as copyedited, and approving individual sections for publication, standing in for the EiC)? I just tentatively updated the deadline again to 2025-09-08 00:00 UTC. I myself will have RR in a publishable form by then, and should also be able to help out with getting several other sections ready, starting with the Technology Report. I already went ahead and approved the Essay (as discussed before, such in absentia EiC approvals should preferably be done for sections that one wasn't significantly involved with oneself).
Regards, HaeB (talk) 05:17, 7 September 2025 (UTC)[reply]
Oh well, since there had been no reaction here (and only three smaller edits on the issue itself), I have changed the template again. Again, we need 1) Signpost team members helping out with the still open tasks mentioned above, 2) Bri or JPxG to give note that they will be available to carry out publication, and when. Regards, HaeB (talk) 02:26, 8 September 2025 (UTC)[reply]
I can publish in the early evening my time which would be around 0100 9 September UTC. ☆ Bri (talk) 14:19, 8 September 2025 (UTC)[reply]
Just ran the publication script in dry-run mode. Something in the Technology report is upsetting the publication script. Can anybody see a problem?Bri (talk) 23:35, 8 September 2025 (UTC)[reply]
Disregard I found the error. Still on track for publishing in a bit more than one hour. ☆ Bri (talk) 23:40, 8 September 2025 (UTC)[reply]
Starting in a few minutes. Please refrain from editing until I come back.Bri (talk) 00:57, 9 September 2025 (UTC)[reply]

21:11 Recent research

[edit]

As usual, we are preparing this regular survey on recent academic research about Wikipedia, doubling as the Wikimedia Research Newsletter (now in its fifteenth volume). Help is welcome to review or summarize the many interesting items listed here, as are suggestions of other new research papers that haven't been covered yet. Regards, HaeB (talk) 06:10, 20 August 2025 (UTC)[reply]

BTW, the doi:10.9734/bpi/mplle/v4/9161D paper is from ScienceDomain International (or rather one of its imprints 'Book Publisher International'), one of the worst predatory publishers out there. Headbomb {t · c · p · b} 19:17, 29 August 2025 (UTC)[reply]
Thanks for noting that. I'll leave it to the author of the review to decide whether and how to mention it.
Generally speaking, as regular readers of this section will be aware, we try to be relatively comprehensive in coverage and not impose a high quality threshold (the idea being that, somewhat related to the debate about ITM above, the sections's purpose is not or not only to provide reading recommendations and highlight the best quality research, but also to give readers an overview on how Wikipedia has been covered in academic research overall recently, and ideally also caution them in cases of bad research. What's more, quality-wise the problem with predatory publishers is the floor, not the ceiling. In other words, while a paper they publish doesn't come with the minimum quality guarantees (or say aspirations) of more reputable publishers (and hence usually fails WP:RS too), that doesn't automatically mean that every such paper is bad - it just needs additional scrutiny, which happens to be precisely what we are offering with such a review in RR.
In this particular case, the author (sadly deceased recently btw, after this paper's publication) was a professor at a reputable (if not quite Ivy League) US university, and had published about related topics before elsewhere.
Apropos this publication: It seems that the ISBN that Citation bot added in Special:Diff/1308483866 doesn't exist (at least it is unknown to several of the search engines I tried out from [2]). Any idea what went wrong there?
Regards, HaeB (talk) 06:43, 30 August 2025 (UTC)[reply]
1) I'm not very partial to the argument that 'they're from a good university, so the paper must be good/not that bad'. Authors from Ivy League Universities have been found to publish in predatory journals for highly dubious reasons.
I'll agree that a paper published in a crap venue does not necessarily mean it's a crap paper (see WP:VANPRED#Use in the real world vs use on Wikipedia) and can be evaluated critically. It is definitely possible that this is a case of honest scholarship, by an honest scholar, that got fooled into publishing into a predatory venue for one reason or another [like being new to a field and not knowing where to publish such research]. Which rather sucks for them, because they got denied the peer-review they were seeking. But it is also possible that this is scholarship they couldn't publish elsewhere, and were happy to find a place to publish it, without asking too many questions about the venue. And at the worse end, it's possible this is scholarship so flawed they looked for a predatory publisher specifically to circuvent the peer review process and try to get a paper published just to meet 'must publish x papers per year' standards.
I am not familiar with Wiggers and have no opinion on which of the above (if any) is the situation here. Likewise I don't blame @Katarzyna Makowska (WMPL): for not knowing about it. To Katarzyna: This doesn't make your review invalid, I'm just pointing this out. No idea if this affects your opinion of the piece, but I think it's at least worth pointing out.
2) ISBN - predatory publishers rarely fill all the paperwork required to ensure ISBN and DOIs are fully functional and registered. It is mentionned here however, and if you google it, you'll be taken to these results, so IMO it's worth including. Headbomb {t · c · p · b} 08:25, 31 August 2025 (UTC)[reply]
Thank you Headbomb for this information, and thank you HaeB for the extra context. I was not aware that this outlet is a predatory publisher, I will keep in mind to check in the future. I have added a mention of this to my review, also drawing from HaeB's arguments about the authors other work. I have not mentioned that he sadly passed away recently, which I also didn't know. If we think it's appropriate I can add a sentence on that as well, but not sure if it's the right place. It's my first time writing reviews for this newsletter, happy to learn. Katarzyna Makowska (WMPL) (talk) 08:05, 31 August 2025 (UTC)[reply]

