User talk:Iridescent

The arbitration committee "assuming good faith" with an editor.

Archives [Jan 06‑Oct 07] [Nov 07‑May 08] [Jun 08] [Jul‑Aug 08] [Aug‑Sep 08] [Sep‑Nov 08] [Nov‑Dec 08] [Jan 09] [Feb‑Apr 09] [May‑Jul 09] [Jul‑Oct 09] [Oct 09‑Apr 10] [May‑Jul 10] [Aug‑Oct 10] [Nov 10‑Apr 11] [May 11‑Apr 13] [May 13‑Apr 15] [May‑Aug 15] [Sep 15‑Jan 16] [Feb‑Apr 16] [May‑Jul 16] [Aug‑Sep 16] [Oct‑Nov 16] [Dec 16‑Jan 17] [Feb‑May 17] [Jun‑Oct 17] [Nov 17‑Jan 18] [Feb‑Jul 18] [Jul‑Oct 18] [Nov‑Dec 18] [Jan 19] [Feb‑Apr 19] [May‑Jun 19] [Jul 19] [Aug‑Oct 19] [Nov 19‑Jan 20] [Feb‑Mar 20] [Apr‑Jul 20] [Aug‑Oct 20] [Nov‑Dec 20] [Jan 21] [Jan‑Mar 21] [Apr‑May 21] [Jun‑Aug 21] [Sep‑Dec 21] [Jan 22] [Feb‑Mar 22] [Apr‑Jul 22] [Aug‑Nov 22] [Dec 22‑]

Universal Code of Conduct

Morning Iri and page watchers

I'm being asked to cast a vote on the Universal Code of Conduct, but I have to say I'm not entirely sure what it entails. I'm very late to this party and haven't been following the drafting or earlier phases of this at all, and to be honest in general I don't pay that much attention to what's going on at the WMF - my focus is on what I can do to make the encyclopedia better for readers. But since I've been asked to vote, I suppose I should give it some consideration. Is there any sort of big en-wiki meta thread or forum where people are debating the pros and cons? Outside of a slight sense of exasperation that the WMF are focusing so heavily on this and not on improving resources for editors or expanding the Wikipedia Library (as some have suggested), the high-level language at m:Universal Code of Conduct looks superficially harmless, but I fear the devil might be in the detail and that this could have a practical negative effect on the way we conduct ourselves. Any thoughts? Cheers  — Amakuru (talk) 10:56, 14 March 2022 (UTC)

Strictly speaking, the vote is about the enforcement guidelines of UCoC and not the UCoC itself. The enforcement guidelines are being discussed at meta:Talk:Universal Code of Conduct/Enforcement guidelines and meta:Universal Code of Conduct/2021 consultations/Roundtable discussions/Questions and meta:Talk:Universal Code of Conduct/Enforcement draft guidelines review, their actual text is at meta:Universal Code of Conduct/Enforcement guidelines. Jo-Jo Eumerus (talk) 11:14, 14 March 2022 (UTC)

Oh right, yes. Enforcement. Thanks... still looks a bit too much to read, so I might just move on for now!  — Amakuru (talk) 16:10, 14 March 2022 (UTC)

My rather rambling thoughts on UCoC are at User talk:Iridescent/Archive 45#UCoC—it was a few months ago, but my opinion hasn't particularly shifted. The TL;DR summary would be "I understand why they're doing it, to allow the WMF to impose control on smaller wikis in the event that they get hijacked by cranks. However I don't think they appreciate how much blowback they're going to get when they inevitably try to use it to impose their own extremely US-centric value system on any of the big wikis, nor do I think the proposed mechanisms contain adequate safeguards against people using it as a score-settling mechanism. (Compare how often "civility" is used as a pretext to silence opponents in arguments as opposed to genuine behavioral issues.) I also have extreme issues with some of the wording; in particular I think forcing Wikipedia participants to endorse The Wikimedia movement does not endorse "race" and "ethnicity" as meaningful distinctions among people is offensive, in some jurisdictions probably illegal, and would have unintended consequences such as making participation in something like Wikipedia:Black WikiHistory Month a blockable offense. As such I'd encourage everyone to oppose it as things stands, even though this vote is largely meaningless since the eligibility criteria allows the WMF payroll to block-vote so it's almost certain to pass even if there's overwhelming opposition from the actual editor communities." ‑ Iridescent 16:12, 14 March 2022 (UTC)

@Amakuru I'm doing the same. To be fair to us, I think most people aren't voting. I dream of horses (Contribs) (Talk) 16:13, 14 March 2022 (UTC)

I voted no, and if it passes, will be resigning my admin bit. I tend to agree with Iri on this, and find the meddling into encyclopedia affairs by the WMF to be them yet again forgetting what they are supposed to be about. Talk about scope creep! Ealdgyth (talk) 16:27, 14 March 2022 (UTC)

@Ealdgyth, the cynic in me says that's part of the thinking behind it; they know this paragraph is going to force mass resignation of admins and functionaries across the entire ecosystem, thus giving the opportunity on the bigger wikis to force a 'temporary' lowering of requirements and the packing of RFA, Arbcom etc with True Believers, and on the smaller wikis to declare an emergency and impose direct rule by the stewards. I'm not conspiracy-minded enough to think this was the reason this is being forced through—the WMF has a long track record of coming up with stupid ideas through the best of intentions—but if the thought has occurred to me, I'm sure it's occurred to them. ‑ Iridescent 18:11, 14 March 2022 (UTC)

The problem for them, of course, is that if they mass-force an admin resignation on en-wiki, I'm not sure they could get any new ones elected. You'd think they'd have figured out from the Fram-mess that much of en-wiki is very ... jaundiced .. towards WMF "initiatives". They may think that because Fram didn't succeed at a second RFA that the majority of people on en-wiki were on the "WMF side" in the dispute, which is the wrong message to have taken from that mess. Heh. It'll be "interesting times" that's for sure. Ealdgyth (talk) 02:01, 15 March 2022 (UTC)

They don't even need to force a mass resignation; if you read the wording of the paragraph I linked, it gives them authority to conduct a purge of heretics a year after ratification (i.e. March 2023).

Remember that, aside from a handful of people like WAID who are still active on the wikis, WMF-ers are either outsiders who never had an idea what goes on, or former editors whose experiences are out of date. Their knowledge of how we actually operate is going to be second-hand (and to be clear, this isn't necessarily a criticism; for at least some positions at the WMF it's completely reasonable that the person doing the job doesn't have any strong on-wiki relationships). Since the people passing on that knowledge are going to be either (a) the type of people who attend WMF events, (b) the type of people who submit reports to Trust & Safety, (c) the type of people who hang around on Meta or (d) those editors they know in real life, I don't see how they couldn't get a skewed view of what we actually do, particularly when you multiply it by the fact that they're having to perform the same "what impact is this going to have?" assessment across around 600+ different projects.

The regular people who just get on with things aren't by-and-large the people who ever come to their attention. The people who come into extended contact with the WMF are a mix of diehard true believers telling them what a great job they're doing; people trying to get WMF funds and/or technical support for a pet project (or latest scam); and problem users so problematic that their cases have been escalated to the highest level. It would be more of a surprise if the WMF didn't have the collective impression both that the wikis were full of argumentative obsessives who need to be kept in line, and that there exists a substantial caucus of people who think the WMF is doing a fantastic job. ‑ Iridescent 05:19, 15 March 2022 (UTC)

re. the WMF skewed perspective part: weren't the UCoC, and its enforcement guidelines, written by volunteer editors though? I know the enforcement guidelines drafting committee included Barkeep49, Vermont and MJL, for example, and I think all three are folks with a good understanding of how Wikipedia, or at least English-language Wikipedias, work. ProcrastinatingReader (talk) 16:12, 15 March 2022 (UTC)

These are the people who wrote UCoC. By my count there were five paid WMF apparatchiks (one of whom was using their personal rather than their WMF account so is listed under 'volunteers' rather than 'staff'), three hyper-insiders from the WMF's byzantine world of makework committees, and one actual non-greasy-pole-climbing Wikipedia editor. Trying to pass it off as "written by volunteer editors" is stretching the truth at best. ‑ Iridescent 18:27, 15 March 2022 (UTC)

Fair enough. I remember the makeup being more representative, but I guess I was misremembering. ProcrastinatingReader (talk) 20:56, 15 March 2022 (UTC)

The enforcement guidelines were written by a broader selection of editors. isaacl (talk) 03:28, 16 March 2022 (UTC)

This myth that WMF staff don't edit just won't die.  A couple of months ago, I tested some polling software, and I picked volunteer edit counts as a handy subject.  Here were the results:

• 9 – no edits as a volunteer
• 11 – one to nine edits as a volunteer
• 19 – enough to get auto-confirmed
• 6 – more than 100 edits
• 20 – more than 1,000 edits

Out of the 65 respondents before we reached the polling software's limit, 86% of staff had made at least one edit as a volunteer, and 20 WMF staff members have made thousands of edits.  That's 30% of respondents in that top tier.

After that, 17 more staff posted that they had made 1,000+ edits; including several who have made more than 100,000 edits.  Two more posted  in the 1–9 category and two more in the 100–999 category.  I don't think the proportions are reliable (this was not an anonymous pool, and it's easier to brag that you've made thousands of edits than to admit that you've only made a few or none), but I do think that it's reliable to say that at least 37 WMF staff members have made more than 1,000 edits.  If you remember your numbers from Template:Registered editors by edit count, you'll remember that this means that there are dozens of WMF staff who have more editing experience than 99.9% of all registered editors. In other words, it's not just me.

Also, if anyone seeing this happens to be looking at the job openings, then I recommend putting your username in your application. The old rule about hiring 50% of staff from the communities is no longer in force, but it is still a significant factor for all jobs, and a near-requirement for some (e.g., my team). WhatamIdoing (talk) 20:11, 15 March 2022 (UTC)

It may well be the case that the perception that WMF staff don't edit just won't die, but then neither will the perception that WMF staff keep coming up with stuff that is at odds with en-wiki community norms. I don't think that the answer is that en-wiki is full of editors who are cranks, or that it's only the cranks who criticize WMF. It seems to me that, even when WMF staff are engaged with content editing, they become constrained by (1) a corporate sense of wanting the editing community to appear like an inclusive HRM paradise, and (2) a corporate sense that it's all-important that the numbers of editors and edits must always be increasing. As much as inclusiveness is indeed a good thing, this mindset leads to blindspots about the things that experienced editors have come to learn from experience. @WhatamIdoing: do you disagree with what I said here? --Tryptofish (talk) 21:44, 15 March 2022 (UTC)

So many thoughts, but let me start with one: I wonder what causes you to think that the WMF thinks the number of editors and edits needs to always increase. What communications from the WMF are you seeing that gives you that impression? WhatamIdoing (talk) 03:07, 16 March 2022 (UTC)

Highlights include increasing monthly active editors by more than 18 percent since 2016, which continues to grow", "Grow Wikimedia's communities", "a growing Wikimedia movement", "growth has been fueled by a global volunteer force and donors who explore and visit the site regularly"… ‑ Iridescent 04:54, 16 March 2022 (UTC)

^That. --Tryptofish (talk) 17:21, 16 March 2022 (UTC)

@WhatamIdoing, as Jayen466 has just indirectly pointed out a few threads up, the WMF has literally just rolled out a massive spam campaign with the strapline "If you donate as little as [insert local currency] today, Wikipedia will continue to grow for years to come". ‑ Iridescent 20:58, 16 March 2022 (UTC)

"Wikipedia" can grow without the number of edits or editors growing. The other sources talk about different kinds of growth, but aren't talking about the WMF having a goal of increasing the number of editors. I didn't note any of them talking about an increase in the number of edits, but maybe I missed that. Some of them (e.g., train-the-trainers programs) seem to be talking primarily about growing the non-editing parts of the movement.

Also: Spot the enwiki admin among those blog posts' authors. WhatamIdoing (talk) 01:58, 18 March 2022 (UTC)

Firstly, the WMF is exponentially growing its own staff numbers. Salary costs are now ten times what they were a decade ago (partly also because C-level salaries are now twice what they were ten years ago). This whole effort is financed by telling the public, year after year, that money is needed to "keep Wikipedia online", which is risible.

Secondly, it is surely without question that the WMF is trying to grow volunteer communities in the Global South, and is investing heavily in this. See, e.g., the budget increases related to "thriving movement" here. Like Abstract Wikipedia (which is teetering on the edge of a CC0 licence to keep Big Tech happy, contributor rights be damned) the ultimate aim of this effort is in part to ensure that the likes of Google, Apple, Amazon and Microsoft have free ("free" as in they don't have to pay for it) content in Asian and African languages to feed their infobox panels and voice assistants with to conquer markets (representing tens of millions of speakers in some cases) that are currently still closed to them because of language barriers – all on the basis of unpaid volunteer labour. (An added kicker is that those companies are by and large tax dodgers – they don't like paying tax in the countries they operate in – so helping them gain control of markets in the Global South actually deprives those countries of tax income needed to fund education and health systems.)

Thirdly, official WMF strategy is that "By 2030, Wikimedia will become the essential infrastructure of the ecosystem of free knowledge, and anyone who shares our vision will be able to join us." Growing both edits and editors is implicit in that. Andreas JN466 11:47, 18 March 2022 (UTC)

Fundraising & expenditure

Did you know that the WMF claims in fundraising emails that 32% of the money people give "will be used to support the volunteers who share their knowledge with you for free every day"?

For reference, total revenue last year (excluding gifts to the Endowment) was $163 million, and after the first two quarters of the present 2021/2022 financial year, the Foundation was already $10 million up on last year. This means the WMF is on course for at least $175 million this year.

32% of this would be $56 million. The total "thriving movement" budget this year, which includes chapter funding (for hundreds of paid staff), is $36.7M, up from $14.3M in FY20-21, so I can't make sense of that 32% figure. I've asked on Meta what they mean.

The second email people get now offers them badges. The third email acknowledges at least that the money is crucial "for our organsiation and movement to grow" – which is honest – but by then people will already have read in these emails that they donate "to keep Wikipedia online for yourself and millions of people around the world", that the WMF "choose not to charge a subscription fee" (duh, their mission statement commits them to not charging one), and that the Foundation needs money "to ensure that Wikipedia remains independent, ad-free, and growing for years to come", and "to keep Wikipedia free and independent." These are all wordings that were deprecated and discontinued on the banners years ago, following vociferous community criticism. Andreas JN466 10:08, 21 May 2022 (UTC)

I assume it depends on what they mean by "support the volunteers". I'd wait until they reply before giving my opinion, as there may be a perfectly innocent explanation such as the money raised in India being spent in India owing to currency transfer fees and the WMF spending disproportionately on chapters and education programs in India. ‑ Iridescent 13:04, 21 May 2022 (UTC)

I was wondering the same. But then I checked the Swedish version of the email linked on the Meta Fundraising page and found it says the same. In both cases, that phrase is illustrated with a picture that seems to have been taken in an African primary school. Andreas JN466 13:26, 21 May 2022 (UTC)

It's been nearly two weeks, and no reply has been forthcoming. Meanwhile, an interview with Raju Narisetti appeared in the Indian Express over the weekend: "We enabled people to edit Wikipedia on the phone, which is a big breakthrough in a country like India, and so that has contributed to the significant growth in languages,” he said, adding this is also the answer to why Wikipedia raises money."

