Jump to content

Wikipedia talk:Wikipedia Signpost/Newsroom

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
< Wikipedia talk:Wikipedia Signpost

This is an old revision of this page, as edited by SGrabarczuk (WMF) (talk | contribs) at 17:47, 7 December 2023 (→Technology report: new section). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

The Signpost
The Signpost
(talk  · chat)
Recent changes:
main · talk


Publication date

I set the publication date ahead a few hours, to 01:00 UTC on the 16th... currently it's about two minutes past, and it looks like some articles need substantial work (ITM, N&N). I erstwhile discovered that there seems to be some weird thing going on with the deadline template where it displays different dates (?). Anyway, I will try to drag some of my drafts out of the closet, because it looks like a pretty thin issue. I'm going to take a walk and ponder it for a bit. jp×g 01:03, 16 October 2023 (UTC)[reply]

I don't think we should publish this week. Maybe not next. We need to consider how we can get the newspaper working on a regular schedule again. IMHO it should start with the EiC showing up and helping out well before the deadline, rather than an hour or so after the deadline. I can't speak for Bri or Andreas, but I suspect they are each thinking something like "This can't continue like this where I'm doing 60% of all the work." Of course there are others who have been contributing, but A&B are really carrying too much of the weight to be healthy. If we committed to addressing the problems with an issue ahead of time - before the deadline - then things would work a bit more smoothly. There really isn't an issue to patch up here, so trotting out a few old drafts won't help. Just my opinion of course. Smallbones(smalltalk) 01:28, 16 October 2023 (UTC)[reply]
Okay, I am back. Yeah, uh -- I have not really been having a great time with the bimonthly schedule, and I've noticed things getting a bit more lethargic. Perhaps these things are related. It's quite a bit more difficult to stay on top of. On one hand, it does allow us to cover things in a more timely fashion. But on the other hand, a lot of our recent issues have been very thin, and I am starting to miss the days when I was writing more stuff (i.e. the last deletion report was Aug '22 and the last time I wrote an arbitration report was very long ago indeed). Of course, it's a little embarrassing to give up on a project, especially one that had such vigor behind it (we were putting out dynamite on the regular when we started the bimonthly distribution!) but I think it might not have flown so good. Who knows. What do you all think? jp×g 03:06, 16 October 2023 (UTC)[reply]

I suppose the logical next step is for me to try to finish up what we've got, and see how close that gets us to an issue. jp×g 09:29, 17 October 2023 (UTC)[reply]

I added one story to N&N and two to ITM. Both sections still need some editing (titles, images etc.) but I think there is enough content now to put the issue out.
Regarding the general issues brought up above:
We need to consider how we can get the newspaper working on a regular schedule again. IMHO it should start with the EiC showing up and helping out well before the deadline, rather than an hour or so after the deadline. - Agreed
I can't speak for Bri or Andreas, but I suspect they are each thinking something like "This can't continue like this where I'm doing 60% of all the work." - I assume you are referring to their recent work on N&N and ITM. I am not sure what this has to do with percentages or workloads. As a reminder, these sections don't currently have a point person listed at Wikipedia:Wikipedia_Signpost/About#The_Signpost_team (unlike, say, the Arbitration report, Featured content or Recent research). But in the Signpost's history of almost two decades both of these two sections have frequently been shepherded by a single editor each. I mean, given their usually recurring contributions, it might indeed be useful if Bri and Andreas gave a heads-up that they won't be able to contribute substantially to an upcoming issue, so that others are better incentivized to jump in in time. But as long as they haven't formally signed up as the point person for either section, they shouldn't feel obliged to do so.
As for the bi-monthly schedule, all the arguments for it (and against a monthly schedule) from the previous discussion still hold. And looking at this year's archive, we have been publishing an average of two issues per month a good while now already (2 per month from January to July 2023, 3 in August, 1 in September, 1 in October so far), including some very voluminous ones. So I'm kind of unconvinced that publishing twice per month is the cause for things getting a bit more lethargic.
Regards, HaeB (talk) 08:14, 19 October 2023 (UTC)[reply]
@JPxG, HaeB, Bri, and Jayen466: I don't see any reason that we couldn't put out an issue this Sunday, but I would certainly like to see the following:
1. A promise that the EiC will be available and involved on Saturday or Sunday. Or at least he gets somebody else involved (other than me) to do the work for him and approve the articles that he can't check himself.
2. That he's ready to publish at approximately the currently posted deadline. Or has found somebody ready and able to do it for him. A guide to the publication process would likely help, since there have been many changes to this process, as I understand it.
I'm sorry if this seems demanding, but I just don't understand how the process of editing and publication can work otherwise.
Smallbones(smalltalk) 16:01, 20 October 2023 (UTC)[reply]
There have been a few pings directed at me. I'll answer that yes I'm still part of The Signpost but am eager to see other people pick up more of the workload. Other stuff in my life has required a diminution of my contributions to Wikipedia generally, and I don't think it would be fair to have people waiting on me to do any regular article beat, including being considered the primary shepherd or whatever for a column. ☆ Bri (talk) 20:52, 20 October 2023 (UTC)[reply]
One thing I can say is that with all the stress of dealing with my dad's death, I've not been able to do what I usually did for the first few months of bimonthly: Go through everything the day before publication, and do what was needed to get it publishable. I was getting credit for up to 5 articles an issue - Featured Content, From the Archives, and often writing enough in order to finish up N&N and ITM to get a credit on them, and often doing a Gallery or something if things were thin (and then usually waiting a week for publication after giving an entire day to getting that all done). Now, I can just about get FC out. Adam Cuerden (talk)Has about 8.5% of all FPs. 21:07, 20 October 2023 (UTC)[reply]
My impression is that writing deadline doesn't mean a whole lot. Whenever I check on things at the writing deadline, it's typically the case (as it was here) that hardly anything's been written, or even started, so there's not a lot for me to do in terms of editing. I don't have a huge issue with this: I would spend the same amount of time editing and preparing an issue regardless of what day the deadline was. So if everyone else is fine with ignoring the writing deadline, then I am fine with it too. But it does require some time, after everything's in, to edit and publish -- which means that if this is the deal, the issue will always be somewhat late in coming out.
I have modified the template that shows up at the top of this page (and at the top of drafts, etc) so that it shows the remaining time for the writing deadline -- maybe this will help emphasize it -- I don't really know. I can promise to be available for publication -- and I was on Sunday -- but publication wasn't the task that needed to be done, the task that needed to be done was writing the articles and then editing them and then publishing them. And I am fine with being available for this, but there was nothing on the newsroom talk about it either, I figured it was just going to be the normal situation where everything's lorem ipsum until 30 minutes before the publication deadline.jp×g 00:26, 21 October 2023 (UTC)[reply]
@JPxG: You are missing the whole idea of a deadline. On the publishing deadline there's really only 3 things you can do: 1) publish as scheduled with what you have (though quite often there's 30 minutes or an hour of work that still needs to be done - in that case it's easy enough on Wikipedia to just take the extra hour to finish up - then publish asap) 2) if there is a breaking story, or a real mixup on a scheduled story so that it needs a complete rewriting, you just write or rewrite the story which might take an hour or two. I think this happened about 3 times while I was EiC. Or 3) you cancel the issue if there is not enough material available. Then you come back ready to go for the next regularly scheduled issue. I think this just happened once while I was EiC.
Then I would turn it over to the publisher, since I didn't have the technical means to publish. Normally it would take 30-60 minutes to publish, but sometimes the publisher wasn't immediately available and in one case IIRC it wasn't published until the next morning.
You notice there is a major difference in the size of the delays then vs. now - a few hours or half a day vs multiple hours or days. The reason for the difference should be obvious - I was going to get the issue to the publisher more or less on time if at all possible come hell or high water. To you it's not a big deal.
Why should it be a big deal? First the readers often like to read the paper at particular times or just get into the habit of looking for the paper at a certain time. Having it available on a regularly scheduled time is - to me - a matter of courtesy to the reader and also leads to increased readership. Perhaps more importantly it lets the reporters know that you take the deadline seriously, so that they need to take it seriously, which allows everybody to coordinate their efforts easily.
You should be coordinating with the reporters more, rather than trying to rewrite or polish what they have written after the deadline, after they've gone to sleep or to deal with other parts of their lives. And you really shouldn't write new stories after the deadline (I know it's difficult at times). That results in articles that have only been read and copy edited by you, which leads to mistakes. We do have a rule here that every story should be gone over by a second person before being published. Also, after deadline, you shouldn't be rewriting chunks of text in other people's stories - they may not agree with your changes. Sure, you're the EiC, but that doesn't mean you get to put words in their mouths. When at all possible, they should have the option to reject your changes (which then leaves you with the option of rejecting their stories or publishing as they want it). There should be some give and take between the editor and writer. That means before the deadline when you have time to email or ping them.
I particularly dislike you changing the style of others' writing. Sure you want to liven-up some text, correct confusing sentences, etc., and some of that is perfectly acceptable (before the deadline). But everybody should be allowed to use their own style, to write with their own voice. Some styles, of course, are not acceptable, but we have a wide range of topics - e.g. traffic report, recent research, special reports, interviews, etc. - that benefit from different styles within the range of acceptable styles. Every writer deserves the chance to develop their own voice. When you change my style, quite often I feel like you're not being serious enough - I feel many of my topics are pretty serious. I don't need to sound like a 20 year old. Or you are making a joke, or just missed the point of what I wrote. Writers need editors, but ...
Well that's a lot to consider. The main takeaway is "please, please take the deadline seriously. It ruins everything if you don't." Smallbones(smalltalk) 02:43, 21 October 2023 (UTC)[reply]
Not to become a man who complains about machinery not being given with an instruction manual, but it has been fairly difficult to figure out what the overall practices and expectations are. Certainly, journalistic skill issue on my part is a factor, but all of this stuff you've said is stuff I would have been glad to hear a long time ago. I even asked a month and a half ago, having completed a large volume of technical tasks, what I ought to focus on next, and it sat with no comments all the way up to being archived by the bot. From my perspective, imagine trying to make decisions based on asking what areas deserve some more attention and getting no response, not even a "whoopty doo", for six weeks: at the very least, it didn't seem like an urgent mandate to do things differently.
But at any rate, now that I've heard all of this, I will do my best to incorporate it into practice. You are correct about the deadlines. For my own part, I have been rather slovenly about it, and things have been run way later than they should have. For this there is not really a good reason (except for the wildfire too close to my house and the laptop getting busted too far from my house, which were great reasons, but the rest were just slovenliness). I really do wish people would say stuff like this more often. jp×g 11:26, 21 October 2023 (UTC)[reply]
@JPxG: Thanks for this. I'm sorry for letting this all build up and then my frustrations came out all at once. They are not as bad as it sounded. One other point that I should emphasize is that all contributors get busy at times and can't contribute as much as they'd like. Similarly, there is going to be attrition of contributors at varying rates - so recruiting new contributors should be a constant job. You're the boss, so if you don't delegate this, you have to do it.
I'll be able to work today, but not tomorrow. I'll guess I can add 3 more sections to ITM and maybe a News from the WMF copied article. Thanks again. Smallbones(smalltalk) 16:39, 21 October 2023 (UTC)[reply]
I've at least polished up everything in ITM and NaN. Adam Cuerden (talk)Has about 8.5% of all FPs. 06:47, 22 October 2023 (UTC)[reply]
  • Good meowning everyone. I see that we are mostly ready, and even have about eight hours to dick around with the issue prior to publication. I will do a more intense copyedit later, but I will make some basic observations in section sections. jp×g 16:30, 22 October 2023 (UTC)[reply]