21:11 Traffic report

[edit]

The table of contents illustration ("piccy") is AI generated but has a cc-by license attached. AFAIK the AI generated stuff is all public domain in the United States, there has to be a human author to receive a copyright and to grant license rights. Don't know if this is worth doing anything about but since we display the license on the TOC, we are kind of perpetuating a bad license in my opinion. ☆ Bri (talk) 18:25, 25 August 2025 (UTC)[reply]

For reference: File:Bot pic IA.png. The uploader already double-licensed it as PD-algorithm, so there should be no issue with changing the piccy attribution accordingly.
The bigger issue IMHO is why this should be considered an adequate illustration at all for the entry ChatGPT in the traffic report. Why is the robot wearing a MediaWiki t-shirt? CCing User:Igordebraga who had added it to Wikipedia:Top 25 Report/July 20 to 26, 2025. Regards, HaeB (talk) 00:41, 26 August 2025 (UTC)[reply]
Why would it be inadequate? Image was on the ChatGPT category on Commons, it was presumably made through it. A robot wearing a MediaWiki shirt seemed a good way to illustrate an entry about the chatbot's Wikipedia page (if only to have fun with it). igordebraga 01:34, 26 August 2025 (UTC)[reply]
@HaeB: can you help me understand how multi-licensing works on Commons? Is it supposed to imply that re-use or distribution can occur under the distributor's preferred license, presumably the less restrictive? ☆ Bri (talk) 05:48, 26 August 2025 (UTC)[reply]
Pretty much, yes, see also c:Commons:Multi-licensing.
(Btw, this is a straightforward consequence of the basic legal meaning of license as a permission to do something. Unfortunately there seems to be a widespread misunderstanding where a lot of people (Wikimedians included sometimes - not saying present company, but e.g. in the 2023 WMF ToU update discussion this was quite present) seem to think that e.g. a CC BY-SA license introduces restrictions - "now that I have put CC BY-SA 4.0 on my content, the people who reuse it will be required to attribute it and share derivative works under the same license" or such; overlooking that without a license, reuse was not allowed in the first place.)
Regards, HaeB (talk) 06:19, 26 August 2025 (UTC)[reply]
Hmm, I am not a lawyer but wouldn't cc-0 be more appropriate for a multi-license scenario? PD means "I never had any rights to this in the first place", which seems at odd with the Commons licenses which say "here are the rights to my work which I am granting to you". In other words: I still don’t see how a PD claim, label, or template is compatible with any licensing. In other, other words: it seems that any license is illogical once something is determined to be in the public domain. However this may be a case where the variance in copyright for AI gen images for the U.S. and other countries comes into play. ☆ Bri (talk) 14:15, 26 August 2025 (UTC)[reply]
CC-0 isn't "I never had any rights to this" but rather "I release this without restrictions on reuse, attribution, commercialization etc...". It is effectively releasing your work directly into public domain. Headbomb {t · c · p · b} 22:10, 26 August 2025 (UTC)[reply]

21:11 News and notes

[edit]

I've started a rudimentary article on the WMF's loss (though they view it as a partial win) in the Online Safety Act case. Sources:

I have a lot on my plate at the moment ... any help welcome. --Andreas JN466 11:11, 27 August 2025 (UTC)[reply]

Have you seen this? The Hill Liz Read! Talk! 23:48, 27 August 2025 (UTC)[reply]