He continued: “More than 75% of the money we raise globally goes to two things. One is to give money back to the volunteer community so they can launch a new language. Two is about half of it goes to the infrastructure. You need to have databases and put it on the cloud and make sure it’s reliable,” he said. Although a lot of the money is raised in the more developed Western markets, most of it is actually flowing into the global south, where the growth will come in languages and users.”

Is there a fact checker in the house? Andreas JN466 08:09, 1 June 2022 (UTC)

I'm willing to believe that "back to the volunteer community" does not mean en-wiki, but rather the languages that need help with more basic necessities. --Tryptofish (talk) 20:36, 1 June 2022 (UTC)

@Tryptofish: First off, have a look at [1]. This shows how much money actually flows into the various regions (outside the US). Is it true that "most of it is actually flowing into the global south"? Andreas JN466 23:17, 1 June 2022 (UTC)

I don't know enough (or anything) about accounting terminology, and the answer is dependent on what they mean by "Total expenditures for and investments in the region". If it includes more general costs rather than purely grant expenditure then it's unsurprising to see such high spending in Europe; the esams server is in the Netherlands, plus presumably a significant proportion of the tech companies with which something like the WMF would do business will be based in Europe. Presumably also this form only accounts for funds disseminated directly by the WMF, so wouldn't show the figures for those countries where donated funds go direct to the local chapter rather than being laundered through SF first.

I don't doubt that the WMF is lying here—when it comes to where the money comes from, where it goes, and who is taking a cut along the way, it would be more unusual to find them being honest—but they're probably not lying quite as much as this makes it appear. ‑ Iridescent 02:18, 2 June 2022 (UTC)

Okay. Even if you limit yourself to the Grantmaking amounts (pages 30 and 31), you end up with a total of a little over $2M going to the Global South, or less than 60% of the $3.5M grant total. $2M is 1.25% of the $163M revenue.

Most striking of all, given that he is talking to an Indian audience: the amount going to South Asia as a whole was less than $80,000 ($3,339 + $75,198). That is 0.05% of the money raised. Andreas JN466 09:06, 2 June 2022 (UTC)

I've expanded on this on the mailing list: [2] Comments and corrections welcome. Andreas JN466 16:31, 2 June 2022 (UTC)

I'd say wait until they reply. I have minimal AGF when it comes to the WMF, but I can think of potential explanations here, such as funds being raised in India going direct to local chapters rather than through WMF central funds (there used to be a similar arrangement for WMUK, so it's not without precedent). ‑ Iridescent 19:25, 2 June 2022 (UTC)

(adding) Ping to User:Whatamidoing (WMF)—WAID I know this particular topic isn't in your remit, but can you nudge whoever's responsible to reply. If there's a legitimate answer, the longer they go without providing it the more the conspiracy theories will grow; if there's not a legitimate answer, the longer they go without retracting the claim the more chance it causes some nasty reputational damage. (For the record, I have absolutely no problem if the majority of expenditure is in the US and Europe—it's exactly what I'd expect since most of the staff and facilities are in the US and Europe—but Although a lot of the money is raised in the more developed Western markets, most of it is actually flowing into the global south is the kind of claim that really pisses off donors if it's not true.) ‑ Iridescent 19:40, 2 June 2022 (UTC)

Memories are short, Whatamidoing (WMF). At around the time of the ill fated IEP a few years ago (another mega WMF cockup) seven figures of funds were squandered in India and our en.Wiki community volunteered to clean the huge content mess left in its wake. What the goody-two-shoes rudderless WMF in their laid back Californian comfort and ideology tend to forget is that corruption is part of the every-day culture here in south and southeast Asia - to the extent that the national economies even depend on it! Comments today: Right now, the W?F keeps spending more and more every year to do essentially the same job they were doing ten years ago with far less spending and ...the WMF is run by incompetents incapable of keeping expenditures in line with income and capital, rather than professionals in nonprofit administration with extensive experience in just that balancing act, are a perfectly apt descriptions. I have been working as an unpaid volunteer consultant for 'do good' NGOs on and off here in Asia, especially in India, Laos, and Cambodia, for longer than Wikipedia exists and their management/executive cohorts nearly all have the same things in common: enjoying high salaries and perks, little effective output during office hours, grossly disproportionate expenditure on their own infrastructure and convenience, paying external agencies for the work they are too lazy to do themselves, squandering the rest of their easy-come;easy-go donations, and a palpable disregard for the expendable community of the hundreds (or thousands) of volunteers who do the actual work for no compensation of any kind whatsoever (well,maybe a $1 T-shirt with a WMF logo if they are lucky). Kudpung กุดผึ้ง (talk) 06:25, 3 June 2022 (UTC)

Your comment reminds me of the fact that while Raju mentions 65,000 Indian volunteers in his interview, the WMF Form 990 says the 2020 South Asia "Grants and Other Assistance to Individuals" total of $75,198 went to 22 recipients (an average of $3,418 per head). Andreas JN466 15:55, 3 June 2022 (UTC)

I don't know where it went, but is that not likely to be assorted costs for the (covid-abandoned) Wikimania 2020? Or are they using some archaic definition of "South Asia" as "former British colonies in Asia"? ‑ Iridescent 17:59, 3 June 2022 (UTC)

They include Burma and Cambodia in "East Asia and the Pacific"; given that Thailand is between Burma and Cambodia, Thailand must be included in East Asia as well. Andreas JN466 20:20, 3 June 2022 (UTC)

The Indian Express article has now been updated, and a Disclaimer added. [3]

The topic of Indian funds always having to go to the WMF first came up on the mailing list recently. So there is no arrangement in India like there is in Germany (and like there used to be in the UK). Best, Andreas JN466 15:19, 3 June 2022 (UTC)

For the benefit of TPWs who aren't following on Meta

For the benefit of TPWs who are following this thread but don't get involved in the cesspit of Meta, the WMF did reply to this one. As per my previous comments, on this specific topic I think the WMF's position is perfectly reasonable; because of the physical fact of the hardware being in Virginia, Texas, Singapore and the Netherlands and the WMF's offices being in California, then by definition a disproportionate percentage of funds allocated to any program is going to be spent in those five areas, even if the area of benefit is somewhere completely different. (I personally think that a lot of these programs are mistargeted, and Africa, India etc would be better served by our concentrating on getting the existing projects up to as good a standard as possible and concentrating on ensuring people everywhere can access them easily and cheaply, rather than the current WMF obsession with missionary activity and trying to recruit as many people as possible to the cause. However, that's a separate issue.) ‑ Iridescent 08:30, 17 June 2022 (UTC)

"then by definition a disproportionate percentage of funds allocated to any program is going to be spent in those five areas, even if the area of benefit is somewhere completely different." This concept is remarkably difficult to get across to some people, even those who should know better. Where it intersects with the concept of 'social value' and commissioned services has been giving me a headache at work recently. (eg, a funder from a specific municipal area queries our spending a portion of said funds on infrastructure physically located elsewhere which supports services in their local area - we have to provide a social value breakdown of all our spending. No you numpty, our network support does not provide social value, but you still have to pay towards it because the people it supports in your area require it!) Only in death does duty end (talk) 09:43, 17 June 2022 (UTC)

Try explaining any government or government-funded job to the public. A ridiculously high number of people—including a depressing number of people important enough to be making actual decisions—seem to genuinely not grasp either that "frontline staff" and "essential staff" aren't synonyms and that admin, tech support, cleaning, quality checking, logistics, marketing, maintenance etc aren't optional extras, nor that it's usually more efficient to have a single central location for the back office functions rather than having a separate phone switchboard, warehouse, garage et al in every single location. (I imagine every local councillor in the world is sick of hearing the question "but why doesn't our small town operate its own [buses/specialist hospital/garbage collections/police force/plastic recycling plant]?".)  ‑ Iridescent 10:46, 17 June 2022 (UTC)

But... but... I thought the goal of all funding was jobs for the boys? And since I'm a local funder, it had better be local people getting my funds.

Only, perhaps you need to ask people to estimate the "economic, social and environmental well-being" they receive from that contract when the network's down, instead of when they don't notice its existence because everything's working. WhatamIdoing (talk) 17:04, 22 June 2022 (UTC)

You don't need me to tell you this as you already know it, but Jayen466 is completely correct in one sense. The WMF may be careful not to technically lie, but they don't exactly go out of their way to dispel the perception of "if I give them money, it will be spent locally and thus not only benefit my community through improved access to Wikipedia, but more directly in terms of creating jobs and improving infrastructure locally". It may be the case that Although a lot of the money is raised in the more developed Western markets, a lot of it is actually flowing into the global south is nominally correct—$2 million, is indeed a lot of money—but it's actively misleading since it gives the impression that it constitutes a significant proportion of total spend. (What's particularly irritating is that there's no need for the WMF to equivocate here and they're just doing it out of habit. "Although a lot of the money is raised in the more developed Western markets, much of it is spent on improving services for the global south" wouldn't put off any potential donors and would actually be accurate.) ‑ Iridescent 17:23, 22 June 2022 (UTC)

Thanks. I've expanded on this theme in the Signpost: "Wikipedia's independence" or "Wikimedia's pile of dosh"?. I'd be particularly grateful for people's views and responses on the Signpost article's talk page, where I've just posted this, concerning the claim in the Annual Report that 31% of WMF spending is for "Direct support to communities":

Here are the audited financial statements for 2020/2021: [4]. They show the following data:

Revenue: $162,886,686

Expenses: $111,839,819, of which:

• Salaries and wages: $67,857,676 (this is for US and non-US employees only; it doesn't include the pay of WMF staff who work as contractors)
• Awards and grants: $9,810,844 (of which $5 million were a grant to the Wikimedia Endowment, see page 14)
• Internet hosting: $2,384,439
• In-kind service expenses: $473,709
• Donation processing expenses: $6,386,483
• Professional service expenses: $12,084,019
• Other operating expenses: $10,383,125
• Travel and conferences: $29,214
• Depreciations and amortization: $2,430,310

According to the Annual Report, 31% of spending is for "Direct support to communities", which is defined as follows: Wikimedia projects have global reach. This is enabled by the diverse contributions of volunteers from local communities around the world. We provide grants and other resources to support local contributors, community outreach events, and advocacy for growing free knowledge.

31% of expenditure is $34,670,344.

There clearly isn't any such item in the above list of expenses. Just the salaries, the $5 million gift to the Endowment and donation processing together account for $79,244,159.

That leaves just $32,595,660 for everything else.

So that means that some of the WMF salaries must be counted as "direct support to communities" – over $2 million, even if you classify all other expenses as "Direct support to communities".

Just tell me – ideally over there – do you think the Annual Report leaves its readers with an accurate impression of WMF spending?

I mean, if we look at the statement "We provide grants and other resources to support local contributors, community outreach events, and advocacy for growing free knowledge", what percentage of this $35 million do we think is "grants", "other resources", "community outreach events" and "advocacy", and what would the average reader think? Andreas JN466 08:13, 29 June 2022 (UTC)

Placeholder note that I've seen this—I'll reply either here or on the Signpost page as appropriate once I've had time to read it fully. ‑ Iridescent 04:33, 2 July 2022 (UTC)

Belatedly replied there, although I've not really much to add. For the benefit of anyone reading this thread ibn a future archive, my comment was:

It was suggested I comment here, but I haven't really anything to add other than my comments in this long talk thread. The TL;DR summary of my view is that I think Jayen466 is slightly missing the point in this case; the problem isn't so much the WMF's fundraising per se, but that they've developed such a culture of instinctive evasiveness that they're misleading and obfuscating even when there's no need to. (No donor, supporter or sponsor would reasonably take issue with "owing to where our employees are located your donations won't necessarily be spent in a particular region, but it will still be for the benefit of that region"; the WMF are only lying because it's become their default communications mode.)

I do think that's the real point here. The exact proportion of funds received from various areas vs funds disseminated in various areas isn't particularly important given that it's natural that funds will disproportionately be spent in areas where WMF hardware and employees are located; the issue is that lying and dishonesty has become so ingrained in WMF culture, that they're lying even when there's nothing to gain from lying. ‑ Iridescent 04:02, 11 July 2022 (UTC)

They're not lying as such. Quoting what Simon McDonald said the other day, "I think that the language is ambiguous, it's sort of telling the truth and crossing your fingers at the same time and hoping that people are not too forensic in their subsequent questioning ..." – except that in their case, it is working.

I agree that spending patterns and the fundraising messages are two different issues.

There is a compelling argument to be made that more money should be spent in the developing world. Many things could be done more cheaply there and in the process provide life-changing opportunities. To be fair, the WMF itself has said it wants to increase such spending, and this is one thing I'd like to see included in the Election Compass.

The other issue is when the WMF makes it sound as if a major part of its spending were already flowing to the developing world. That's not okay, irrespective of what one may feel about geographical spending priorities.

I also agree that the WMF could fundraise pretty well without scaring people with the spectre of a subscription fee, or the prospect of Wikipedia blinking out of existence. But given their commitment to A/B testing I'm quite sure that every word is weighed. I believe they would take significantly less money if they used "honest" banners – they've trialled some and then not used them.

You say the WMF has an ingrained culture of lying and dishonesty – do you think that affects communications with volunteers as much as external communications with the public? What if anything could or should the community do about this culture, in your opinion? Andreas JN466 08:14, 12 July 2022 (UTC)

As I understand it, the main constraint against spending more money in the developing world is one I think they're actually correct on. The WMF for good reason is reluctant to have operations in any country where there's any possibility that editorial independence will be compromised, which severely restricts where the core operations can take place. My issue here is that nobody would have a problem if they just said "most of our operations need to be in the US"—even the proudest patriot in India, Nigeria etc would concede that the US is the only place currently with the right combination of technological infrastructure and free-speech laws for Wikipedia to operate—but with the deliberately misleading messaging.