Thanks and recap

It's probably time to admit that I cautioned JPxG when he was considering taking this job, that it is often thankless, and gets the E-in-C a lot of attention. IMHO anybody who's willing to do it is taking taking a big hit for the team and deserves thanks. Now on to making JPxG as successful as possible. I think some of the guidance on hitting deadlines is well intentioned, though a bit raw, and it would be good of some other folks were available to volunteer to make the whole thing work. One or two people can't do it alone. ☆ Bri (talk) 18:03, 23 October 2023 (UTC)[reply]

Personally, I feel the core of the issue is the twice monthly schedule, rather than a monthly schedule. If twice a month is too much to handle, let's slow down and that'll give people more time, and more predictable/memorable deadlines. Headbomb {t · c · p · b} 01:18, 24 October 2023 (UTC)[reply]

November 19 issue

We are nearly 10 hours past the publication deadline, with several meaty stories basically ready to go out but the EiC's last Signpost-related edit dating almost five days back. Any updates? What is the hold-up this time? Regards, HaeB (talk) 05:56, 20 November 2023 (UTC)[reply]

Hacker News

--Andreas JN466 00:22, 7 November 2023 (UTC)[reply]

Not to get all politics on main, but (Redacted). jp×g🗯️ 06:07, 7 November 2023 (UTC)[reply]
Yeah, that lead section is terrible. Andreas JN466 08:07, 7 November 2023 (UTC)[reply]

Earlier this week, the "Fringe" paper (item 1 above) also attracted the attention of Rebecca Watson aka Skepchick: [1] [2] (Thanks to a recent issue of Wikipedia’s Signpost newsletter, I learned about a really interesting study on Wikipedia itself. [...] it’s interesting that while Reddit, Twitter, and Facebook continue to circle the drain on their inevitable descent to the sewer, Wikipedia has only seemed to get better.) I took the liberty to add a brief item to ITM to close the circle ;)

Regards, HaeB (talk) 03:53, 20 November 2023 (UTC)[reply]

Cool! Andreas JN466 09:00, 20 November 2023 (UTC)[reply]

Pageviews, 2023-11-06

Wikipedia talk:Wikipedia Signpost/Single/2023-11-06

Pageviews for 2023-11-06 (V)
Subpage Title 7-day 15-day 30-day 60-day 90-day 120-day 180-day
Wikidata Evaluating qualitative systemic bias in large article sets on Wikipedia 577 691 732 774 802 819 863
WikiCup report The WikiCup 2023 351 482 510 544 584 638 712
Traffic report Cricket jumpscare 476 633 680 721 758 772 832
Recent research How English Wikipedia drove out fringe editors over two decades 12788 13224 13417 13534 13592 13629 13725
Opinion An open letter to Elon Musk 1264 1623 1713 1781 1836 1862 1925
News from Wiki Ed Equity lists on Wikipedia 434 566 594 632 668 686 730
News and notes Board candidacy process posted, editors protest WMF privacy measure, sweet meetups 815 980 1017 1086 1128 1153 1213
In the media UK gov bigwig accused of ripping off WP articles for book, Wikipedians accused of being dicks by a rich man 1114 1378 1440 1520 1565 1608 1703
Featured content Like putting a golf course in a historic site. 465 623 678 723 753 765 807
Arbitration report Admin bewilderingly unmasks self as sockpuppet of other admin who was extremely banned in 2015 53006 54419 54996 55482 55738 55879 56196