Just saw it now over on de:WP (someone mentioned this German report on the matter over there) and added a link to ITM. But one could argue that NaN is a better place to cover this. What say ye? Pinging JPxG, Smallbones, HaeB, Bri. At any rate, it's very much a developing story; we're in the very early stages. No idea what it may evolve into. Andreas JN466 22:44, 28 August 2025 (UTC)[reply]
@Liz Note the the letter is available as well. The probe appears basically related to the Heritage Foundation story we covered a few weeks ago. There is a Village Pump discussion:
Andreas JN466 23:06, 28 August 2025 (UTC)[reply]

I'll take 20 minute break for dinner, talk to the family, then start writing in ITM. Smallbones(smalltalk) 00:17, 29 August 2025 (UTC)[reply]

Thanks. There is something about the UK in N&N. I might have a little time tomorrow to brush it up—right now it's crap. For the MAGA story go by the Republicans' letter (link in ITM). The Hill glosses over all the important bits. Best wishes - Andreas JN466 00:31, 29 August 2025 (UTC)[reply]
Maybe, we could post the WMF/Diff statement on the UK as news from the WMF? Smallbones(smalltalk) 00:42, 29 August 2025 (UTC)[reply]
Thanks, you guys are awesome! But take those family vacations, too. Liz Read! Talk! 03:06, 29 August 2025 (UTC)[reply]
TechDirt has some analysis:
https://www.techdirt.com/2025/08/13/uk-court-ruling-leaves-wikipedia-facing-years-of-uncertainty-under-online-safety-act/
There was actually quite a lot of press on August 11:
https://headless.courthousenews.com/wikipedia-loses-bid-to-shield-anonymous-contributors-from-uk-online-safety-law/
https://www.forbes.com/sites/emmawoollacott/2025/08/11/wikipedia-may-have-to-impose-identity-verification-on-readers/ Andreas JN466 08:14, 29 August 2025 (UTC)[reply]
Diff post for reference:
https://diff.wikimedia.org/2025/08/11/wikimedia-foundation-challenges-uk-online-safety-act-regulations/
Longer, more detailed version on Medium:
https://medium.com/wikimedia-policy/wikipedias-nonprofit-host-brings-legal-challenge-to-new-online-safety-act-osa-regulations-0f9153102f29
Both with the Aug. 11 update. Andreas JN466 08:25, 29 August 2025 (UTC)[reply]

I've done some work on the Online Safety Act piece in N&N. It's now in better shape. --Andreas JN466 11:28, 29 August 2025 (UTC)[reply]

Thanks for this, I have just augmented it with more context and relevant detail (including link to previous Signpost coverage, although one could still go further and mention the preceding year-long efforts by Wikimedia UK and the Foundation in persuading UK lawmakers to avoid this situation, which appear to have been largely futile, unlike some analogous efforts in other countries and the EU).
I feel a bit ambiguous about ending the article with pointers to previous Signpost articles about editor imprisonment. While I understand why you put it there (the connection being that WMF argues, with justification, that the OSA "could expose users to [...] even imprisonment by authoritarian regimes"), many readers might miss that nuance and take away the incorrect impression that the Online Safety Act directly threatens editors with prison in the UK, or also confuse this with the recent controversies and free speech concerns about widespread arrests in the UK for social media posts (which I understand to be pertaining to different laws). Regards, HaeB (talk) 05:37, 3 September 2025 (UTC)[reply]
Thanks, HaeB, nice work; that looks much better now. I see your point about the final sentence. I have added "by authoritarian regimes around the world" after "imprisonment". This should help somewhat. Best, Andreas JN466 23:04, 3 September 2025 (UTC)[reply]

21:11 Disinformation report

[edit]

ETA about 3 hours, in order to give it some polish. Thanks @Bri: for copy editing, but I'll also change a few sentences along the way. @Jayen466: I'm looking for some Signpost articles on "dark arts" and Bell Pottinger. I'da thunk you wrote the one I'm looking for. There is a good Bell Pottinger Wiki-article. Smallbones(smalltalk) 22:01, 1 September 2025 (UTC)[reply]