Regarding do you think that affects communications with volunteers as much as external communications with the public?, I'm obviously jaded in this given my experience during Framageddon. My best guess is that for the most part, it's not so much a culture of dishonesty per se, but that WMF staff and the typical editor are starting to have radically differing views of the roles both of Wikipedia and of the WMF. When I talk to WMF people lately, I start to get the feel that they see it almost as a religious movement in which any activity that advances the goals is by definition doing good no matter how bad it appears to outside observers; when viewed from that "if it can be said to advance the goals then it can't be wrong" perspective, a lot of the weirder WMF statements and decisions make perfect sense. This isn't a particularly new observation—some of my earliest experiences of Wikipedia infighting are of Kelly Martin complaining that Wikipedia was becoming cult-like—but it's definitely a tendency that seems to be accelerating recently. (It also, to me, explains some of the more apparently eccentric overreactions. If you're doing God's work, then those who disagree with you aren't just in disagreement but are actively evil; as such going out of they way not just to disagree with but to actively humiliate those not of the body, be it Fram, Eric Corbett, or the monkey selfie guy isn't being obnoxious, it's a testament of faith.)

As to what we could do about it, I honestly don't know. Over the long term, I assume the missionary zeal will burn itself out and the WMF will just become another part of Big Tech that happens to have the quirk of being owned by a charitable foundation rather than shareholders. My worry is that when that happens, the existing model will be unsustainable; while too much faith is a bad thing, we rely on it to some degree to keep people volunteering. A more professional WMF would likely have less issues recruiting admins as people love the chance to tell others what to do, but I suspect would flare out and fall apart fairly quickly. (If you have a spare half hour, it's worth reading this very long thread in full for my thoughts on this.) If frorced to make a long term prediction, it would be that eventually we'll end up with a content fork—it bears repeating that Google could send Wikipedia the way of Myspace tomorrow with the addition of one line of code to PageRank—but for that to happen we'd need the backing of someone with both technical infrastructure and deep pockets, and none of those seem very appealing. For all the WMF's faults, I imagine I speak for every editor in saying I'd take Jimmy Wales over Elon Musk. ‑ Iridescent 09:00, 12 July 2022 (UTC)

Over half of all WMF expenses are US salaries – even though a large proportion of the US staff work remotely, doing work that could really be done anywhere in the world.

On the page you linked, User:力 mentioned (in 2018) that "fake references ... will not be detectable by AI anytime soon". This reminded me that some smart people at Meta (Zuckerberg's) earlier this week announced an AI tool that does precisely that – see [5] and [6].

One thing volunteers could do about the culture in my opinion is point it out (off Wikipedia) when the WMF uses the Boris style of truth in its public communications. Getting away with it does them no good.

I smirked about doing "God's work" ... there is definitely an element of that. :) Cheers, Andreas JN466 18:45, 15 July 2022 (UTC)

The WMF has posted mock-ups of the Jimbo mails it plans to use in the upcoming English email campaign (Email 1, Email 2, Email 3). You can now get badges if you donate a few more dollars to add to the WMF's giant stash of cash, to "keep Wikipedia online". Andreas JN466 11:52, 10 August 2022 (UTC)

Votes cast

(ec) I am currently looking at the vote log and it says 903 votes cast. That's a small portion of all Wikimedia users, I think, but I am not sure how many of the "active" participants. Looks like I am #160 and there are a lot of regulars on enwiki there. Jo-Jo Eumerus (talk) 16:30, 14 March 2022 (UTC)

We have until March 21 to vote, right? The WMF has so lost their way (if they had ever found it); the level of group think and impenetrable writing is still mind-boggling (and for as long as their writing has been pure gibberish, one would think they would have done something about it by now). Easy to see why a sensible admin like Ealdgyth would turn in the bits of this sort of thing. SandyGeorgia (Talk) 17:13, 14 March 2022 (UTC)

This type of bug on a secure voting site really inspires confidence: phabricator:T303735. --MZMcBride (talk) 21:24, 14 March 2022 (UTC)

In case anyone wants to validate Iridescent's theory that the WMF payroll [will] block-vote so it's almost certain to pass even if there's overwhelming opposition from the actual editor communities, there have been a total of 28 votes by (WMF) accounts so far, including several by people I've never heard of before even though I'm very into obscure WMF geekery. And only two of them (SOyeyele (WMF)/Jamie Tubers, Zuz (WMF)/Celestinesucess) meet the first set of voting requirements considering only their volunteer account. * Pppery * _{it has begun...} 01:28, 15 March 2022 (UTC)

The Universal Code of Conduct is a bit of a Curate's egg. It certainly does have some good parts, and some of the sillier features were fixed during its long gestation. But at heart it reads like a blurb from a company HR department for how you treat colleagues after the recruitment process has filtered some people out. Take two examples where I have tried and failed to influence the process, age and linguistic ability. In the past I have supported RFAs of teenagers who I thought were ready for adminship, my tests for such candidate's include questions such as "Do your parents or guardians trust you with an internet account that you can't tell them the password of". But I'm aware that there are others who don't think that anyone should be an admin until they are legally adult. I suspect that the WMF and all the chapters will quickly exempt themselves from this one, leaving us with the practical issues that come up when children of surprisingly young ages start editing. No age discrimination is a very different thing when your HR department is only recruiting university graduates. Language fluency is another area, these days I spend as much time on commons as here, and of course Commons is one of our multilingual projects. Wikipedia however has 300 or so seperate wikis based on different languages, it makes absolute sense that we require some fluency in English for most roles here and in Scots for people who write the Scots Wikipedia. As a monolingual Brit who has worked for a chapter and might apply for some WMF jobs, it would be in my interest if the WMF weren't able to prefer multilingual candidates when recruiting. But as a Wikipedian I absolutely see that it would be an advantage for those WMF staff who interact with the community to have at least a working proficiency in more than one language. A rule against discrimination on language fluency in our multilingual projects would make sense, as would a ban on age discrimination among community members who are legally adult. But the code still contains "This applies to all contributors and participants in their interaction with all contributors and participants, without expectations based on age, mental or physical disabilities, physical appearance, national, religious, ethnic and cultural background, caste, social class, language fluency, sexual orientation, gender identity, sex or career field." As usual with the WMF one comes away from any interactions thinking that for a twenty year old organisation with only a few hundred employees, it has somehow acquired the inflexibility and ossification of an organisation several orders of magnitude larger and older. ϢereSpielChequers 11:05, 15 March 2022 (UTC)

• I learn something every day. I had never heard the phrase "curate's egg." I have no interest in Wikipolitics but I would like to do my bit for the survival of Wikipedia. So, O Wise Curator of Eggs and Others of Esteemed Opinion, should I vote yes or no on the enforcement guidelines of the UCoC? I will likely take your advice.Smallchief (talk)
  • I have voted No, but then used the comments box to give two or three examples of where I disagree with the current proposal. My hope is that this leads to the proposal being further improved. I would suggest that you only vote if you have the time to read the code, if you then vote "No" please give some pointers as to where you think the code needs improvement. Feel free to echo mine if having read the code you agree that each language version of Wikipedia would benefit by continuing to be allowed to require some fluency in that wikipedia's language for at least some roles in the pedia. ϢereSpielChequers 15:37, 15 March 2022 (UTC)
    • I'd encourage anyone who isn't sure both that they want this and that both UCoC and the enforcement guidelines in this form are what they want, to oppose. This is a vote on a change that's effectively going to be irrevocable, since any change to it will itself require a consultation exercise across the whole of the WMF ecosystem. (Ever tried to get the wording to WP:ARBPOL amended? Multiply that by the 325 languages in which we currently operate.)
      If you're not certain that both that UCoC is something we need and that this particular wording is the wording we need, the only rational choice is to tell them to reconsider it, preferably with an explanation of exactly which aspects you consider potentially problematic but if not, a simple "no" is fine. Wikipedia isn't going to fall apart if we go a few more months without something we've got along without for 21 years, but Wikipedia potentially is going to fall apart, or at least become unrecognizable, if we're plunged into a permanent civil war of wikilawyering over whether any given comment potentially violates something or other. ‑ Iridescent 18:40, 15 March 2022 (UTC)
      • Somehow I doubt my oppose of 'this policy is both structurally and procedurely racist in its wording and intended implementation' is going to make much difference to the WMF staff there. Anyone who can with a straight face support the 'do not recognise' section (which isnt even the most problematic of the document as a whole) despite it being blatantly discriminatory and outright illegal in various countries is already way past the point of rational argument being a useful approach. The only options left realistically are to amend WP:ADMIN to explicitly forbid enforcement of UCoC provisions (where they are not supported by ENWP's existing policies) with removal of tools if not followed, and to forbid advanced tools (admin/crat) from WMF staff members/contractors due to the conflict of interest. Only in death does duty end (talk) 18:53, 15 March 2022 (UTC)
        • The WMF have already anticipated that one. All admins and arbitrators (and their equivalents on other projects) are going to be required to affirm (through signed declaration or other format to be decided) they will acknowledge and adhere to the Universal Code of Conduct. This is why the talk of mass resignations isn't just hyperbole; if this passes, we're putting people in a position where it's literally the case that the only honorable course is resignation. (I suppose I—or anyone else—could point-blank say that I'm not going to adhere to UCoC and challenge Arbcom to come and take it, but I doubt Arbcom would thank us for dumping an existential crisis onto their collective doorstep.) ‑ Iridescent 19:24, 15 March 2022 (UTC)
          • I've considered that option (not doing it and seeing what happens) also. We'll see. It would get interesting if a group of admins with a lot of social capital on en-wiki all took the "we aren't going to affirm, so come and take it from us, ArbCom" attitude. @Dennis Brown, Cullen328, and Hog Farm:... Ealdgyth (talk) 19:33, 15 March 2022 (UTC)
            • Well, you've definitely nailed the camp I'm in although I think you overestimate my ability to persuade others. I've discovered that after 57 years, two divorces, and 16 years of this place, I'm not as soft spoken as I once was. I absolutely will resign before "affirming" my allegiance to the WMF (cue USSR style music). Or it may be better to let them take it from me, although 100 admin piled up in Arb might be one hell of a backlog. I joined a movement, a project, I didn't sign on to be slave labor for a corporation. And of course, I will vote in every RFA, and if there is any concern about the candidate that is legitimately troublesome, I would politely oppose with a reasonable rationale, because it would be wrong to stuff the ranks with unqualified candidates. Don't underestimate that portion of our combined strength. Dennis Brown - 2¢ 20:11, 15 March 2022 (UTC)
              • It will come as no surprise that I'm not going to take a loyalty oath. Nor an indoctrination program. Unfortunately I think I used up my social capital during Framgate. Or, maybe that gave me more?! I don't really know how these things work. Surely threatening to resign loses it's power the 3rd or 4th time I try to use it. If nothing else, I'd be the only person to have ever been desysopped twice by WMF (I hold out some hope that the en.wiki arbcom would never desysop for this themselves).--Floquenbeam (talk) 20:17, 15 March 2022 (UTC)
                • We need box better than the lame box I'm using on my talk and user page. And maybe matching shirts. Dennis Brown - 2¢ 21:34, 15 March 2022 (UTC)
                  • I refused to sign the confidentiality agreement required to become an Arbitration electoral commission person, although I think it was more that I wanted to thoroughly understand and appreciate what I was expected to adhere to, and just couldn't do it. And I can pinpoint several fuck-ups in my life that I can trace back to signing things thinking they would be a good idea but turned out not to be. So you can imagine what my response will be to sign anything - procrastinate and hope it goes away. I'm mindful to oppose, but like WSC, I would like to put some constructive reasons why first. Ritchie333 ^{(talk) (cont)} 11:54, 16 March 2022 (UTC)
              • Ealdgyth, I think you're probably overestimating the amount of social capital I have here, or even the percentage of editors who would recognize my username. But yes, I will NOT be swearing fealty to the WMF, nor undergoing the "training" program. Like Dennis, I'm not sure if it's best to just resign and not fight it or to make the WMF and/or ARBCOM take it. Would likely depend on how much energy I have at the time, which is something that I've been in short supply of the last several weeks ... I would hope that conditions would still be good enough for me to stick around even after the mess, but if the WMF screws stuff up too much, I guess there's always Missouri Historical Review. Hog Farm _Talk 23:43, 15 March 2022 (UTC)
          • Ah you misunderstand me Iri, changing WP:ADMIN is specifically to counter that. Its largely irrelevant if an admin affirms their allegiance or not if an ENWP community policy explicitly denies admins the authority to take an action. At that point the WMF will have to overtly take control of ENWP. Essentially its putting policy in place to prevent anyone other than WMF Staff accounts from enforcing UCoC (specific) violations. The end result of such a policy is this: Admins take their oath of bullshittedness and sit through the re-education camps and then either: never actually enforce the UCoC and so the WMF de-tools them, or they enforce it and ENWP de-tools them, or they dont enforce it and nothing happens because the WMF are not actually interested in hiring the staff and paying them to do all the jobs advanced tool using volunteers do because it would upset their gravy train. The alternative is actually a lot worse, in that all the admins of character (eg, everyone who refuses to kowtow) leave, and we are left with a bunch of WMF yes men. What happened to Fram will seem like a happy memory once the ideological purges start with checks in place. Dont get me wrong, I dont think in the long run it will stop the WMF from doing whatever it wants to do, it will just force it out into the open a lot sooner. RE Dennis and any other admins thinking of resigning. Dont do it, force the WMF to de-tool you. Why make it easier for them? You gain nothing and they get an easy ride. Only in death does duty end (talk) 21:37, 15 March 2022 (UTC)
            • You misunderstand: I'm not pledging loyalty to anyone, least of all the Foundation. Even if they offer free puppies. The place is full of power hungry asshats with too much of other people's money and no understanding of the community. I won't sign as a matter of principle. I will walk away. Dennis Brown - 2¢ 21:58, 15 March 2022 (UTC)
              • I voted “no” on the basis that the text is so longwinded that I was exhausted by it before finding anything of substance. I also haven’t been able to find a page to offer feedback. SmokeyJoe (talk) 22:07, 15 March 2022 (UTC)
                • I'm pondering how they will define holders of advanced permissions. I think I might be OK with requiring checkusers and oversighters to sign the affirmation. We're assuming that administrators will also be included, but would it necessarily stop there? Template editors? Autopatrolled? Rollback? There's an awful lot of room for a lack of common sense in determining who would have to swear fealty. --Tryptofish (talk) 22:43, 15 March 2022 (UTC)
                  • Since the oath is to uphold and enforce the UCoC, one would think they mean admin bit. But I've learned to never underestimate the silliness that comes from the Foundation, so who knows. Dennis Brown - 2¢ 10:27, 16 March 2022 (UTC)
                    • I agree entirely. I've been thinking about this some more, and I'm toying with proposing something in the event this thing passes. We could have an RfC at Village Pump Proposals, in which the community could choose to define the meaning of "advanced permissions" at this project for these purposes. And the criterion could be, rather reasonably, permissions that require identification to the WMF. That would mean checkusers, oversighters, and arbitrators (the last because they typically get the other two permissions). It would thus exclude administrators, which I think would be entirely a good thing. I find the reasons that admins would resign to be honorable and appropriate, but I'd much prefer that they wouldn't have to do it. Of course the WMF might object to the community deciding this, as it wasn't what they would have wanted us to do, but I like the idea of putting them in the position of saying no to a community consensus. --Tryptofish (talk) 17:20, 16 March 2022 (UTC)