Not sure why these stats are showing up for some articles and not others; some kind of weird 24-hour rollover thing. Unsurprisingly, the ones that Andreas posted to HN look like they have about a hundred bajillion times more readers than the others. Much to think about! jp×g🗯️ 06:09, 7 November 2023 (UTC)[reply]

Numbers aren't everything. What kinds of people does the Signpost want to reach? Ed [talk] [OMT] 06:12, 7 November 2023 (UTC)[reply]
The Hacker News community is a pretty good match for Wikipedia demographically. Similar interests. Andreas JN466 14:42, 7 November 2023 (UTC)[reply]
To be clear, Andreas, I wasn't trying to implicitly criticize you with that post! (Apologies if it came off that way.) It was just an observation about valuing pageviews/popularity vs. intentionally reaching an audience. Ed [talk] [OMT] 15:35, 7 November 2023 (UTC)[reply]
Thanks, Ed. Best, Andreas JN466 22:16, 8 November 2023 (UTC)[reply]
Ed raises a valid point. Then again numbers are also not nothing, and I do think we should continue to quantify the impact of our work here.
Apropos, Ed, since there are a few parallels (and differences, of course) and it's always nice to talk shop, perhaps you would like to share a bit on how you folks are handling this nowadays in the Wikimedia Foundation's Communications department: How does the department (or your team) define the audiences that you want to reach with your communications, and what metrics are you using to measure the impact and success of that work?
Regards, HaeB (talk) 08:35, 9 November 2023 (UTC)[reply]
Hell of a question, and worth thinking about in more detail. Personally, I am not totally certain. I think that, as far as English Wikipedia power users, we are probably at or close to saturation (lots of getting-started guides, help pages, welcome messages and the like mention the Signpost). I think we could call this our core audience, and the people who we care the most about informing (or at least not misinforming). Hence all the stuff about the WikiCup, the traffic reports, featured content, arbitration and deletion reports, et cetera; I think this is primarily interesting for en.wp editors and there's no problem with it lacking "broad appeal". On the other hand, I think there is a much wider base of people who would be interested in learning about Wikipedia, and Wikimedia projects, across the wider WWW: a website or a project or a movement needs to continually bring in new participants, and Wikipedia is famously opaque to outsiders. I think that a news outlet that provides accessible, interesting stuff that teaches you about how things work in a way that isn't obtuse or boring serves a role in getting people comfortable and excited and familiar with the "whole deal of it".
Even if it doesn't, and they stay readers forever, I'd like to think that reading intelligent and accurate stuff about stuff that happens on Wikipedia would help people to be more intelligent and informed consumers of news and of media and of our own articles. For example, how often do you see some noob on a talk page saying that they're going to "stop donating to Wikipedia" if we don't XYZ? Or hear someone talk about Wikipedia needing to "hire more editors" or whatever? I think the wider public just lacks a bunch of key information about how we work, and part of that is that there isn't a lot of news coverage that explains it in a way that is easy or fun to read. Maybe the most poignant examples are the regular kerfluffues where someone (whether it's a rando online or a famous pundit) accuses us of being biased, and refers to some random editor who said a dumb politics thing as a "mod" who's sopping up your hard-earned donation money to et cetera. Or there's some big circus about a "censorship campaign"... referring to one person making a goofy AfD which doesn't even close as "delete". And so on and so on.
Anyway, this is kind of a long and disjointed post, but I really do think that if we're able to get a bunch of people to read what we write, we do a service both to ourselves, to the broader editoriat, and to the project, although of course this is always contingent on writing things that are fair, accurate, true and good. jp×g🗯️ 23:25, 9 November 2023 (UTC)[reply]


I was curious what articles published this year have been popular. Here are the data from the pageviews widget used above:

2023 Signpost articles with >20,000 pageviews so far
Issue and item Title Pageviews
2023-02-20 Disinformation report The "largest con in corporate history"? 20393
2023-05-22 News and notes Golden parachutes: Record severance payments at Wikimedia Foundation 53796
2023-07-17 Recent research Wikipedia-grounded chatbot "outperforms all baselines" on factual accuracy 47453
2023-08-01 News and notes City officials attempt to doxx Wikipedians, Ruwiki founder banned, WMF launches Mastodon server 31496
2023-08-15 Special report Thirteen years later, why are most administrators still from 2005? 22455
2023-08-15 Serendipity Why I stopped taking photographs almost altogether 28205
2023-11-06 Arbitration report Admin bewilderingly unmasks self as sockpuppet of other admin who was extremely banned in 2015 54589

It's an interesting mix, and I'm pleased to see a diversity of authors and article types represented, including research. ☆ Bri (talk) 22:40, 8 November 2023 (UTC)[reply]

These are interesting, thanks for posting! — Frostly (talk) 06:14, 16 November 2023 (UTC)[reply]
I think Recent Research is a bit of an unsung hero here. It's the Signpost's most consistently high-quality segment. Andreas JN466 14:40, 16 November 2023 (UTC)[reply]

More

Pageviews for 2023-11-20 (V)
Subpage Title 7-day 15-day 30-day 60-day 90-day 120-day 180-day
Wikimania Wikimania 2024 scholarships 343 471 547 590 614 627 664
Traffic report If it bleeds, it leads 539 743 794 846 868 886 951
Recent research Canceling disputes as the real function of ArbCom 996 1263 1332 1405 1435 1469 1533
News and notes Update on Wikimedia's financial health 1168 1516 1877 2020 2072 2154 2320
In the media Propaganda and photos, lunatics and a lunar backup 1108 1535 1632 1735 1803 1822 1882


19-22 In the media

@HaeB: Hello, I've just incorporated my "In brief" bit into the current draft for next issue, as you suggested.

Actually, I was thinking about doing another one: I've found an article by an Italian sports magazine that explicitly cites the article for the Peruvian Clásico (an association football derby between Alianza Lima and Universitario) and encourages readers to take a look at it. That's because the latest, title-deciding match between the two teams took place just a few days ago and ended with some... interesting post-game antics, let's put it that way.