Ouf, I can't recall writing anything substantial about Bell Pottinger, Smallbones. Some past references to that story are: Wikipedia:Wikipedia Signpost/2011-12-12/In the news, Wikipedia:Wikipedia_Signpost/2012-07-02/In_the_news, Wikipedia:Wikipedia_Signpost/2016-03-16/In_the_media. Best, Andreas JN466 22:22, 1 September 2025 (UTC)[reply]
Thanks, Smallbones(smalltalk) 00:36, 2 September 2025 (UTC)[reply]
See also [3].
Generally speaking I can highly recommend bookmarking the Signpost's archive search form (or even better, add a search shortcut to your browser [4][5]). Regards, HaeB (talk) 05:44, 3 September 2025 (UTC)[reply]
Yes thank you for reminding me. It took a few more steps than usual but I found something pretty good. Wikipedia:Conflict of interest/Noticeboard/Bell Pottinger COI Investigations I was thinking about @Jayen466:'s Kazakhstan related articles, this one was related to Uzbekistan. Smallbones(smalltalk) 22:42, 3 September 2025 (UTC)[reply]
Ah, the Kazakhstan article. I remember I was in Washington for the 2015 WikiConference North America when that issue was published ... good times. The world and America have changed a lot since then. :/ Andreas JN466 23:17, 3 September 2025 (UTC)[reply]

Smallbones – the lead paragraph promises to go over these topics: how the Wikipedia community is organized, the limited role of the WMF, and how the English-language Wikipedia deals with disinformation. It's not immediately clear where to find those topics in the rest of the piece. Maybe different sub-headings would be helpful? And I'm not sure how DEI, which gets a paragraph or so, fits in with those subjects just mentioned. ☆ Bri (talk) 03:58, 4 September 2025 (UTC)[reply]

@Bri and JPxG: I've had some trouble logging in today. Have the servers been down? In any case I can log in now and the basic gist of Bri's requests are in there. There's some grammatical stuff, a few other things to check, But it could be copy edited now and be ready to go. It is too long.
BTW Verge has a story that should go in ITM. at [6] and on some archive. Long-as-hell and twice as boring! Nah. Smallbones(smalltalk) 21:47, 4 September 2025 (UTC)[reply]
That Verge report is really good. Hawkeye7 (discuss) 04:05, 12 September 2025 (UTC)[reply]
I spoke too quickly. I still can't edit the Disinformation Report page. Here's what I'm trying to change. The word "start" is duplicated in one sentence. Is Trust & Safety still called Trust & Safety. In the bottom half several times the word "you" should be changed to "Congress" or similar. As far as the editing problem, I hope somebody else can edit it and I didn't break the whole page! Go ahead please. Smallbones(smalltalk) — Preceding undated comment added 21:58, 4 September 2025 (UTC)[reply]
No trouble with system access for me, but I remain logged in more or less continuously because 2FA is a pain to reset. This diff covers your concerns, I think. ☆ Bri (talk) 17:48, 5 September 2025 (UTC)[reply]

Disagreement about a use of Template:Press

[edit]

Talk:CNN#Template:Press_BRD If you have an opinion, please join. Gråbergs Gråa Sång (talk) 18:22, 5 September 2025 (UTC)[reply]

21:11 publication

[edit]

Publishing in a few minutes. Please refrain from editing until I come back.Bri (talk) 00:58, 9 September 2025 (UTC)[reply]

It is done, no drama; just doing some spot checks. ☆ Bri (talk) 01:04, 9 September 2025 (UTC)[reply]
Still looks good, spot check of local delivery is Thumbs up icon. Feel free to copyedit typos or whatever again. ☆ Bri (talk) 01:07, 9 September 2025 (UTC)[reply]
Thanks for saving the day!
I removed some leftover placeholders from N&N. Come to think of it: We may want to have the publication script throw a warning if any of the sections lined up for publication contains the string "TKTK". Regards, HaeB (talk) 01:45, 9 September 2025 (UTC)[reply]
Good idea for more sanity checking. I'd love to have it throw those warnings during the pre-publication "dry run". Also: here is a reader feedback link for the issue. ☆ Bri (talk) 03:01, 9 September 2025 (UTC)[reply]

WikiProject report submission

[edit]

Wikipedia:Wikipedia Signpost/Next next issue/WikiProject report seems to be in limbo. Should it be returned to userspace draft, or ...? ☆ Bri (talk) 20:05, 9 September 2025 (UTC)[reply]

See previous discussion at Wikipedia_talk:Wikipedia_Signpost/Newsroom/Archive_45#21:8_WikiProject_report: Unless User:Valorrr still plans to work on it to make it a story in the usual format for this section (in particular, conducting interviews with the WikiProject's participants), this indeed seems like a case for userfying. Regards, HaeB (talk) 00:01, 10 September 2025 (UTC)[reply]

















Wikipedia talk:Wikipedia Signpost/Newsroom