                      As far as I'm aware, there are no longer any permissions that require identification to the WMF other than (for obvious practical reasons) those that create an actual contractual relationship. It was always a stupid requirement—I could knock up an ID document in about ten minutes in any name I like that would be adequate to fool the WMF (who obviously don't have access to the biometric data etc that genuine ID documents contain). ‑ Iridescent 20:08, 16 March 2022 (UTC)

                      Doggone it. We'll just have to say CHU, OS, and ARB. --Tryptofish (talk) 20:21, 16 March 2022 (UTC)

                    • I'm reading the proposal now, in preparation to vote, and I see that it says: "Designating functionaries will be done, whenever possible, by local communities, following the principle of subsidiarity that online and offline communities across the world should make decisions for themselves whenever possible." That would actually make it very practical for us to make such a consensus stick. --Tryptofish (talk) 17:32, 16 March 2022 (UTC)

So far this discussion seems to have come with no reasons at all to support the enforcement guidelines. The UCoC itself seems mostly harmless, and certainly includes things I support, but given that our informal standards of conduct here aren't that far away from the UCoC, what arguments are being made in support of having enforcement for it? What bad things happen if we don't have agreed enforcement guidelines; if we continue to expect each wiki to manage these issues themselves? The only argument I can recall is that small wikis need some such mechanism. I can see that -- there have certainly been some horror stories about small wikis -- but the larger wikis are quite different kettles of fish. Are there legal pressures on the WMF to have enforcement? I could see voting for something that applied to small wikis (maybe less than 100 active editors/month) but not for en-wiki. I have little interest or involvement in WMF politics, and I usually find myself silently disagreeing with the vocal majority on WMF issues (e.g. Framgate) but this seems misguided and I plan to oppose it. Mike Christie (talk - contribs - library) 22:08, 15 March 2022 (UTC)

I can't really argue that the UCoC enforcement guidelines are the correct solution, but a review/adjustment of conduct enforcement seems valuable in theory. Surely at least some people here agree that conduct dispute resolution on enwiki can be dysfunctional? For example: I don't think requiring people facing actual harassment, especially non-experienced editors who might even be using real names (as in some past cases), to show up at ANI for widely-attended public discussion is an effective way to deal with those problems. I can't imagine any company having a 'reporting harassment in the workplace' policy that functions analogously to enwiki's ANI route - it would be widely denounced. So I like the ideas in meta:Universal_Code_of_Conduct/Enforcement_guidelines#Recommendations_for_the_reporting_and_processing_tool and the technical tools suggested there.

While there aren't any reasons to support expressed here, I don't see any real disadvantages either. The main point of contention above seems to be that the UCoC requires admins to affirm their 'acknowledgement' of, and 'adherance' to, the UCoC. I don't know why that clause was inserted, but I don't see why it's a big deal either. As a practical matter, AFAIK almost everything–if not everything–disallowed by the UCoC is already disallowed by local enwiki policy and practices, either as-written or as-applied, and the enforcement guidelines leave most enforcement to local admins. So personally I'm curious what exactly people feel is bad about it? ProcrastinatingReader (talk) 22:52, 15 March 2022 (UTC)

I want to question your characterization of ANI as the place we want reporting of harassment to be conducted. (Although ANI is certainly an easy target for criticism.) We have ways of privately contacting ArbCom, for example. I don't dispute that inexperienced users can be unaware of the best options available to them, but there is no reason to assume that they would be any more aware of the WMF. --Tryptofish (talk) 23:03, 15 March 2022 (UTC)

I will also take a stab at answering the question that you asked. Not too long ago, there was a widely noticed example of a problem where the UCoC might have been substituted for community practice, if enforcement were in effect: [7]. On the one hand, there was a class assignment, bringing in new student editors. And there was an instructor who was very interested in a topic that can reasonably be described as related to inclusiveness. And there was even a vivid case of a young editor facing considerable off-site harassment. But on the other hand, there were some minor concerns about canvassing, and major concerns about our content rules about notability. The community did not do a perfect job of dealing with it, but I think anyone who felt compelled to be guided exclusively by the UCoC would have been inclined to violate our own policies and guidelines. --Tryptofish (talk) 23:15, 15 March 2022 (UTC)

Highly visible and public reporting of harassment is not necessarily comfortable for victims.but there is no reason to assume that they would be any more aware of the WMF True but the WMF isn't doing enforcement. I understood the plan being a prominent/easy-to-use reporting tool that forwards reports to the appropriate place, though I don't think the details are ironed out yet.

but I think anyone who felt compelled to be guided exclusively by the UCoC would have been inclined to violate our own policies and guidelines how so? Specifically, how do you think the issue would've been dealt assuming the UCoC enforcement guidelines had been ratified at the time? ProcrastinatingReader (talk) 23:24, 15 March 2022 (UTC)

Maybe we edit-conflicted, but I answer that just below. --Tryptofish (talk) 23:43, 15 March 2022 (UTC)

We did EC, however, why do you think the UCoC mandates the deletion (or non-deletion) of an article? ProcrastinatingReader (talk) 00:08, 16 March 2022 (UTC)

I'm about to logout, and the indenting here is getting strange, but what I mean is that the content was related to inclusiveness, and deletion could be construed as insensitive to inclusion. If we're going to be generally accepting of women and minorities (which of course we should), but as a matter of overriding policy, policy that overrides our current norms on notability and neutral point of view, then deleting content that reflects cutting-edge theory about that would be excluding stuff the UCoC wants us to welcome. It's like treating WP:RGW oppositely to the way that we currently do. This is what can happen when a simplistic corporate formula tries to replace what editors have developed through years of experience. --Tryptofish (talk) 00:19, 16 March 2022 (UTC)

Putting that another way, that class project resulted in content that was AfD-worthy, but it might well have been contrary to the UCoC to delete that content. --Tryptofish (talk) 23:22, 15 March 2022 (UTC)

I've now read the enforcement proposal in its entirety. (Gee, what a novel idea! Read something before giving an opinion about it, rather than after!) Now, I have to say that I'm less bothered by it than I expected to be. It does a better job than I expected at deferring to local wikis and at staying out of purely content issues. There are still issues for me, and the reasons that I will vote no, in that I think that it goes too far in requiring oaths of loyalty rather than just awareness, and it opens up (at larger projects) too much risk of disconnects between the centralized U4C and local consensus about how policies should be applied. --Tryptofish (talk) 17:44, 16 March 2022 (UTC)

Per my comment above, my reason for opposition and for advocating that others oppose isn't that I consider most of it particularly problematic. Rather, it's that once it's passed it will be effectively set in stone given that we can't subsequently amend it without consultation once it's been translated into 325 languages and I assume nobody wants to give the WMF carte blanche to amend it unilaterally. Thus, the usual wiki model of constantly tinkering until we find something that will stick doesn't apply—if we're not confident that this is genuinely the best we can do, we shouldn't be approving it. ‑ Iridescent 20:11, 16 March 2022 (UTC)

Funny that you put it that way. In my comment along with the vote, I also said something approximately like "Wikimedia has always relied on the local communities to develop content policies through experience, and Wikimedia should likewise trust the communities with conduct policies", or something like that. --Tryptofish (talk) 20:21, 16 March 2022 (UTC)

Thanks for articulating those thoughts. There's something there I agree with and am going to amend my vote accordingly, but making clear I would vote yes if the amendment process was adequete. (I don't think U4C Committee-proposed changes or WMF-facilitated reviews are adequate.) I do note as an example that WP:ARBPOL is hard (though possible) to change directly by the community, and as such it's never had a community-proposed modification made to it, so a high bar to change is not a disadvantage IMO, but there needs to be the possibility for direct community change. ProcrastinatingReader (talk) 20:59, 16 March 2022 (UTC)

The process last time we made a significant change to ARBPOL was formal ratification by every member of the then-committee followed by a full vote of anyone who cared to comment, and ARBPOL is an extremely inside-baseball process affecting a single project and about which 99% of editors don't care, whereas this affects every participant on every project. If we end up approving something—either UCoC itself or the enforcement guidelines—and end up accidentally creating something that fundamentally changes Wikipedia's internal dynamics, or even makes one of the projects potentially illegal somewhere, it will be virtually impossible to get it right. (The idea that the WMF could unintentionally introduce an initiative that makes a project illegal isn't hyperbole. Check out the unhappy history of the entirely well-intentioned Wikipedia Zero scheme, for instance.) ‑ Iridescent 20:07, 17 March 2022 (UTC)

The UCoC looks like plain and simple common sense. However, many policies and guidelines started out as what people thought was common sense, and turned out to be distorted and browbeaten into something else over time. So I don't trust that people will enforce the guidelines properly and in a sensible manner. For example, consider everybody who didn't !vote "keep" at Wikipedia:Articles for deletion/Marina Ovsyannikova - I don't think there's any regular editor who didn't think they were all commenting in good faith, and a lot of replies from inexperienced editors were not the sort of comments the AfD process typically uses, but would a random WMF staffer with no experience of this read comments like "Fully agreed and people speaking out against war and for peace should never be censored! Why is this page even considered for deletion? While not every country sees free speech as a key value, and that should be respected, the internet is a place of free speech, in these times please keep this! She is a voice for peace!" and see it the same way? Ritchie333 ^{(talk) (cont)} 12:01, 16 March 2022 (UTC)

Ritchie, further above you ask for reasons to oppose; your first three sentences here sum up why I am opposing. Mike Christie (talk - contribs - library) 12:05, 16 March 2022 (UTC)

Thanks Mike; I have now opposed giving a variation on that. Ritchie333 ^{(talk) (cont)} 12:09, 16 March 2022 (UTC)

It isn't just the delete !votes in that one, several keep votes accuse the delete !voters of being pro Putin. In normal times we wouldn't sanction a bunch of newbies and IP editors for making such aspersions, but with the UCOC, we'd be obliged to hold them to the same standards as we hold the regulars. Where is John Cleese when we need him? ϢereSpielChequers 13:37, 16 March 2022 (UTC)

They took his mic away. Dennis Brown - 2¢ 18:44, 17 March 2022 (UTC)

Okay, I'm voting no, because I don't think the English Wikipedia is broken and, thus, why mess with it? I have no idea what's going on with wikipedia in other languages, so maybe the enforcement mechanism is needed there. But I'm not persuaded.Smallchief (talk) 15:40, 17 March 2022 (UTC)

My take would be slightly different. I do think that English Wikipedia is if not broken certainly strained, because we have policies on civility and neutrality but no definition of what 'civility' and 'neutrality' actually mean. I don't think UCoC in its current form answers either question, though; all the current wording of both UCoC itself and the proposed enforcement mechanism will do is transpose the endless "where I come from this isn't considered an offensive term", "is it racist to point out that non-speakers can't rely on machine translation?" and "what do we actually mean by 'maturity'?" arguments into a different venue. ‑ Iridescent 08:55, 18 March 2022 (UTC)

Good discussion and needed for voter clarification (although I voted "no" awhile back due to not wanting the foundation to force "our-way-or-the-highway" admin signings). A question: As presently written are administrators who sign on (signed in visible ink, blood, or invisible ink) required or encouraged to report an editor who voices an opinion off-Wikipedia which, if said in print, would be a code violation (i.e. daring to act as if free speech were still a thing)? 'hanks. Randy Kryn (talk) 12:13, 18 March 2022 (UTC)

As a general reminder to those who may not have read Meta:Universal Code of Conduct/Enforcement guidelines/Voter information in full and thus missed the part the WMF has tried to slip through, a "no" vote here doesn't mean what it does everywhere else. If a simple majority votes "yes", it moves on to ratification; if the "yes" vote fails to meet the 50% mark then the enforcement guidelines (not UCoC itself which is inviolable) are reworded and the vote is held again; the vote is repeated until it passes. (Personally, I could make a case that even if I thought the wording were perfect, the sheer arrogance of the WMF holding an 'election' in which the only choices are "Yes" or "Yes but not yet" would be grounds enough to come back and oppose it every time until they give up, just to discourage them from trying to pull this fake-legitimacy stunt in future.) ‑ Iridescent 14:52, 18 March 2022 (UTC)

@Randy Kryn: As written, the UCoC forbids you quite categorically from "sharing information concerning [another contributor's] Wikimedia activity outside the projects" without their prior consent. Whether this is in print or not doesn't come into it. It includes sharing such information verbally in the privacy of your home, in the pub, on Twitter, Facebook, Discord, by email to ArbCom, on the phone or in correspondence with a reporter or academic researcher ... sharing information concerning another contributor's Wikimedia activity anywhere, by any means, violates the code as written. This would even include quoting what someone has said on Wikipedia. See related discussions here and here. Andreas JN466 18:04, 18 March 2022 (UTC)

@Jayen466 As I read it, Randy Kryn is asking the opposite question; that is, whether the civility provisions apply to someone's activity elsewhere—e.g. if I make a comment in real life that constitutes "name calling based on perceived political affiliation" or "implicitly suggesting the possibility of unfair embarrassment", and it's possible for someone to join the dots between my Wikipedia account and my real-life identity, can I be sanctioned for it on Wikipedia. The wording of UCoC on the matter is vague; it's not clear whether The following behaviours are considered unacceptable within the Wikimedia movement translates as "these are things you can't do on Wikipedia" or "people who do these things aren't permitted on Wikipedia".