I fear this would be a little more off-topic than the "In brief" bits you usually publish, though. What do you think about it? Oltrepier (talk) 17:33, 11 November 2023 (UTC)[reply]

@Bri: Thank you for taking a look at my entry! I don't know if it really deserves to be a lead story, though, since the original article I used as a source is pretty brief... Oltrepier (talk) 11:21, 12 November 2023 (UTC)[reply]

Odd heading for item

There's a heading under In brief that says "Israel and Palestine" but the text that follows says nothing about that, rather, it's about translation of Stephen Harrison's Slate article into Spanish. ☆ Bri (talk) 19:20, 16 November 2023 (UTC)[reply]

19-21 Essay

@Theleekycauldron: Looks short. I'm anxiously awaiting the 2nd half! Is this a more general problem? Is there a good solution for the specific problem? Smallbones(smalltalk) 00:21, 13 November 2023 (UTC)[reply]

not nearly ready for copyedit yet – it's gonna get a lot longer – but glad you like the pitch so far! theleekycauldron (talk • she/her) 00:32, 13 November 2023 (UTC)[reply]
Considering that this hasn't progressed further since the above conversation and that this issue's publication deadline has passed, this piece should probably be dropped or postponed. Regards, HaeB (talk) 02:19, 20 November 2023 (UTC)[reply]

Another for the pile

Wikipedia pays more attention to events in richer countries, say data scientists. Contra: "Wikipedia would love to pay more attention to events in poorer countries but can't because the damn news refuses to write anything about them, say Wikipedians"? jp×g🗯️ 21:13, 14 November 2023 (UTC)[reply]

Toronto bomb threat

Is anybody dealing with this right now? See Wikipedia:Wikipedia Signpost/Newsroom/Suggestions/Archive 35#Suggestion by SYSS Mouse (2023-11-12) --Andreas JN466 14:38, 16 November 2023 (UTC)[reply]

I was there. I have some screenshots of things said on social media and talked with people about what they know and experienced.
One story is that this happened; another story is that this is the latest event in a pattern of harassment or even terrorist threats targeted at the Wikimedia community.
I am not sure if I will make deadline for next issue on covering all of this WikiConference but I did do video interviews of about 20 people and plan to publish them first in a Signpost article about the conference.
I am sure I can contribute to a little coverage by the next issue; I am not sure if I can fully cover the conference by deadline. I will join anything anyone else is doing. Thanks, ping me anyone who wants to talk about the conference. Bluerasberry (talk) 16:29, 16 November 2023 (UTC)[reply]
Red-tailed hawk started a section at Wikipedia:Wikipedia Signpost/Next issue/News and notes#WikiConference North America held in Toronto some days ago which has some relevant notes and links already (I added [3] and [4] yesterday).
Regards, HaeB (talk) 21:09, 16 November 2023 (UTC)[reply]
I need to expand that, yes. Can try to interview an attendee or two. — Red-tailed hawk (nest) 02:51, 17 November 2023 (UTC)[reply]
@Red-tailed hawk, Bluerasberry, HaeB: This still needs work. Anyone have time? I am not au fait with what did or didn't happen. Best, Andreas JN466 15:26, 19 November 2023 (UTC)[reply]
Let me see what I can get up in the next few hours. — Red-tailed hawk (nest) 21:41, 19 November 2023 (UTC)[reply]
@Red-tailed hawk Curious about Special:Diff/1185973996. It looks like you know something that I don't? Regarding the specific content of the link, I don't see anything that could remotely be conceived as problematic, not do I on an (admittedly cursory) review of other recent tweets from the account. And The Forward and Slate magazine are not exactly WP:BADSITES (in fact we prominently featured one of his Slate articles in In The Media not long ago). It also looks like the event's registration was subject to the usual policies, which would usually make one assume that conference participants are unlikely to be well-known banned users, in case that was the concern?
Currently, the WikiConference piece focuses almost entirely on the bomb story, so the link would be helpful for those of our readers who are more interested in the content of the conference itself. Again, you may be better informed about the relevant background here, but generally speaking we shouldn't make content decisions based on rumored past revdels. Regards, HaeB (talk) 05:20, 20 November 2023 (UTC)[reply]
My understanding is that both the Twitter account and real-life name of the associated user are not disclosed on Wikipedia, and that for WP:OUTING reasons we'd hit a wall in revealing those sorts of connections. There was something at a noticeboard a while back that I recall where someone connected a Tweet or an article with the Wikimedian and it got revdel'd; I'll try to find it if I can. — Red-tailed hawk (nest) 05:25, 20 November 2023 (UTC)[reply]
Thanks for explaining, but since we are not connecting the Twitter account and real-life name to his on-wiki identity (which, for the record, I don't know and don't need to know), I fail to see how this could be a WP:OUTING violation. (Note also that he has been cited in mainspace for over a year without any clueless admins or oversighters hitting the revdel button, even though on the site linked there he likewise identifies as "veteran Wikipedian who has edited more than 30,000 pages on the site.")
Since we didn't get to resolve this in time for publication, I have left a comment on the talk page, in the hope it's still useful for some readers. Regards, HaeB (talk) 17:38, 20 November 2023 (UTC)[reply]

19-21 Recent research

As usual, we are preparing this regular survey on recent academic research about Wikipedia, doubling as the Wikimedia Research Newsletter (now in its thirteenth year). 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) 21:13, 16 November 2023 (UTC)[reply]

@JPxG: Should be publishable now if need be, although I may still be adding a bit unless I hear the publication engines revving up. Regards, HaeB (talk) 20:04, 19 November 2023 (UTC)[reply]
  • My piece is rather long but I have so much more I'd like to contribute regarding analysis of what the author's conclusion means to us. I'd like to say that a) it was reinforced by another paper [5] that called the arbcom decisions amoral or "purely procedural ethics". More of this still resonates with me, especially whether Arbcom's form and function are a deliberate design from the beginning: that although it's described by outsiders as a "Supreme Court" of sorts for us [6], its function to quash editor conflict at the cost of everything else is "just how it is". ☆ Bri (talk) 21:29, 19 November 2023 (UTC)[reply]
    I do think there is room for another sentence or two; we also often include "see also"s for related research publications (ideally linking our own previous coverage). Feel free to add in case there is time (looks like we have a significant publication delay). Regards, HaeB (talk) 05:29, 20 November 2023 (UTC)[reply]
  • Can anybody figure out if Law & Social Inquiry that has the research I reviewed is the same as Law and Social Inquiry? It's very confusing, American Bar Association might be involved or might not. The website says "Published by Cambridge University Press on behalf of the American Bar Foundation". Never mind, I determined the ISSNs are the same. ☆ Bri (talk) 00:55, 20 November 2023 (UTC)[reply]

News and notes section about the Foundation’s audit report

Happy to see the Foundation’s audit report being covered in the Signpost. Question for you all: would you consider swapping out the image for something more neutral and less charged? The article points to our reserves, so maybe this image would add more context? Or even just a generic image could work fine. A couple of additional notes for the editors’ consideration:

1. “The Foundation took $180 million in total support and revenue, vs. total expenses of $169 million.” It might be helpful to contextualize these numbers a bit more for the reader. We could add a sentence like “Both the revenue and expenses were within 5% of the $175M budget set in the Annual Plan. Revenue was exceeded primarily due to unrealized investment income, which is determined by fluctuations of investment values that can’t generally be predicted.”
2. “Its net assets grew by $16 million over the year prior, to $255 million at the end of the year.” I think context would also be valuable here, to understand what the $255 million in assets represents. Could we add something along the lines of “This is in line with the Foundation’s working capital reserve target of 12-18 months of operating expenses - which is standard practice for non-profits of the Foundation’s size.”


Let me know if you want to discuss any of this further. Thanks again for helping bring the audit report to a wider audience. –JBaldwin (WMF) (talk) 23:13, 16 November 2023 (UTC)[reply]

I didn't pick the picture, but as for the text, I am rather against having you micromanage our coverage and insert corporate spin. (The 12 to 18 month reserve line in particular is trotted out every year ... it is misleading if you don't at the same time tell people by how many tens of millions you've increased your planned expenses.) Up to the EiC, of course, but my answer would be a "No". A "Hell, no", in fact.
If you want your version in the Signpost it would be best to submit a text for the "News from the WMF" slot. Regards, Andreas JN466 02:22, 17 November 2023 (UTC)[reply]
Question for you all: would you consider swapping out the image for something more neutral and less charged?
Is this request about the image a joke? Sometimes humor does not translate well and I am having trouble understanding the request. Bluerasberry (talk) 16:06, 17 November 2023 (UTC)[reply]
The quark image was a snarky placeholder that I substituted for the original. See Special:Permalink/1185359699 for what was there when Foundation personnel requested a change. ☆ Bri (talk) 16:50, 17 November 2023 (UTC)[reply]
I've now put in a bar chart showing WMF finance data from 2003-23. Andreas JN466 19:57, 17 November 2023 (UTC)[reply]

Credit for DALL-E prompt engineers

I recently read of a future job description, or maybe present already: prompt engineer, for a person who expertly interacts with an AI to produce a result. Well, I've already been working with a fellow Wikimedian who I guess is doing that; they have created many images used in The Signpost starting a bit more than a year ago (see their last five uploads). Do you think they should be credited in the byline for images created just for us? I recommend the following:

In the media

Propaganda and photos, lunatics and a lunar backup

By Bri, Oltrepier, Smallbones, and HaeB

Original illustration by Mayopotato

Is this fair? ☆ Bri (talk) 01:53, 20 November 2023 (UTC)[reply]


According to Wikipedia, The byline [...] gives the name of the writer of the article. A frequent journalistic practice (in newspapers etc.) seems to be to credit illustrators in image captions instead, which yes, seems to be a fine idea (even if not required by copyright).

I generally think AI-generated images as illustrations area a great idea - thanks Mayopotato for contributing these! But can I ask why we are apparently using the rather outdated DALL-E 2 model like it's 2022? This is like eons ago in this area and the quality of those models has improved immensely since then. Compare File:DALL-E Austronaut Library.png with [7] (click side arrows to see all four versions), which is what I just got on first try with the successor model DALL-E 3 (freely accessible via Bing AI) using the same (apparent) prompt. By the way, per c:COM:AI, the model and prompt that were used should be documented in the image description.

Regards, HaeB (talk) 02:07, 20 November 2023 (UTC)[reply]

That is nice. I switched in the Bing image (if you want image credit in Commons, go ahead and add your name there). ☆ Bri (talk) 02:23, 20 November 2023 (UTC)[reply]
I was accidentally pinged above and didn't have a clue, but I think I've figured it out now. I'd go with credit in the caption, perhaps with "(AI)" in there somewhere. For the bigger picture, I've noticed that we've had a lot of editorial cartoons/photos lately, and more-or-less I like them in ITM. I'll suggest @Mayopotato: make a bunch for every issue (depending on the amount of time he has) in the style of editorial cartoons (but photo-like is ok). Maybe even one for the menu page, that's like the main editorial cartoon on a newspaper's opinion page. Or even something like our old series WikiWorld. Maybe we could even get Annie Rauwerda to suggest articles/text? I'd say go with it as far as we can take it. If it doesn't work- it's pretty easy to just stop. Smallbones(smalltalk) 03:26, 20 November 2023 (UTC)[reply]
BTW, I was tempted to put this photo in Bri's Recent research piece but the "Equal justice under law" carved in the building doesn't show up very clearly. Well maybe we could get an AI photo of the lady with the sword and the scales, with a tear coming out from under her blindfold. Smallbones(smalltalk) 03:46, 20 November 2023 (UTC)[reply]
  • We previously used some DALL-E 2 images, the August 22 issue had a lot of them (and we had images on the main issue page as well). Me and Frostly made a few around that time. It's a little time-consuming and expensive, and making thumbnail images for articles is not really well-supported by our current layout code, so I haven't been doing it lately. In general, I'm not a gigantic fan of the Wikipedia image credit model -- you're supposed to totally abstain from a text image credit, which is doing too little, and you're also supposed to put a giant pain-in-the-%@# link on the image itself, which is doing too much. I think it'd be saner to just have a text credit below images, and I think it's good to have done this here. I wouldn't be opposed to a general style-guide where we just do this for our images overall. jp×g🗯️ 09:52, 20 November 2023 (UTC)[reply]
    Expensive? One can use DALL-E 3 for free on Bing now [8].
    the August 22 issue had a lot of them - to clarify for others reading along, I guess JPxG means one of the (several) August 2022 issues, already linked above.
    making thumbnail images for articles is not really well-supported by our current layout code - well, they don't have to be thumbnails.
    I think it'd be saner to just have a text credit below images, and I think it's good to have done this here. - to clarify, we went without any credit in the new ITM (as the prompter of the currently used image I'm not complaining, although I might still correct the image description on Commons per Bri's suggestion above).
    Regards, HaeB (talk) 17:14, 20 November 2023 (UTC)[reply]

19-22 In the media

Hello! I've noticed that a snippet from a previous thread of this page has been incorporated in the Newsroom's "In the media" panel for the next issue: was it a glitch, or did anyone do it on purpose?

By the way, I would be more than happy to write a short entry for that article about the Peruvian Clásico, and I've already got another interesting story on my mind, as well! Oltrepier (talk) 21:12, 22 November 2023 (UTC)[reply]

I think the glitch is that the comments in the Newsroom page don't automatically clear out after the new issue, maybe not until the page for the new issue is created, or even maybe not at all! @JPxG: I think this is an artifact of your recent work on this issue. In general, though, that work has been useful.
@Oltrepier: thanks for the contributions in the last issue. I think *everybody* liked them very much (or should have), and you should keep on contributing as much as you like. In fact we should all encourage you. It's especially good that we're getting (through you) more non-English language sources since we are supposed to be covering all Wikipedia languages. Keep up the good work. Smallbones(smalltalk) 18:30, 25 November 2023 (UTC)[reply]
@Smallbones I genuinely hope I can keep doing so, thank you so much for your support! : ) Oltrepier (talk) 21:46, 25 November 2023 (UTC)[reply]
@Oltrepier In a narrow sense this appears to have been your fault, because you created that section with the wrong heading - it should have started with "19-21" (like #19-21 Essay or #19-21 Recent research above) instead of "19-22"; for the same reason, it also wasn't included in Wikipedia:Wikipedia Signpost/Newsroom at the time when it should be been (i.e. until the publication).
In a broader sense though, as Smallbones already alluded to, this is a problem with the new newsroom system itself that JPxG implemented recently, which relies on these cryptic numbers. (Or who among the Signpost team 1. always knows the volume and issue number of the upcoming issue and 2) that the sections actually have to use the volume and issue number of the *preceding* issue because JPxG wasn't able to find a better technical solution?) Hate to go "told you so", but this is exactly the kind of problem I predicted in the recent discussion about these changes (where I also suggested an IMHO less confusing solution, which admittedly would take more work and skill to implement). As said there, JPxG's sloppy work on such tech issues has again and again been causing significant confusion and lost time for others on the team. This is another such instance. Regards, HaeB (talk) 05:31, 27 November 2023 (UTC)[reply]
Since you asked, I know the issue number because I always record my contributions in my private roster using it. ☆ Bri (talk) 06:20, 27 November 2023 (UTC)[reply]
@HaeB @Bri @Smallbones Ok, thank you all for clarifying.
By the way, I've just added a new short entry to the "In the media" draft, and I've also got an interesting idea for the second lead story, hopefully I can add it sooner rather than later... Oltrepier (talk) 21:29, 27 November 2023 (UTC)[reply]
The template that generates the "create a new section" link from the Newsroom automatically inserts the section with the correct name. Previously, the system was to just not transclude items from here to the Newsroom at all, and there was a separate set of comments there and here; this also had to be reset regularly. @HaeB: I am a little saddened to hear this; I've pinged you before (when I've had a lot of free time) to ask what specific technical improvements you think I should focus on, and gotten no response at all; if your preference is for the issues to remain unfixed and you just like calling me a fool every few weeks, I guess that is fine, but let me know so that I can stop trying to actually think of what should be addressed. jp×g🗯️ 19:02, 28 November 2023 (UTC)[reply]