I'm fairly certain the drafters meant the narrower "these are things you can't do on Wikipedia" definition. If they genuinely meant "nobody can say anything in any context which any other person might find objectionable" it would essentially be a carte blance to block every single editor whose real-life identity could be connected to their Wikipedia account; as worded, it would mean a parent could be sitebanned if it could be demonstrated that they'd threatened to punish their child for being naughty. ‑ Iridescent 14:27, 19 March 2022 (UTC)

Yes, I meant if I ever become friends with an admin, God forbid, and text them "I believe that ______ has the home-grown manners of a _________ ___ and if only the sheep could talk she'd say '______ ___ _____, buddy!'" does the admin have a contractual duty to block me on Wikipedia or just on their phone. Randy Kryn (talk) 14:59, 19 March 2022 (UTC)

Me, I'd ask why they say "Wikimedia movement" rather than "on Wikimedia websites and physical spaces" but that's like knitpicky. Jo-Jo Eumerus (talk) 15:08, 19 March 2022 (UTC)

That line at least I think is reasonable—there's more to Wikimedia than just Wikimedia websites and physical spaces, even though the websites account for 99.9% of it. This particular wording reduces (albeit doesn't remove) the gray areas over whether (for instance) comments made over Zoom, or at local meetings which aren't held on WMF property, still get covered. Think of it as finally clarifying the 20-year-old question over whether the IRC (and now Discord) channels are exempt from the rules. ‑ Iridescent 03:05, 22 March 2022 (UTC)

Right. I did get hung up on the "if said in print" bit, given that the UCoC doesn't differentiate between "print" and "non-print" and just sticks with "sharing". Thank you for the correction, Iridescent; and I see what you meant now, Randy Kryn. You were talking about the UCoC's considered opinion that anyone who calls another person, say, a "murderous dictator", or a "pure thug", or a "war criminal", based on the other person's politics, thereby marks himself out as the kind of low-life we wouldn't want to have in the Wikimedia movement. Got it. Andreas JN466 18:58, 19 March 2022 (UTC)

Well, not in so many words. I was talking about a sheep or Wikimedia pulling the wool over its eyes or something. Randy Kryn (talk) 19:20, 19 March 2022 (UTC)

Andreas JN466 20:21, 19 March 2022 (UTC)

Pardon the new subheader...

Here's a vote count so far, courtesy of Xeno (WMF):

•   enwiki: 564 (37.3%)
•   dewiki: 168 (11.1%)
•   frwiki: 90 (6.0%)
•   eswiki: 69 (4.6%)
•   ruwiki: 71 (4.7%)
•   plwiki: 65 (4.3%)
•   metawiki: 50 (3.3%)
•   zhwiki: 46 (3.0%)
•   jawiki: 44 (2.9%)
•   itwiki: 45 (3.0%)
•   commons: 29 (1.9%)
•   arwiki: 20 (1.3%)
•   cswiki: 19 (1.3%)
•   ptwiki: 18 (1.2%)
•   nlwiki: 17 (1.1%)
•   kowiki: 17 (1.1%)
•   trwiki: 15 (1.0%)
•   cawiki: 11 (0.7%)
•   idwiki: 10 (0.7%)
•   78 others: 144 (9.5%)

Seems like it's the big wikis dominating so far, although I wonder how much off-home wiki voting (i.e editors mainly active in project A voting through project B) there is going on, I saw a few such votes. Jo-Jo Eumerus (talk) 12:18, 18 March 2022 (UTC)

If I recall correctly, SecurePoll determines your home wiki based on what is identified as your home wiki on Special:CentralAuth. Everyone has to vote from Meta. Risker (talk) 13:26, 18 March 2022 (UTC)

IIRC CentralAuth tends to over-represent en-wiki as it's based on the wiki where the account first edited, and a lot of people start off on en-wiki even if it's not their native tongue just because it's so much bigger and thus more likely to have the article they were looking for and (importantly) more likely to be accurate and up-to-date. A lot of the wikis outside the big global languages are essentially just collections of pages on the interests of the half-a-dozen people who've declared themselves that site's owners. ‑ Iridescent 14:25, 18 March 2022 (UTC)

For anyone who's not aware, "yes" has scraped a decidingly unconvincing win. For the record, I'll refuse to participate in their 'mandatory training' reeducation scheme and (depending on the exact wording on what's proposed) would probably go so far as to say that if anyone does voluntarily participate in it, I'd consider that grounds in itself for a loss of confidence in that person's judgement to the extent that I'd consider them unfit to hold advanced permissions. ‑ Iridescent 19:03, 11 April 2022 (UTC)

In case you're wondering what's ahead...

I made a single comment on mediawiki.org that "Wikimedia Foundation Inc. is corrupt and bad. The Web team is a particularly egregious demonstration." which is demonstrably true and probably some of the mildest criticism you can offer. I got a 1-month block from the TechConductCommittee account. Lawwwwwwl, this place is hopeless. --MZMcBride (talk) 17:31, 19 April 2022 (UTC)

Behold the future. To give a tiny bit of benefit of the doubt mediawiki.org and Meta have always had a slightly puritanical streak despite (or perhaps because of) being the haunt of some industrial-grade wackoes. I'd like to think they wouldn't pull that kind of crap on any of the large wikis, although if the WMF ever did decide to set up a similar squad of civility-enforcement Death Eaters on en-wiki, I could probably already predict the names of everyone who'd volunteer. ‑ Iridescent 17:42, 19 April 2022 (UTC)

It remains to be seen what happens when somebody criticizes another user, but we now know for sure that criticizing the WMF will result in a trip to the gulag. --Tryptofish (talk) 19:59, 19 April 2022 (UTC)

It looks like my Phabricator account has been disabled as well: <https://phabricator.wikimedia.org/p/MZMcBride/>. For calling the organization corrupt and bad. Huh. --MZMcBride (talk) 02:57, 20 April 2022 (UTC)

I wonder what they were thinking they would accomplish by blocking you for a month when you on average edit that wiki far less than once a month. For what it's worth, blocking an account on MediaWiki.org or Wikitech automatically blocks the corresponding account on Phabricator, so it's probably not a deliberate decision to disable your phabricator account. And, as another admin on MediaWiki.org (albeit one who isn't very active and uses the block button rarely) I agree the block wasn't warranted. * Pppery * _{it has begun...} 16:40, 20 April 2022 (UTC)

Regarding I wonder what they were thinking they would accomplish by blocking you for a month when you on average edit that wiki far less than once a month, I assume it's a conscious attempt to send a "nobody is safe" message. This is the first block they've placed for four years, and there was no attempt made to warn or discuss beforehand; it's impossible to believe this wasn't both calculated and deliberate. ‑ Iridescent 19:01, 20 April 2022 (UTC)

I think the other interpretation is that in the context (on a page where Tuvalkin was warned about the Tech CoC for this edit), MZMcBride chose to attack the Web Team in a similar not-very-personal way and the TCC then blocked him because they thought that way he would Respect Their Authority. —Kusma (talk) 19:59, 20 April 2022 (UTC)

That group doesn't always bother to use the official account. Many of the members have been admins for years. See, e.g., this block from a year and a half ago. I've only rarely seen them block people, but it is generally true that they enforce the level of civility that you'd expect in an actual office, when speaking face to face with a co-worker. If a comment would get you fired or escorted out of the building at a real-world job, it'll get you blocked in the technical spaces (if you persist after being warned). It's different from enwiki's approach. Some people even think it's better. WhatamIdoing (talk) 05:30, 3 May 2022 (UTC)

By the level of civility that you'd expect in an actual office I assume you mean "the level of civility that would be expected in the WMF's office" and not actually "in an actual office". I've been around the block a few times, and have never (and I mean never) worked anywhere where MZM's comment would get you fired or escorted out of the building. ‑ Iridescent 23:24, 3 May 2022 (UTC)

Really? Corruption is a serious crime. Just how often did your colleagues accuse each other of committing serious crimes, with nobody minding? WhatamIdoing (talk) 15:33, 4 May 2022 (UTC)

Hi WhatamIdoing. Who said or even suggested legally (in the U.S.?) corrupt? I think it's perfectly possible that Wikimedia Foundation Inc. is ethically and morally corrupt without meeting a strict legal definition.

I'm not a lawyer so you'd have to ask one if collecting millions of donor dollars under false or highly misleading pretexts and wasting them is legal. Regardless of the legality, I'm within my rights to call it out. I might write an essay at some point. --MZMcBride (talk) 18:39, 4 May 2022 (UTC)

An advantage of living in a free society is that you can share your belief that your government is corrupt without having to fear repercussions just for saying so – and plenty of people do so. The governments that put you in jail when you say anything they feel discredits them are generally the authoritarian ones.

One trick used by the latter is to have laws that practically everyone has broken, but use them only to punish the "right" people. The UCoC seems not unlike that. Take this mailing list post from Wikimedia Taiwan, urgently asking for the UCoC enforcement guidelines to be implemented now, rather than waiting for another vote. The post "shares information concerning other contributors' Wikimedia activity outside the projects", in breach of the UCoC as written, and clearly hopes that enforcing the UCoC per the draft enforcement guidelines will somehow thwart or sanction the contributors they're talking about. The reply by a WMF board member is entirely sympathetic, the UCoC breach passed over without mention.

This idea – to have rules that most anyone will have broken, and then apply them selectively to sanction only the "right" people – has some very obvious short-term benefits and equally obvious long-term drawbacks, as any track record of using rules in this way will expose the actor to credible and effective charges of hypocrisy and abuse of power from outside their immediate sphere of control – exactly the same consequences that are experienced by authoritarian governments.

Unfortunately, it looks more and more like the WMF is hell-bent on adopting that authoritarian government pattern. The new board (s)election method, for example, arguably follows the Iranian model by requiring candidate vetting and approval by another body than the electorate. Andreas JN466 15:45, 5 May 2022 (UTC)

@WAID, I'm quite sure you (wearing either your editor or your WMF hat) understand the concept of 'figure of speech' and don't really believe Wikimedia Foundation Inc. is corrupt and bad. The Web team is a particularly egregious demonstration. actually constitutes "accus[ing] each other of committing serious crimes". (If I wrote "a crowd of 70,000 watched the Rams beat the Bengals last February", would you wonder why none of those 70,000 people had reported the assault?)

Even if one takes it literally, it's perfectly possible to say that elements of an organization have questionable integrity without it rising to the level of a criminal allegation. (We are talking about an organization which employed someone who had a side hustle running a website for people to trade and rate crime-scene photos of murdered children; which employed someone who shared their fantasies about murdering women in internal discussions; which employed someone who was caught whitewashing articles for pay; which employed someone who was caught using a sockpuppet to harass other Wikipedia editors; which oversaw the entire Knowledge Engine fiasco; which has at least arms-length responsibility for the chapters and affiliates some of which are outright moral vacuums; and which going further back had a Chief Operating Officer with a lengthy string of serious criminal convictions.)

Yes, anything that gets to this kind of size is going to have a few bad apples, but there's a fairly consistent pattern over the years of the WMF trying to cover misconduct up and style it out (every one of those examples I've just given only came to light as a result either of leaks, whistleblowers, or external detective work by someone else); it's not unreasonable for people to raise concerns. (It's now been three years; is the WMF able to say yet what the total amount of 'grants' given to Laura Hale over the years added up to?) ‑ Iridescent 19:07, 7 May 2022 (UTC)

I'm not convinced that it was a figure of speech. I think it is far more likely that it was a non-figurative claim of moral and ethical corruption.

Imagine that you are dealing with a pattern of behavior. One day at work, someone walks over to your team – he has nothing to do with your team, he's not a manager, etc. – and offers his unsolicited but apparently heartfelt opinion: "You know, if it were up to me, you'd all be fired". Another day, he goes to your team and calmly says "Your project should never have been started, and it ought to be cancelled." The next time, he announces "You produce garbage". Another day, he tells you "You all are morally and ethically corrupt".

Is this normal and expected workplace behavior, in your experience? Is it behavior you would recommend to a colleague? Do you think your team should be expected to accept this treatment as "just part of the job", indefinitely, without complaint? WhatamIdoing (talk) 17:57, 15 May 2022 (UTC)

I notice that you frame your example as one of repeated, continuous expression, which I can appreciate would lead to a cumulative effect. But, unless there is more that I'm unaware of, the comment by MZMcBride was a single one. So I don't think it could be characterized as an ongoing problem over time. You also describe something where the person keeps approaching coworkers face-to-face and frames the criticisms as "you". That's more confrontational and personalized than what happened here. In addition, there are differences between a paid workplace and volunteer editing at a website, in terms of contractual obligation to abide by the management's expectations (although the UCoC seeks to significantly reduce those differences). If the example were of the leader of the team telling the team members "I've discovered that you are all corrupt and I'm going to take action", that might be a jolting statement, but it would also be accepted at many workplaces as at-will employment. Here, I think it might be closer to compare it to someone who is not part of the team offering an unsolicited but heartfelt opinion that "I think you guys have been doing a bad job, and it would help the rest of us if you get it together", one time, and then being docked pay. --Tryptofish (talk) 20:43, 15 May 2022 (UTC)

I believe that MZMcBride has earned his reputation as a very persistent and consistent critic of the team that supports mw:MobileFrontend, including face-to-face communications. This is the team he called "a particularly egregious demonstration" of the WMF being "corrupt and bad". If you are looking at this incident without taking the whole history into consideration, then of course you might think it was an overreaction. (I would hope for a first-time incident, management would work out an apology and a promise not to re-offend; for people who are already living paycheck to paycheck, docking their pay can mean food insecurity or losing their housing. But I think that around the third or fourth time, you could expect to be fired.)

It's also worth noting here that this action was taken by a group of volunteers who were implementing their local community-approved policy. WhatamIdoing (talk) 02:47, 16 May 2022 (UTC)

You compare these interactions to interactions among colleagues – peers – all working for the same employer. But staff and volunteers are not peers. Staff are paid, volunteers are not. To me these interactions feel more like the sort of interactions that happen between labour and management. I think you'll readily admit these frequently tend to be quite adversarial, with management accused (often justifiedly so) of all sorts of malfeasance, employees taking strike action and so forth. (Hell, I remember WMF staff being quite vocal about Lila back when.)

And as the WMF has recently voiced its commitment to human rights, it's worth mentioning that these labour rights – right to unionise, right to take strike action, i.e. to be adversarial – are also recognised as fundamental human rights.

Now you can say that MZMcBride is not a union representative and has no democratically earned mandate. That is true. But a lot of people think that he usually has a point, just as was the case with Fram. Andreas JN466 14:02, 16 May 2022 (UTC)

MobileFrontend is bad and should not exist. The team that's working on it is actively harmful. Saying so isn't the issue here, squandering millions of dollars and community goodwill on a bunch of bad and failed projects is the issue. But of course you already know all of this. --MZMcBride (talk) 16:06, 16 May 2022 (UTC)

I don't know enough of the back story here so will take you (WAID's) word for it on has earned his reputation as a very persistent and consistent critic of the team that supports mw:MobileFrontend. What I would say is whether that's a bad thing or not depends purely on who's in the right. "A very persistent and consistent critic" could describe

1. Someone who camps outside a politician's house screaming abuse through a megaphone because they're not telling the truth about the lizard people;
2. A heroic Galileo figure martyred for refusing to bow to orthodoxy in the face of evidence, and who will subsequently be lauded as a hero;
3. Literally anything in between.

I don't know what he's like in his interactions with the WMF, but in my experience MZM has consistently over the last 15 years been someone with whom I may not agree, but whose opinions are pretty much invariably worth listening to. If he's saying he thinks there's a problem and is giving a coherent explanation as to why he thinks it's a problem, it's almost certainly worth listening to.