@Smallbones: For the record, I think I've found the English version of that Haaretz article you've been looking for, hopefully it helps. Didn't we already address the news in a previous issue, though? Oltrepier (talk) 18:03, 29 November 2023 (UTC)[reply]

Wow @Oltrepier:, you're right. It looks like User:Ca added the two line blurb, which we used as is, on Nov. 17 when the article first came out. On Nov. 26 Haaretz's standard email notification sent me the same story in Hebrew (identical except for the headline) and I just missed the connection. Thanks - likely I'll just delete the section for this issue. Smallbones(smalltalk) 19:46, 29 November 2023 (UTC)[reply]
@Smallbones Oh, ok, but I think it might actually be notable enough to keep it!
Oh, and I'm also completing my own entry for the second lead story... Oltrepier (talk) 20:16, 29 November 2023 (UTC)[reply]
Ok, my draft should be live and ready for copyediting now! Oltrepier (talk) 21:41, 29 November 2023 (UTC)[reply]
It looks good to me. I just added a link to when we covered her Nobel Price acceptance speech. Smallbones(smalltalk) 17:58, 30 November 2023 (UTC)[reply]

As far as the Haaretz article. I'm pretty sure I don't have the time or the skills to do a good write up. It looks like a minefield to me, but anybody else who wants to write it should go ahead. Smallbones(smalltalk) 17:58, 30 November 2023 (UTC)[reply]

@Smallbones I think I can at least help you write a proper entry, but I won't have enough time to do it until tomorrow... Oltrepier (talk) 21:14, 30 November 2023 (UTC)[reply]
Thanks (if you want to take on the whole thing!) Smallbones(smalltalk) 22:38, 30 November 2023 (UTC)[reply]
Actually, I'll likely be late with it, I'm so sorry... Oltrepier (talk) 21:40, 1 December 2023 (UTC)[reply]
@Smallbones Ok, surprisingly enough, I've managed to prepare and submit a last-gasp write up for the Haaretz article: since it's under pay-wall, I couldn't add many details aside of the Yoram Cohen dispute, but I tried to do my best. I hope it's good enough! Oltrepier (talk) 18:09, 2 December 2023 (UTC)[reply]
OK @Oltrepier:, for some reason (maybe because I accept their email reports) they let me see the full article. I've got an hour of work to do on my Disinformation report before deadline! after that I'll go over the Haaretz section in ITM, then there's a bunch of other stuff I've left hanging... Smallbones(smalltalk) 18:55, 2 December 2023 (UTC)[reply]

Images for lead story

I have created some images for the first ITM item "Edit wars over real-life war in Gaza" (Haaretz report on editwarring). I'd like the team to select one.

Cheers ☆ Bri (talk) 22:23, 2 December 2023 (UTC)[reply]

I think #3 should be eliminated because I've seldom seen Israelis (or Palestinians) wear ties. #2 is better than #1 IMHO. Smallbones(smalltalk) 23:47, 2 December 2023 (UTC)[reply]
  1. 2 is better, if only because the typewriters in #1 make no sense.
Headbomb {t · c · p · b} 00:12, 3 December 2023 (UTC)[reply]
  1. 2 looks good to me, as well! Oltrepier (talk) 09:06, 3 December 2023 (UTC)[reply]
Please include the prompts used (per c:Commons:AI-generated_media#Description). Regards, HaeB (talk) 23:06, 3 December 2023 (UTC)[reply]
  • I have been working on some CSS/html/template/js shiznit lately and I think I might be able to make the "images on the Signpost front page" thing work again, without it requiring 800 hours of labor like it did when we tried this last year. jp×g🗯️ 02:00, 4 December 2023 (UTC)[reply]
    I did and it works. Article headers now take a piccy= param which can be used for a header image that'll automatically get put on the main Signpost page and also in the archives (which, incidentally, also have subheadings too). That and some CSS fixes for templates and the external site. Ready to be ready now. jp×g🗯️ 07:01, 4 December 2023 (UTC)[reply]

Publishing schedule

It's a good time to discuss the Dec.-Jan. schedule now. We're scheduled to publish Sunday Dec. 3, and then the next issue would be Sunday Dec. 17, and we're into the Christmas-New Years period which can be an interesting time to publish. On the regular 2 week schedule, the next after that would be Dec. 31, which would be a horrible publishing date (December 24 would be just as bad). I'll suggest we just wait another week and go past the New Year (with a new volume number! #20) and publish on Jan. 7, 2024.

I've always planned to be out of touch over the holidays, but it never worked out that way. My wife always has 2+ weeks off over the Xmas-NewYears holiday, so we always plan to travel. This year though she'll be traveling abroad alone to meet family, and I'll be taking care of the other family here. But I'll be fine with the schedule I outlined above.

BTW volume 20 just means that we'll be completing 19 years of publication soon, it doesn't mean that it will soon be our 20th anniversary. That'll be January 10, 2025 (or Sunday Jan. 12, 2025). We should probably start something though that leads up to the 20th birthday. Smallbones(smalltalk) 23:41, 27 November 2023 (UTC)[reply]

On the regular 2 week schedule, the next after that would be Dec. 31 - that's not "the regular schedule"; per this discussion, the current default schedule is the first Sunday and third Sunday of the month (i.e. twice per month, as also stated in the Signpost's "About" page, not every two weeks). In other words, your proposal is exactly the default schedule. Regards, HaeB (talk) 03:02, 28 November 2023 (UTC)[reply]
Whoops! My mistake - then nothing to worry about. One possible thing to do, though it would likely mean a lot of extra work for @Igordebraga:, would move the year-end summary of Traffic report, to a Jan-November summary and publish it on December 17, rather than in a January issue. Would that be possible? Smallbones(smalltalk) 17:10, 28 November 2023 (UTC)[reply]
Better keep the Annual Report in January, it's an yearly deal so it only really works with 365 days of traffic data. December is business as usual. igordebraga ≠ 01:15, 29 November 2023 (UTC)[reply]
Nineteen years is old enough to get a beer in Canada (as everyone who's ever turned 19 in Michigan will already know). The first sunday of January is the 7th, but I think it may be more apropos to shoot for the 10th (since that is the 19th anniversary and all). jp×g🗯️ 20:40, 28 November 2023 (UTC)[reply]
@JPxG @Smallbones @Igordebraga So, the plan is to address the WMF report on this year's Top 25 list only in January, right? I hope I'll be able to help you with it, by the way! Oltrepier (talk) 21:04, 6 December 2023 (UTC)[reply]

Traffic report blurb

To me, by far the most interesting item on the 20 articles in the traffic report is the possible existential threat to humanity. I wobbled on whether to bump the apparently fascinating (to some) cricket games from the blurb, and chickened out, and put them first. If somebody wants to take a crack at the other thing, go for it! ☆ Bri (talk) 00:35, 2 December 2023 (UTC)[reply]

I've never been good at TR blurbs, or blurbs in general, but here's an off the cuff try.
If you think one sport taking more than a quarter of the slots in these lists isn't cricket, look for movies, celebrities, and political follies. Or a possible existential threat to humanity from an unusual fired-for-the-weekend CEO!
Maybe a bit over-the-top, do with it what you like. Smallbones(smalltalk) 17:25, 2 December 2023 (UTC)[reply]
@Bri: Maybe relevant: there's a prediction market on this. jp×g🗯️ 20:01, 5 December 2023 (UTC)[reply]
Holding steady at 25%, that's actually pretty high for a predicted breakthrough. ☆ Bri (talk) 20:35, 5 December 2023 (UTC)[reply]

19-23 Featured content

@Adam Cuerden Do you need help to complete your draft? Sorry for not asking you before... Oltrepier (talk) 09:04, 3 December 2023 (UTC)[reply]

If possible. I'm kind of in a crunch time here. Adam Cuerden (talk)Has about 8.6% of all FPs. 15:59, 3 December 2023 (UTC)[reply]
@Adam Cuerden I've tried to knock out some of the short poems for the FA section (at least for the articles I'm more familiar with) and make a few more adjustments, hopefully it helps. Oltrepier (talk) 17:48, 3 December 2023 (UTC)[reply]

3 December issue

I am trying to take care of a bunch of stuff before we publish. I will be around tomorrow. jp×g🗯️ 09:19, 3 December 2023 (UTC)[reply]

"Tomorrow" was stated about 7 hours ago, so I hope this actually means publishing will be on time about 2000 UTC. ☆ Bri (talk) 16:19, 3 December 2023 (UTC)[reply]
In the far future of the year 2000 humanity will have advanced past the need for time zones. jp×g🗯️ 18:18, 3 December 2023 (UTC)[reply]

Looks like 7 pretty good articles, some copyediting needed. Smallbones(smalltalk) 00:15, 4 December 2023 (UTC)[reply]

@Smallbones: Wanna take a look at Wikipedia:Wikipedia_Signpost/Next_issue/Comix? jp×g🗯️ 01:42, 4 December 2023 (UTC)[reply]
@JPxG: - yes, it's pretty good, say 96.5/100. Did you plagiarize the captions? (joke) I'll copy edit, which should take three minutes. The best humour article we've had in a long time. Smallbones(smalltalk) 02:21, 4 December 2023 (UTC)[reply]
OK CE done - I could only verify the last part of the blurb quote [9] but it looks ok. I don't know how you got a laptop on the table in the 3rd cartoon, but if possible you should remove it. Smallbones(smalltalk) 02:29, 4 December 2023 (UTC)[reply]
LOL -- I don't actually know what that is. The image is a direct scan from a Danish magazine from 1877; I think it's just some kind of briefcase or craft kit that happens to look like one. It would be a pretty stupid thing to add, though, if it wasn't there to begin with. jp×g🗯️ 07:07, 4 December 2023 (UTC)[reply]

Out

Image script worked, so there are piccies on the main page and also in the archives. I will incorporate this into the preloads and documentation -- it's an invisible param that goes into the article header with the title and author. jp×g🗯️ 10:53, 4 December 2023 (UTC)[reply]

@JPxG Thank you for implementing this!
The only problem I've got with the current script is that some images (most specifically, three of them on today's issue) look giant in comparison to anything else... Is there an easy way to make all the pictures fit in a "standard", proportionate format? Oltrepier (talk) 20:25, 4 December 2023 (UTC)[reply]
Yeah, I am messing around with {{Signpost/snippet/sandbox}} as we speak. I am not 100% on what to do; I think that some kind of CSS crop may be condign, but I will have to take a look at {{CSS crop}} and see how grotesque of a hack it is. It might just be too half-assed to be worth using, but if it's actually good, I will test it out. jp×g🗯️ 20:27, 4 December 2023 (UTC)[reply]
Nice to have the images back. Some notes:
  • The first image is broken right now in Wikipedia:Wikipedia Signpost/Archives/2023-12-04 (but not in Wikipedia:Wikipedia Signpost).
  • Let's make sure that the image for each story is actually also used in the corresponding story. That is not the case right now for the photo of the mysterious gentleman in a suit. (Unless the reader is a knowledgeable fan of a particular sport, they would need to scroll through much of the tables in Wikipedia:Wikipedia Signpost/2023-12-04/Traffic report and do some skilled guesswork and clicking on various biography articles to find out that he is the same as the athlete in sportswear with sunglasses.)
  • This is also important because the current solution likely violates CC licenses, as the images are not clickable in Wikipedia:Wikipedia Signpost - i.e. the file description pages with the required attribution and licensing information are not accessible. (For some reason they are clickable in Wikipedia:Wikipedia Signpost/Archives/2023-12-04 though.) I doubt that having the image with file description link on the linked page is sufficient either, but at least it would ameliorate the situation a bit.
  • Looks like the template's documentation still needs to be updated.
Regards, HaeB (talk) 02:49, 5 December 2023 (UTC)[reply]
Are you sure the image isn't a caching issue? It was messed up earlier today but I thought I fixed it.
The whole thing is pretty experimental overall; I haven't added documentation yet because I am not sure if this is the best way to implement it. That is: having SPS constitute markup for archive/issue pages based on information in the articles, then having that stored separately on different wikipages. For a simple example, changing one of the pictures for one of the articles right now requires a sequence of three separate edits (article metadata, Signpost front page and archive issue page). I don't know if it is smart to stick with this, so I don't want to commit it to documentation just yet (and everything still works fine without including a piccy param in the header templates). The image attribution thing is another issue I'm not quite sure about. A bunch of the images (i.e. the midjourney/stable diffusion ones at least) are public domain; for the ones that aren't I am trying to figure out an attribution overlay CSS.
As far as I can tell, there is no other website besides Wikipedia where "making the image a hyperlink to a different website that has information about its author provided if you scroll down" is considered necessary and "crediting the author with an attribution line below the image" is forbidden. This seems to be a quirk particular to Wikipedia; the Creative Commons website doesn't say anything about needing images to only be displayed with the full image as a hyperlink to an attribution page. About the extent to which notices are required, they say

In 4.0, the manner of attribution is explicitly allowed to be reasonable to the means, medium, and context of how one shares a work.
In the 1.0, 2.0, 2.5, and 3.0 licenses, attribution may be reasonable to the medium or means, and applied to all elements other than certain notices where the requirement is firm. In 4.0, this explicit permission applies to the medium, means, and context of use. We believe this to be a clarification rather than a change: attribution reasonable to the means, medium, and context of use should be permissible for works under any CC license. Additionally, the pre-4.0 licenses specified that credit in adaptations and collections should be at least as prominent as credits for other authors; 4.0 is not specific in this regard.