Statement of the obvious perhaps but if mw:MobileFrontend refers to the m.wikipedia.org site and the iOS/Android apps, then MZM is hardly alone in thinking MobileFrontend is bad and should not exist. It's not exactly one of Wikipedia's state secrets that readers complain about the design and usability of the mobile site. ‑ Iridescent 12:47, 21 May 2022 (UTC)

@Jayen466, regarding One trick used by the latter is to have laws that practically everyone has broken, but use them only to punish the "right" people while I agree that the UCOC is a good example of this, the WMF certainly can't be blamed for introducing the "create a rule which technically every person in the world has broken, but only enforce it against those considered undesirable" approach. Fortuitously, Internet Archive captured a full snapshot of Wikipedia the week before the WMF was founded which means it's possible to recreate what Wikipedia looked like in the Last Days of Bomis without having to go through hundreds of logs to confirm which page was created on which date and what was linked from where. As such one can confirm that even back when the entirety of Wikipedia:Policies and guidelines was 566 words long the intentional ambiguities like "treat others with respect" and "represent differing views fairly" were already starting to creep in. ‑ Iridescent 02:46, 8 May 2022 (UTC)

Yes, this idea to tailor the application of rules to the one at the receiving end of them has a long history in the volunteer community as well. As for "represent differing views fairly", I remember being really impressed when I first arrived here by the concept of "Writing for the enemy" – or, as the current version puts it (much less pithily, and much more politically correct), Wikipedia:Neutral_point_of_view/FAQ#Writing_for_the_opponent. The current version of the FAQ says,

• "I'm not convinced by what you say about "writing for the opponent". I don't want to write for the opponents. Most of them rely on stating as fact many statements that are demonstrably false. Are you saying that, to be neutral in writing an article, I must lie, in order to represent the view I disagree with?
• The great thing about NPOV is that you aren't claiming anything, except to say, "So-and-so argues that ____________, and therefore, ___________." This can be done with a straight face, with no moral compunctions, because you are attributing the claim to someone else. Even in the most contentious debates, when scholars are trying to prove a point, they include counter-arguments, at the least so that they can explain why the counter-arguments fail."

We live in a world today where what you're told about the world depends to a very great degree on where you live. The problem assumes such proportions that we don't even know any more (or perhaps we never did) what people elsewhere are told and genuinely believe. Occasionally one may find articles that try to bridge the gap – the New York Times for example had an article recently on Russian reporting of the Ukraine war, and the Guardian has published similar efforts, describing how press reporting of the conflict differs in places like Brazil or South Africa, or what Chinese social media users are saying (apparently, Putin is portrayed as a hero standing up to the West in most posts).

If Wikipedia were true to its original idea, it would be a place where you would see competing narratives, neutrally presented, each attributed to its origin. I'm pretty sure this is not what the present WMF management would want and I suspect it is not, by and large, what is really happening in Wikipedia ... but would it even be a "good thing"? Or was neutrality too amibitious a goal, given that we've entered the age of fake news? I have no idea.

I do think though that it would be good to have a place that would simply report, without judgment, what the politicial propaganda machines are spewing out in the various corners of the earth (ours not excepted). Maybe it wouldn't be an encyclopedia, but it would be instructive – and perhaps bring people to a common baseline. Andreas JN466 19:09, 12 May 2022 (UTC)

All the above is only true for the infinitesimal proportion of the content that deals with contentious topics. Even with the big topics, for something like Horse or Locomotive, there are no alternative points of view that need to be mentioned (unless we're going to pander to obvious looney-tunes ideas like Baraminology). For the first five pages I got clicking Special:Random (Sandalinas, 1962 Icelandic Cup, Autonomy Liberty Democracy, Eric of Brunswick-Grubenhagen, Coast station (PAAC)) there's nothing—other than arguably the definition of "centre left" in the article about the Italian political party—that would be any different were the article written by anyone else, from anywhere else, of any other political persuasion. I've no reason to believe that I couldn't repeat the experiment virtually indefinitely with the same result. NPOV gives the impression of being a major existential crisis for Wikipedia because the handful of articles where potential bias is an issue get a disproportionate amount of attention from critics, but in reality Wikipedia's main problem is bad or nonexistent sourcing, not cultural bias.

(If I were in charge, Wikipedia wouldn't cover current events at all, and we'd have some kind of "nothing gets an article until it's been mentioned in a book" rule. The solution to Wikinews being a useless joke of a website should not have been to surreptitiously import its problems over here. As far as I'm concerned, Wikipedia's entry on the Ukraine war should be "It is currently too early to establish what has happened, here's a list of external sources where you can see coverage of it". See also about 95% of BLPs other than genuinely major figures where it would both be perverse not to cover them, and where the subject is so high profile that the expectation of privacy no longer applies.) ‑ Iridescent 19:49, 12 May 2022 (UTC)

Of course, such contentious articles and biographies can attract a very large number of pageviews... So while they represent a small proportion of what's there, they represent a much larger proportion of what's actually being read on any given day (outside of wrestling articles). In the tech world, whose perspective is dominant in Wiki(m/p)edia, traffic is all that counts. Not having biographies and articles on current affairs would run counter to Wikipedia's entire reason for being: they're among Wikipedia's big traffic success stories. I have to admit that looking up actors, TV shows and developing news stories etc. makes up a significant part of my own Wikipedia usage. Quality-over-quantity thinking would require a completely different management. Andreas JN466 23:29, 12 May 2022 (UTC)

I don't really buy the 'pageviews=importance' argument—it just sounds to me like a slight variation on the 'core topic' idea which is one with which I've never agreed. (Besides, quite often the articles with the highest readership are on topics which no sane person would consider important.)

Wikipedia's strength is the breadth of its coverage, and the fact that people can find out about Little Thetford as easily as they can find out about France. IMO the measure of the importance of a Wikipedia article is "if we shut Wikipedia down, how easy would it be for someone to find out about the topic from a Google search?". To me, Apororhynchus is a more important article than Ant.  ‑ Iridescent 05:14, 13 May 2022 (UTC)

A nice example, and I do get you point. To be clear, I didn't say I was endorsing the pageview criterion. It's just important to be clear that this is a key criterion according to which this project is being run.

Take for example this report on Project Tiger, which mentions that "Google has also been pushing for translation of popular Wikipedia articles in Indian regional languages ... While Wikipedia is run by an independent non-profit, it does work closely with Google to identify pages that need quicker updates or more translations based on search volumes."

So if you look at it from this angle, the list of Most Popular Wikipedia Articles of the Week (I had been looking at the April Signpost Traffic Report, which shows much the same ...) makes perfect sense. It all boils down to how you define "important". As Jimmy Wales said on Facebook, there are outside interests that don't necessarily have Wikipedia's core values at heart – values which he characterised thus: "that we are a community-first project, that we are a charity, that we are neutral, that we strive for quality, and that we work towards governance that means safety for all these values in the long run."

I believe that all these values are expendable to some of Wikimedia's biggest supporters today. What's important to a Big Tech company, for example, is to have as many eyeballs as possible on their pages for as long as possible. It's how they make their ad money. To them, Wikipedia is there to deliver copyright-free, cost-free, freely reproducible and freely mineable text on popular topics in as many languages as possible. This helps them do good business. Andreas JN466 09:52, 13 May 2022 (UTC)

What's important to a Big Tech company, for example, is to have as many eyeballs as possible on their pages for as long as possible. It's how they make their ad money. To them, Wikipedia is there to deliver copyright-free, cost-free, freely reproducible and freely mineable text on popular topics in as many languages as possible. describes the situation back in the day, but I don't really think it's the case now. Wikipedia's value to Google and Facebook isn't as a source of text, it's as a source of data.

From the perspective of Big Tech, the platonic ideal of Wikipedia is as a huge repository of free data, on which eventually they'll run Reasonator-style algorithms to generate articles on the fly tailored to both the requirements of their readers, and the interests of their sponsors and advertisers. (How much would Disney be willing to pay if everyone looking at United States who'd recently been searching for holidays saw mention of Disneyland in the first couple of paragraphs of text, but equally importantly didn't give the same prominence in the case of users whose histories didn't give any indication they were considering a vacation, to avoid people thinking "I'm sick of hearing about this" and developing subconscious negative associations?). Strip away all the 'metaverse' buzzwords and "create the appearance of giving the people what they want, to the point where they assume that if they're given something it must be what they want" is the long game on which all the Big Tech companies, in particular Facebook, are betting the farm. In this context, Wikipedia editors and non-English versions of the projects are just the unfortunate price they need to pay in order to convince people to create and curate their database for free.

(I'm not sure as many languages as possible was ever true. I doubt Mark Zuckerberg would lose a second's sleep if we shut down even German Wikipedia, let alone Lezgian or Cornish; from their perspective the main point of non-English versions is to encourage editors on topics that aren't popular in English-speaking countries so they can extract the data from those as well. Google in particular also likes minority-language wikis as when people translate pages from English/French/Spanish/etc onto Atikamekw Wikipedia they tend to translate directly so it gives their translation software a free dataset of vocabulary and grammar.) ‑ Iridescent 18:27, 13 May 2022 (UTC)

You're right to mention data – there wouldn't be a Wikidata if what you said weren't true. And of course Wikidata hoovered up a huge amount of CC BY-SA content from Wikipedia and turned it into CC0 data for reusers – exactly what Wikimedia originally promised wasn't going to happen: "Wikidata does not plan to extract content out of Wikipedia at all. Wikidata will provide data that can be reused in the Wikipedias. And a CC0 source can be used by a Share-Alike project, be it either Wikipedia or OSM. But not the other way around. Do we agree on this understanding?" That's what Denny Vrandečić said ten years ago.

However, by the same token, there wouldn't be an Abstract Wikipedia, designed to have volunteers create CC BY-SA Wikifunctions articles which can then potentially be machine translated into CC0 (!) human-language articles, if there weren't a significant interest in having Wikipedia articles – text – in those currently underrepresented languages. If you ask Google a question, e.g., it'll quote text passages from relevant pages found in an answer box, and now even highlight the passage when you go to the page in question.

But another key reason why texts are wanted is the rise of voice assistants. I mean, these days it's not just Amazon's Alexa, Siri and Google's voice assistant that read you Wikipedia articles, even your car speaks Wikipedia when it talks to you. And many of these African and South Asian languages have tens of millions of speakers, so there are markets to be conquered there that may become very lucrative in a decade or two – if volunteers somehow, be it via Abstract Wikipedia or incentivised native Wikipedia editing, create the requisite texts.

(I think Reasonator was an earlier incarnation of this idea that Magnus Manske came up with. As both your Rajneesh example and this one for Obama illustrate, Reasonator output never really got to a stage where it would have been any use for a voice assistant ... nobody'd want to listen to that.) Andreas JN466 07:29, 14 May 2022 (UTC)

I didn't know until now about the CC0 issue. It seems to me, if I understand correctly, that taking CC-by-SA content from here, and making it CC0 at Wikidata is not only a broken promise, and not only a bad idea functionally, but also downright illegal. At least if some content contributor here were to make a legal issue out of it (something I have zero interest in doing, myself). (Oh, wait, does that mean that the UCoC is going to give me an NLT block?) --Tryptofish (talk) 20:16, 14 May 2022 (UTC)

See Whither Wikidata? in the Signpost. Heather Ford later wrote a great chapter for the "Wikipedia @ 20" MIT book that touches on this as well: Rise of the Underdog. Andreas JN466 21:00, 14 May 2022 (UTC)

I strongly suspect voice assistants in their current form are a fad that will be forgotten in a few years. They're great as glorified hands-free remote controls, but I'm not sure I've ever known anyone other than small children who carries on using the "Tell me about…" functions once the initial novelty wears off.

If I had to gamble on where the next big jump takes place in terms of how Wikipedia's information gets used, I'd put my money on tailored versions of what Britannica pretentiously calls "Micropedias"; an algorithm that, based on your previous search history and known profile, serves up a one- or two-paragraph summary of an article giving only those parts in which you're likely to be interested. Lila's Silicon Valley gibberish was impenetrable, but I think something like that is what the whole Knowledge Engine boondoggle was ultimately supposed to be. ‑ Iridescent 14:15, 14 May 2022 (UTC)

I am really curious how the voice assistant thing will turn out too. I don't use Alexa on the Kindle and originally thought much the same like you, until I observed a mate (middle-aged, not technically minded at all) who would routinely ask his mobile questions while I was visiting with him ... suddenly I wasn't so sure any more. The Echo seems to sell well too for now. Andreas JN466 14:59, 14 May 2022 (UTC)

Alexa is a special case because Amazon run it as a loss leader to pull people into their ecosystem in the hope that once they're there, they can sell them things. The signs that the bubble is bursting are already there—Microsoft has for all practical purposes killed Cortana, Apple is keeping Siri on life support but no longer appears to be doing any significant work on it, Samsung Bixby is a bad joke. The only one that's still being kept alive is Google Assistant, and I suspect that's as much an artefact of bet-hedging and not wanting to hand a monopoly to Bezos as anything else. ‑ Iridescent 19:00, 14 May 2022 (UTC)

@Jayen466 As a daughter of parents who are substantially less technically-minded than both their children, I wouldn't bet on people such as them sustaining any technological market. Any substantial (even if substantially cosmetic) update will confuse the easily confused...and if substantial cosmetic updates wouldn't confuse your mate, he's at least slightly more technically inclined than my dad. I dream of horses (Contribs) (Talk) 19:25, 14 May 2022 (UTC)

There isn't much of an interface that one could update. My mate says "Hey Google", asks a question and either gets an answer or not.

For what it's worth, carmakers – not just premium brands like Mercedes (typically driven by older people), but even volume manufacturers like Volkswagen – are currently falling over themselves building voice assistants into their cars. There's intense competition, and I'd be really surprised it they were to remove them in five or ten years' time, or were to stop working on them to make them better. Andreas JN466 20:52, 14 May 2022 (UTC)

@Jayen466 There isn't much of an interface that one could update. Most likely true. I was assuming that there was something updat-able that I hadn't thought of myself. I dream of horses (Contribs) (Talk) 21:00, 14 May 2022 (UTC)

With a hat-tip to WO – a Professor Philip M. Parker at INSEAD is working on Botipedia – a content engine to generate millions of articles for underserved languages. This could make editors obsolete, and he says it'll cost 1% of what Wikipedia costs a year (well, heard claims like that before ...). One interesting feature illustrated in Parker's video on Botipedia is the ability to choose which parts of the political spectrum the article's sourcing should come from. Andreas JN466 09:13, 17 May 2022 (UTC)

I can't see that working, for the same reason Reasonator doesn't work and the same reason I can't see Abstract Wikipedia working. That approach is fine for creating stubby articles that just deal in facts and figures, but I've yet to see an AI script that can handle nuance convincingly. When one is purely looking for "how many goals did David Beckham score during his spell at Preston North End?" information this works, but there's a chasm between the "infobox converted into text" approach and an actual article that passes the "would somebody not already familiar with this topic find it interesting?" and "why is this apparently uninteresting topic actually significant?" tests.