As far as I can tell, the "image as hyperlink" practice exists here because of our practice of refusing to give attribution in article text or image captions, necessitating something else be done (and to be honest, I'm really not sure that Wikipedia actually provides reasonable attribution -- I certainly wouldn't be satisfied with a website using my photographs and then saying "no it's okay bro we don't give you any acknowledgement but we made the image be a hyperlink to a Commons page which incidentally mentions your username 2 screens down").
But that is another thing for another day. Here I will try to figure out something today for getting attribution publicly on the front page as well as the article pages.
Overall, I'm thinking that having separate static pages for everything might not be a great idea. Module:Signpost has proven quite powerful and useful wherever it's been used -- e.g. there are about a hundred tag and series pages like Wikipedia:Wikipedia Signpost/Series/RfA reform and Wikipedia:Wikipedia Signpost/Tag/fromtheeditor, with thousands of articles between them. But since these use the module, they can all easily be modified by simple edits to list-maker templates (i.e. if we decided we wanted to include bylines or subheadings on the tag pages or something). Contrariwise, if we wanted to include bylines on archive pages -- even though this information is already stored in the module's data indices -- it'd still require a gigantic bot run across hundreds of pages running weird bespoke code to properly format the bylines onto the articles (which would itself require a lot of oversight and review because of oddball formatting issues on the archive pages). So I am not completely certain that adding another parameter to our weird "static archive page" thing is the best way to go -- it might just be better to use all the metadata from the module indices and generate pages from that instead. Who knows. jp×g🗯️ 05:32, 5 December 2023 (UTC)[reply]

19-23 Recent research

@HaeB: Just saw this: "Toxic comments are associated with reduced activity of volunteer editors on Wikipedia, PNAS Nexus, Volume 2, Issue 12, December 2023". Seems slightly interesting but also kind of bizarre. Mostly, what they do is carry out sentiment analysis on a bunch of user talk page diffs, run them through a black-box algorithm to determint their "toxicity", and then compare them to user contribs graphs. The conclusion they draw is that, since many users' contributions cease after receiving a "toxic message", "Wikipedians might reduce their contributions or abandon the project altogether, threatening the success of the platform".

I am not so sure about this. Their methodology is somewhat troubling. They go into great detail on what a shame it is that people are leaving Wikipedia due to these toxic comments, and how the community needs to take a strong stand against "toxicity", but do not actually give examples of what is or isn't a "toxic" comment. Most user warning templates, like {{uw-vandalism4}}, have pretty harsh language ("You may be blocked from editing without further warning the next time you vandalize Wikipedia").

This becomes very important when combined with the next omission, which is rather disturbing: they do not seem to have made any attempt whatsoever to determine whether the users were blocked. The words "Block" or "ban" does not appear anywhere in the paper and the topic is never mentioned; it seems like they simply assumed that a user's contributions stopping at a certain date indicated that they "abandoned the platform".

It doesn't really seem to me like the data demonstrates what they're claiming it does, at least based on the graphs they show. They've done a bunch of regressions on the data, sure, but it seems equally likely that what the data actually demonstrates is "user accounts that get warned for vandalism often stop editing afterwards", which is trivially true, and definitely isn't evidence for this:

A suggested solution to this problem has been the red-flagging of harassment and harassers (46). However, the opinion that toxic comments are negligible and should be seen as merely over-enthusiastic participation is still present among editors (25). Furthermore, various anti-harassment measures have been de-clined multiple times by the community, as they were seen to slow the process of content creation (57, 58). Based on our find-ings, we believe there is a need to reevaluate these policies, and more research attention is required to understand the impact of potential interventions.

Of course, this isn't false, but it really seems questionable to me whether any of the data in this paper is actually evidence to support it. What do you think? jp×g🗯️ 00:16, 6 December 2023 (UTC)[reply]

Just quickly: I haven't read the full paper yet, so I don't have an opinion on several of the issues you mention. But the preprint version, which came out some months ago, has been on the the to-do list for RR for some months already. As noted there, this comment (by a researcher from EPFL) seems worth considering: "I've found the title of this paper a bit misleading — the finding is purely correlational: there is no attempt to control for the confounders that might be present."
Now, that title was changed in the now published peer-reviewed version, from "Toxic comments reduce the activity of volunteer editors on Wikipedia" to "Toxic comments are associated with ..." (my bolding). But evidently tons of people still interpret it as a causal claim. Earlier today, after (re)tweeting the paper from @WikiResearch, I reached out to the authors (or the two of them I was able to find on Twitter) about their thoughts on this issue, let's see if they respond.
If you want to write up a review for RR, I'd be happy to assist.
PS: The supplementary material contains some concrete examples of comments with their toxicity ratings.
Regards, HaeB (talk) 00:33, 6 December 2023 (UTC)[reply]
Oh, I am a fool and missed that. Some interesting stuff there:
These diffs go back a very long way (some as far back as 2004). But then the paper says stuff like: "A suggested solution to this problem has been the red-flagging of harassment and harassers (46). However, the opinion that toxic comments are negligible and should be seen as merely over-enthusiastic participation is still present among editors (25)." The idea, I guess, or at least the implication, is that Wikipedians have never done anything about civility, or that we think it isn't a problem, or that we need to take action because there's so much of it, et cetera -- but the evidence of this is a bunch of comments that people already got blocked for, and also some people were assholes in 2005... jp×g🗯️ 02:54, 6 December 2023 (UTC)[reply]
I guess we can expand the list of spurious causation-inferrers to the authors of the paper. jp×g🗯️ 05:37, 6 December 2023 (UTC)[reply]
Looking into this deeper: googling the string from the 0.99 toxicity comment gives https://storage.googleapis.com/kaggle-forum-message-attachments/300575/8842/oof_greatest_hits.xlsx which is an Excel file called "oof_greatest_hits.xlsx". In the fifth tab, "threat_ranked_high", the topmost row is this: Please stop. If you continue to ignore our policies by introducing inappropriate pages to Wikipedia, you will be blocked. Much to think about. jp×g🗯️ 06:03, 6 December 2023 (UTC)[reply]
I am not sure about the provenance of that file, although it does seem to match up rather nicely with the data they give. But nonetheless, I have obtained developer access to the Perspectives API, and I can confirm that the text of {{Uw-own3}} does indeed give 0.8039029 for ATTACK_ON_AUTHOR (above the .80 threshold they give). I will have to look into this in some more detail. jp×g🗯️ 08:52, 6 December 2023 (UTC)[reply]
Not sure what the ultimate source of the corpus was. meta:Research:Detox/Data_Release seemed likely; it was used in a Jigsaw paper in 2017. The comments in it are from between 2001 and 2015. But one of the diffs above is from 2021(?) so I don't really know. jp×g🗯️ 09:46, 6 December 2023 (UTC)[reply]

Technology report

Hey everyone, because I have some experience in working with you already, I've boldly added a post I'd very much like to see in Signpost. Of course, as a Wikipedian working for the Foundation, I know that Signpost is totally yours, so if you decide not to have this text or to move it around, I'll understand. At the team, we just wanted to make sure that this news reaches as many editors as possible. Signpost would be perfect for that. Thanks! SGrabarczuk (WMF) (talk) 17:47, 7 December 2023 (UTC)[reply]

Retrieved from "https://en.wikipedia.org/enwiki/w/index.php?title=Wikipedia_talk:Wikipedia_Signpost/Newsroom&oldid=1188784532"