When someone can create a bot that can explain coherently why The Princess Bride is considered culturally significant despite its lack of commercial success, the difference between rugby union and rugby league and why fans in different areas tend to support one or the other, what the Holy Roman Empire was, why so many musicians cite Mark E Smith as an influence despite his absence of commercial success and lack of apparent musical ability, or how tontine financing worked and why it's no longer used, then we'll talk. Until then, I'll continue to consider every "I can create an internet that writes itself" claim to be just another piece of tech bro bullshit. ‑ Iridescent 17:41, 23 May 2022 (UTC)

There's a difference between an in-car voice assistant and the voice assistant on your phone, even if they rely on the same underlying cloud-based voice recognition and mass data collection technologies. There's an obvious use case for someone controlling a moving vehicle to be able to input a destination into a satnav, make a phone call, or control the in-car sound system, without taking their hands off the wheel or their eyes off the road. There's considerably less of a use case for someone controlling a moving vehicle to be able to find out the name of the bass player in Aerosmith. The sort of questions for which a voice-controlled search function is useful when driving ("Google, is it going to rain in the next 45 minutes?", "Siri, where is the nearest florist?", "Alexa, how long will it take to drive to the airport?") are precisely the kind of real-time questions for which Wikipedia/Wikidata are of absolutely no use. ‑ Iridescent 06:13, 15 May 2022 (UTC)

I wouldn't exclude the possibility of in-car conversations touching upon the Aerosmith line-up, but you're right that those use cases will probably be of minor importance. On the other hand, the report I linked earlier on the Mercedes MBUX system specifically mentions Wikipedia articles. One use case that has been around for years is Wiki local, designed to be accessed while on the road. This provides access to Wikipedia articles whose geodata match the vehicle's location, so you can learn about places you've passed, nearby points of interest etc. --Andreas JN466 13:59, 15 May 2022 (UTC)

If notability standards were like the way you wanted, we'd have less articles, but would still find a way to argue about notability. I dream of horses (Contribs) (Talk) 02:06, 13 May 2022 (UTC)

We'd have arguments about notability whatever the definition was, since as long as an edge exists there will be edge cases. Taking a more active stance against recentism wouldn't affect the scale of arguments about notability and if anything might increase it, but it would have an impact on the potential for us to get things wrong, and on the potential real-world damage caused when we do get things wrong. (This is the entirety of what Britannica currently has to say about the current situation in Ukraine; a full and accurate summary of those facts on which there's broad agreement, but a complete absence of he-said-she-said or "according to some reports".) ‑ Iridescent 05:01, 13 May 2022 (UTC)

WhatamIdoing writes, "It's also worth noting here that this action was taken by a group of volunteers who were implementing their local community-approved policy." WhatamIdoing is actively employed by Wikimedia Foundation Inc. And the "group of volunteers" she's referencing includes m:User:Martin Urbanec (WMF), m:User:ASarabadani (WMF), and m:User:MusikAnimal (WMF). Kunal recused in this case, but he previously worked for Wikimedia Foundation Inc. as well. And of course at least two of the auxiliary members are actively employed by Wikimedia Foundation Inc. So when you say this action was taken by a group of volunteers, it's—charitably—misleading.

Most organizations have ethics policies that prohibit taking retaliatory action against people who are accusing their organization of malfeasance and corruption. It's also fairly relevant that this group itself was not elected by the community and new members are chosen by current members. This is painfully obvious self-serving bullshit. --MZMcBride (talk) 16:53, 16 May 2022 (UTC)

That community is largely populated by current and former WMF staff, so it's unsurprising that their volunteer structures would also involve a lot of current and former staff (and even the occasional person who hopes to be hired at some point in the future – Ladsgroup was on the committee for several years before the WMF hired him). However, people who serve on that committee do so as volunteers, in their own spare time, and without compensation or direction from the WMF. WhatamIdoing (talk) 22:38, 16 May 2022 (UTC)

As per my comments elsewhere in this sprawl I know nothing about the Mobilefrontend team so can't really comment here. What I would say though is that the WMF does have recent form for packing committees with the payroll vote and omitting to mention it on the grounds that they're participating as volunteers not employees/grant recipients; regardless of whether MZM is right on this occasion, it's not conspiracy-theorizing to suggest it might be happening. ‑ Iridescent 13:14, 21 May 2022 (UTC)

Also, the TCoC was not approved by the community; the promised process for the full text as a whole never happened. I don't think this group of WMF appointees (or appointees' appointees' etc) can be considered to be a community group. --Yair rand (talk) 18:12, 24 May 2022 (UTC)

Break—Analytics and automated curation

I can't keep up with the indenting, but on the point regarding eyeballs and attracting users, I'll point again to the opening paragraphs of m:experiments, which I think very clearly and concisely capture some of the issues specific to Wikimedia. It's now nearly a decade old, but the points are still very relevant in my opinion.

Regarding content translation and data mining, I think you're right that Facebook and Google and others want to algorithmically create content for readers using Wikimedia wikis as one of many sources/backends. There are actually active efforts backed by Wikimedia Foundation Inc. and Google Inc. to do this with Abstract Wikipedia / m:Abstract Wikipedia. I remain pretty skeptical that what we need is a metalanguage to solve our problems. Apologies if this topic has already been thoroughly discussed here. --MZMcBride (talk) 18:57, 13 May 2022 (UTC)

If they don't hurry, they'll miss the window for being able to blame it all on Katharine Maher, and will need to wait until the next chief exec resigns before they can cancel it without losing face…

I don't think algorithmically created content tailored for specific users is necassarily a bad idea in theory even from the point of view of Wikipedia, let alone from an advertiser's perspective. If I visit the articles on Toronto, Montreal and Ottawa and in each case jump straight to the sections on architecture and sports and ignore the rest, I'd consider it a benefit if when I visited Calgary the sections on sports and architecture were at the top. (Whether that would justify the intrusive surveillance and mass data retention on reader activity that would be necessary to make it work is another matter, but we have examples like Mozilla to show that it's possible to at least try to reconcile "personalized user experience" with "allow the user control over their own privacy". If it's clear that algorithmically created output is the destination, then for all the WMF's many faults I'd probably trust them more to be the ones driving the train than Zuckerberg and Bezos's goons.)

Another issue is that during the transition phase when we're still writing articles rather than just inputting data to a giant database it would mean having to rewrite every piece of content from a modular approach as we'd no longer be able to rely on "as previously mentioned…" etc. I personally think we should at least aspire over the very long term to be doing this anyway, to cater for mobile readers who are being directed straight to a particular section and for whom the rest of the page is collapsed and may as well be invisible, but it would be a massive cultural shift.

Auto-generating articles for those languages in which the articles haven't been written and for which Google Translate doesn't work (like the Haitian Wikipedia experiment) certainly makes a degree of sense. To stick with the Little Thetford example I've already used, if I were a Haitian who wanted to find out about it then the page that's autogenerated by the WMF's experiment, even in its current very rough form that resembles MS Access c. 1995, is certainly more useful than a red link (and I'd argue, probably more useful than French Wikipedia's human-written effort). ‑ Iridescent 05:05, 14 May 2022 (UTC)

Allow me to briefly recommend DeepL Translator. It has fewer languages than Google Translate but in my experience does a much better job with the ones it does have. Andreas JN466 11:25, 14 May 2022 (UTC)

Metalanguage is the right word. What I don't get is how someone can argue with a straight face that while the article in the metalanguage should be CC BY-SA, the human-language derivative – the translation into Swahili, or Kannada, or whatever – should be CC0.

To give an analogy, if I translate a short story into a metalanguage like Wikifunctions and then have the result machine-translated into Swahili, this does not magically remove the original story's copyright. Andreas JN466 21:45, 14 May 2022 (UTC)

Ucoc vote rationales posted

See meta:Universal Code of Conduct/Enforcement guidelines/Voting/Report/Comments- posting this here as I think it is in the interest of those who frequent this page and the community in general, and it could spark some constructive (or not) discussion. Some PII was been removed but they otherwise seem pretty complete. I believe you are allowed to out yourself, so I will say I was #237, “I don’t understand this really, so I cannot support.” The meta pages were too obtuse and what meant what confused me, so I opposed it: don’t support something you don’t understand. That might seem silly, but I believe it’s fair. Moneytrees🏝️(Talk) 05:08, 20 May 2022 (UTC)

I'm pretty sure 349 is me, or someone very much on my wavelength. Not sure if whoever put my words there under a CC-BY-SA licence understands what the BY bit of CC-BY-SA stands for, but understanding community norms such as attribution hasn't been the WMF's strong point. ϢereSpielChequers 12:41, 21 May 2022 (UTC)

They're damned either way there—if they dont publish the comments they'll be accused of secrecy, if they publish the comments without attribution they're breaching CC BY-SA, and if they attribute the comments they'll upset people who spoke frankly on the assumption that their comments wouldn't be attributed. ‑ Iridescent 13:06, 21 May 2022 (UTC)

I can't remember what was promised when I wrote those comments, but if they'd published them as CC0 rather than CC-BY-SA then I'd be inclined to give them the benefit of the doubt and assume they collected them under CC0. If anyone from the WMF notices this discussion, this could easily have been avoided by collecting the comments on a CC-BY-SA basis and making it clear when people voted that they were making and signing an "optional public comment". Then just copying them with signatures. And to encompass anyone who wanted anonymity, they could have given the option to sign as "Anon". ϢereSpielChequers 13:34, 21 May 2022 (UTC)

Thinking about it, if anything I'd hope it wouldn't even be possible for them to attribute the comments. People are less likely to be honest if they know the WMF are keeping a log of who said what. The WMF isn't exactly noted for being open to constructive criticism; if I thought there was any possibility that in future I might want to apply for a grant from them, I definitely wouldn't want to have my name on their records as the one who said their pet project wasn't going to work. ‑ Iridescent 13:28, 22 May 2022 (UTC)

I see your point, but I was approaching this from the issue that they were publishing these unattributed comments under CC-BY-SA. WMF pet projects come and go: Threads, Image filters, knowledge engine, strategy (two unrelated ventures a decade apart). The WMF people and pet projects of five years time may have little connection to current projects, and while they rarely respond well to criticism of current pets, their very American hire and fire practices limit their institutional memory. ϢereSpielChequers 16:07, 22 May 2022 (UTC)

Agreed. Even if the people reading over the comments hypothetically weren't in charge of grant funding, what if they leaked your username to those who were, intentionally or not? I dream of horses (Contribs) (Talk) 18:05, 22 May 2022 (UTC)

Unless I've radically forgotten what I wrote (I mentioned the "ethnicity" line), mine isn't there. Clearly they didn't like it. Black Kite (talk) 19:47, 21 May 2022 (UTC)

I can't remember what I put but I can't see anything there which I remember writing. ‑ Iridescent 13:20, 22 May 2022 (UTC)

I do remember what I wrote and it is not there. So that list is either omitting some, has refactered the comments to the point where I dont recognise mine, or the vote was disregarded for some reason. Only in death does duty end (talk) 16:23, 22 May 2022 (UTC)

FWIW, what I wrote is there, though it took me a while to find it. (I would have put more care into it if I'd known it would be published.) Andreas JN466 05:50, 23 May 2022 (UTC)

@Jayen466 (I would have put more care into it if I'd known it would be published.) Yikes! I agree, you should've known that it would've been published, even anonymously. I dream of horses (Contribs) (Talk) 06:20, 23 May 2022 (UTC)

• That's a lot of comments and I can't remember which one I wrote but put me down for any of them (or they didn't publish it). Obviously releasing them is only a clearly pruned but routine exercise in the name of 'transparency'. The WMF's carefully selected drafting committee or any other governing body isn't actually going to do anything about it. A majority pass is a majority pass, just like UK elections and referenda... Kudpung กุดผึ้ง (talk) 00:57, 23 May 2022 (UTC)

Break: I think the WMF are clear in this case

Assuming File:SecurePoll interface - UCoC Enforcement Guidelines Ratification vote.png is a genuine screenshot and hasn't been subject to after-the-fact tampering—and I've no reason to doubt it's genuine—then the "comments" box on the voting form was labeled "Comments will be summarized and posted on wiki without identifying information". Assuming that's the case, they do make it clear both that the comments will potentially be published, that what's posted will be a summary rather than the exact wording, and that whatever's republished won't be attributed. The WMF may have a lot of faults but this doesn't seem to be something over which to attack them. It looks to me like the only issue here is one of good design practices given that numerous people, all of whom have significant experience with the Wikiverse and with the WMF specifically, all managed not to notice the caption despite it being fairly prominent. ‑ Iridescent 10:20, 23 May 2022 (UTC)

• Excellent point. Thank you. --Andreas JN466 12:17, 23 May 2022 (UTC)

  I agree it was collected CC0, but it was then published CC-BY-SA. Call me a pedant, but this isn't the first time that content has moved between those two statuses in ways that imply those two copyright licences are more similar than the legalese might indicate. ϢereSpielChequers 19:50, 23 May 2022 (UTC)

  We (collectively) don't want to be strict about material that's actually in the public domain being marked at CC BY-SA. Wikipedia is full of material taken from public domain sources but marked as CC BY-SA, and in general even when the page is marked correctly that it contains PD text, it's not at all clear which parts are PD and which still need attribution.

  And those are on pages where there's at least nominally a commercial implication as material from mainspace articles gets reused in print where attributing correctly is a huge nuisance. With regards to this vote summary page, it shouldn't make any difference since other than the complete-mirror content forks (which I'm not sure even exist on meta) nobody's ever going to re-use it. ‑ Iridescent 04:14, 24 May 2022 (UTC)
  Disclaimer as on re-reading this, I appreciate it makes me look like I'm advocating ignoring legitimate copyright concerns. Attribution is essential for Wikipedia to function—it's deeply annoying when other people pass their work off as their own—but there needs to be a de minimis point. Strict compliance with our own attribution rules would mean a Herculean task of going through every single page with any kind of "this article incorporates material from…" tag and clearly marking up which parts were verbatim from 1911 Britannica (or whatever), which parts were paraphrased, and which parts were sourced elsewhere. I do not volunteer to do this and I can't imagine anyone else doing so either. ‑ Iridescent 15:11, 24 May 2022 (UTC)

  I think what threw me off was the word "summarized". I sort of envisaged them summarising what people had said, rather than quoting them in full and verbatim (which is what they seem to have done with what I wrote, and given the length of some of the other comments, seem to have done with what others wrote too). Similarly, at meta:Universal_Code_of_Conduct they said (and still say), "A summary of comments will be posted when available." But anyway, no big deal. Andreas JN466 17:24, 25 May 2022 (UTC)

  I agree with you in the interpretation of "comments will be summarized", but "we won't post every comment but we'll post a representative sample" would be a valid meaning of summarized. As you say, this really is no big deal; there are a lot or reasons the WMF deserve criticism but "slightly confusing wording on a form" is really not up there.

  FWIW, if you still run Wikipediocracy can I take vague exception to "Iridescent's talk page is a great place for people to have discussions about how Wikipedia is going to hell, but no one there actually seems engaged in doing anything about it except complaining."? The reason myself and NYB keep being asked questions despite our having been essentially inactive for years is precisely because we have a fairly decent track record of coming up with concrete proposals rather than just saying "something must be done". (In the particular case of UCOC, my first draft at squaring the circle would be "Declare its literal wording a guideline rather than binding policy, but have a committee comprising stewards, T&S staff and representatives from the big wikis with the authority to make it temporarily binding on individual projects against their will should credible concerns be raised about the editing environment of that project and the steps being taken to address it".) ‑ Iridescent 18:02, 25 May 2022 (UTC)

  Pot and kettle. --Tryptofish (talk) 22:47, 26 May 2022 (UTC)

  To be fair, I imagine their collective answer would be "the solution is to close it down". My position is different, that of "Even if we shut it down it would just continue under a different name, so the job of critics is firstly to think of ways it could be improved, and secondly to think of how to put those proposals into action". I at least try to make concrete suggestions rather than just kvetch about the Good Old Days. ‑ Iridescent 19:24, 27 May 2022 (UTC)

  I haven't logged into Wikipediocracy in five years or so, though I still read the site quite regularly. (After Kohs had my wife thrown out the second time, my time there was up, really.) Andreas JN466 18:17, 25 May 2022 (UTC)

  As for your suggestion, I think it's a good one. Much more poportionate. Note that Meta pages have now been opened to discuss changes both to the UCoC and to the Enforcement Guidelines:

  • meta:Universal_Code_of_Conduct/Policy_text/Revision_discussions
  • meta:Universal_Code_of_Conduct/Enforcement_guidelines/Revision_discussions Andreas JN466 13:03, 27 May 2022 (UTC)

  Ironically of the diehard critics Kohs was probably the one I got on best with. I do think he was slightly different from the norm—most of the critics are just people grumpy that at some point someone didn't do exactly what they wanted, but he was genuinely treated appallingly by the WMF, and demonized for his "noindexed incubator space for potentially COI content that can then be vetted by neutral third parties" proposal that turned out to be pretty much verbatim what the WMF themselves adopted a decade later when they introduced draftspace. I can't blame him in the least for being bitter about it. ‑ Iridescent 19:30, 27 May 2022 (UTC)

They quoted me in their report, and they butchered the quote

I was comment 234 on the list of comments. They included me in their Feedback report. I believe I was supposed to represent the most extreme end of negative feedback. I will post my comment below for reference, and I'll highlight in yellow the portions that they DELETED from my comment before they included it in their report:

The Foundation displayed a gross lack of respect for the community, and a gross lack of understanding of the community, in trying to push through a Code without bothering to seek community approval of the Code. The enforcement guidelines are irrelevant, and revising the guidelines won't change my vote. The process to create the Code was botched, the Code itself was botched. If you want to try again you need to start from scratch, letting the community develop something new and seeking consensus for it.

Here's a link to the revision as-they-posted-it. It is ironic and sadly predictable that the three deleted portions are EXACTLY the three most important points of my comment:

1. They deleted the key point about refusing to submit the Code itself for community approval. Their edit arguably misrepresents me as blindly hostile to producing any code of conduct at all. They turned me into an imaginary strawman for unconstructive blind opposition.
2. They deleted that the Code itself was botched. They arguably misrepresented me as just making frivolous whines against a perfectly good Code. Again this edits my comment into a strawman.
3. They deleted the solution - seeking consensus (on a new Code). I don't think this is so much misrepresenting me, but rather reflects a failure mentally to process "seeking consensus" as a meaningful part of the sentence.

The Foundation's biggest problem has always been a pathological unwillingness or inability to acknowledge or engage the concept of consensus. They literally deleted it twice when they "quoted" me for their report. They also have a significant pathology regarding the concept that any pet product could be so badly flawed that they need to scrap it and potentially start again. That was the middle deletion.

I presuming their rationale for the deletions was was to trim for length, so I re-edited the quote. I put all of the key points back in, while ensuring the new text is strictly shorter. Alsee (talk) 08:47, 16 July 2022 (UTC)

Thanks. Filed as another wonderfully illustrative example of WMF "sort of telling the truth and crossing their fingers at the same time and hoping that people are not too forensic ..."

Iridescent is right, this approach towards the truth is just ingrained by now, second nature. Andreas JN466 10:15, 16 July 2022 (UTC)

Wow. My comment about the Code was that I opposed it because I didn't trust the Foundation. I didn't think that off-the-cuff comment -- I only learned of the vote the last day it was open -- would prove so insightful. -- llywrch (talk) 17:20, 21 July 2022 (UTC)

I opposed it because I didn't trust the Foundation sounds about right. I haven't been following this in great detail, but my general attitude towards the WMF has become so jaded, I now assume that even the most innocuous proposal is part of a hidden agenda to impose control. In the last few years I've seen little to change my mind. ‑ Iridescent 10:19, 19 September 2022 (UTC)

Words, words, words

Does anyone know of an Wikipedia essays about the problem of etymology and "rational" ideas of what words mean?

I have found Wikipedia:The problem with elegant variation by Popcornfud, which I think is the closest.

I was hoping for something like Wikipedia:Lies Miss Snodgrass told you, except that not even Miss Snodgrass would have told you that Chronic fatigue syndrome means that you're tired all the time, or that just because people used a word one way doesn't mean that it still has that meaning and the same connotations one or two centuries later. Miss Snodgrass might have her limitations, but even she knows that kids these days don't say unbelievable to indicate that they don't believe it, or fantastic to indicate that there's fantasy involved, and she definitely knows that spinster no longer means someone who spins yarn, regardless of age or marital status.

Wikipedia:Queen Elizabeth slipped majestically into the water is delightful but not apparently relevant. What have I missed? WhatamIdoing (talk) 02:34, 11 August 2022 (UTC)

Not a Wikipedia essay, but I think etymological fallacy may be relevant? Caeciliusinhorto (talk) 06:08, 11 August 2022 (UTC)

There is a learned aside mentioning semantic drift in Wikipedia:Red_flags_in_edit_summaries. That's all I found in essays. A nice (!) example of semantic drift is nice. Andreas JN466 10:57, 11 August 2022 (UTC)

@WhatamIdoing Perhaps, if you can't find the essay you need, you can write it yourself, if you want. I dream of horses (Contribs) (Talk) 17:41, 12 August 2022 (UTC)

Writing it means that you have to choose a title, and mw:Naming things is hard. I open the bidding with Wikipedia:The twin temptations of perpetrating etymological fallacies and ignoring semantic drift. That's awful enough that anyone here ought to be able to offer a better option. (Awful is another one of those words.) WhatamIdoing (talk) 19:27, 12 August 2022 (UTC)

@WhatamIdoing Writing it means that you have to choose a title, and mw:Naming things is hard. No advice, but this is relatable. Heh. I dream of horses (Contribs) (Talk) 02:37, 13 August 2022 (UTC)

There's a similar discussion at Wikipedia talk:No original research#Lexical cohesion in sources. WhatamIdoing (talk) 19:20, 26 August 2022 (UTC)

This was something Eric Corbett was always interested in—you might want to poke him to see if he has any input. (Although he's currently blocked, I imagine he still keeps a vague watch on his talkpage.) The example he used to use was an old description of St Paul's Cathedral as "awful and terrible" (that is, awe-inspiring). If you want live, current examples where you can see semantic drift in action, I'd nominate 'rape' as an obvious example—The Rape of the Lock makes no sense without knowing the way the meaning of the word has changed recently. I'd also say the numerous words and phrases that mean different things in different varieties of English are an obvious one to bear in mind ("Everyone invited to the ambassador's reception must arrive in their best fancy dress"). ‑ Iridescent 10:19, 19 September 2022 (UTC)

Touching Base

Hi, I just wanted to let you now that I am really sorry for all the problems that I have caused and am doing my very best to follow the Wiki rules and feel really bad if I have frustrated you in the past. Whenever someone tells me now that I have done something wrong like added a copyright link ect. I will not question it and will remove it. I also have not been editing other people comments anymore like I have been told not to. I want to be on good terms with my fellow editors and administrators and will do my best to make sure that I do so. Davidgoodheart (talk) 17:17, 31 August 2022 (UTC)

No need to apologise. Just remember (a) that Wikipedia has fought very hard to fight off its reputation for inaccuracy so we take reliable sourcing and fact-checking very seriously, and (b) that when you're writing anything that mentions anything biographic you're writing about real people with real lives and real families so you have both a legal and an ethical duty to do everything you can to be both accurate and balanced in anything you write. ‑ Iridescent 10:18, 19 September 2022 (UTC)

Mood

Is everyone a bit more irritable than usual this week? Or is it just the pages I'm reading? WhatamIdoing (talk) 02:38, 3 September 2022 (UTC)

@WhatamIdoing I haven't edited enough to tell the mood on-wiki; however, off-wiki, IRC and discord seem normal. I dream of horses (Contribs) (Talk) 03:50, 3 September 2022 (UTC)

WAID, I'm not sure that it's just this week, but I'd say for the past couple of weeks, yes. I'd also say that this has been a steady and mounting trend for a few years. --Tryptofish (talk) 22:06, 3 September 2022 (UTC)

@WhatamIdoing I think it's probably more to do with the community's exponentially increasing distrust of the WMF and its employees - fund raising based on deliberately misleading claims, dubious use of funds, a BoT election that is underway, and a community appeal to the Foundation that is likely to be one of the most heavily subscribed campaigns in Wikipedia history. Kudpung กุดผึ้ง (talk) 07:06, 4 September 2022 (UTC)

@Tryptofish, early in the pandemic, it felt like people were really stressed, but we had an explanation, so we expected some of it. Right now, the only explanation that I can find is more like it's hot outside, so we're all irritable. (This isn't completely unreasonable; violent crimes go up during heat waves. Maybe the prevalence of snippy comments increases, too.) Also, there's a change in [my perception of] the tone: we used to be struggling; now we seem irritated. WhatamIdoing (talk) 19:44, 4 September 2022 (UTC)

@WhatamIdoing, @Kudpung, @Tryptofish: I know that I've noticed an uptick in people IRL staring at me. I'm not sure if it's related to the pandemic or not. It probably changes my perspective on tone on Wikipedia. I dream of horses (Contribs) (Talk) 20:00, 4 September 2022 (UTC)

I could certainly go ballistic over WMF, but I wasn't even thinking of that. More like the trend towards editors just not respecting one another. I think there's an accumulation of real-world irritations – economic, political, meteorological, viral, and so on – that put people on edge, and it accompanies a growing willingness of editors to feel like they don't want to give one another the benefit of the doubt. (A few weeks ago, I gave the finger to a motorist who cut me off while I was crossing the street. At 66, I don't think I've ever done that to anyone before, and I was quite surprised at myself.) --Tryptofish (talk) 22:34, 4 September 2022 (UTC)

We've just had one of the best months at RFA in recent years, overall editing levels are holding up and at least where I am it is no longer "far too hot". Yes the board elections are going on, we have an incident where a WMF staffer has been blocked, and the NPP petition has topped 400. But compared to Framgate or various other incidents I can think of, there isn't a storm in this teacup - more like irritating drizzle. ϢereSpielChequers 14:04, 5 September 2022 (UTC)

I'm looking forward to it no longer being far too hot. We've reached 40 °C (104 °F) a couple of days recently. Today's down a smidge, and tomorrow is expected to be down even further, and we might reach normal end-of-summer weather by the end of the week.

Tryptofish, I had to drive somewhere today, and I wasn't irritated at any of the pedestrians or the other drivers. However, I point out that the car has air conditioning and I had it turned up high, so it was a welcome break from the heat. (Hopefully, I didn't irritate any of them, either.) WhatamIdoing (talk) 01:47, 7 September 2022 (UTC)

So that was you whom I flipped off! (joke) --Tryptofish (talk) 19:06, 7 September 2022 (UTC)

That's the problem, right? Because the driver who cut you off probably didn't see you, and therefore probably also doesn't know that it's only your vigilance that kept "a normal trip" from becoming "a couple of hours of repeating 'But, honest, officer, I didn't see him...'." So it could have been me, or someone like me, and I'd never know it. And that means that I don't know how many other vigilant drivers prevented me from causing a car wreck. I only know how many times I prevented them from hitting me. It's a system that makes you think you're an above-average driver and that everyone else is intentionally discourteous, even when both of these are statistically unlikely. WhatamIdoing (talk) 21:36, 7 September 2022 (UTC)

She definitely saw me, because she gave me an exasperated look and pointed at the green turn light. Unfortunately, it lit at the same time as the pedestrian walk light, and she ignored the sign that says "yield to pedestrians". I was at the back of a group of pedestrians, and the driver was trying to make the turn before the light turned red. And it was a very hot day. --Tryptofish (talk) 22:38, 7 September 2022 (UTC)

I think I'm losing patience with comments that declare that some unspecified group of editors should do work that the commenter could do, but is obviously not doing. WhatamIdoing (talk) 02:40, 7 September 2022 (UTC)

Hi WhatAmI, I'm in England where we have an interesting relationship with the weather gods. Traditionally the heavens open as soon as a hosepipe ban is declared. So we've had a few cloudbursts in the last week or so. ϢereSpielChequers 15:59, 7 September 2022 (UTC)

Since you're from that area: Why's a dedicated section of a street called a "bus gate"? WhatamIdoing (talk) 21:37, 7 September 2022 (UTC)

The one that used to be near me was a scheme to relocate a traffic jam from a stretch of narrow single lane road to the preceding wider road that had a bus lane next to it. The buses could flow freely along the bus lane beside a long queue of traffic, then flow freely through the single lane stretch up to a major bottleneck junction. There were lights on the general traffic lane besides the end of the bus lane, they held the traffic back there and released it in batches to maintain a short queue at the junction. As a bus user I thought it was brilliant, but the car lobby got it taken out. Sitting in a queue of cars whilst buses whiz past you seems to have annoyed lots of car drivers. I don't know if every scheme called a bus gate follows that same design. ϢereSpielChequers 09:10, 8 September 2022 (UTC)

I don't know the specifics of this incident, but while I know there are some "petrol heads" who oppose any restrictions on car usage, some of the more reasonable complaints come from Amazon, Deliveroo, Curries etc delivery vans that simply want to do their job and know that time is money. I was recently debating this with someone on Canterbury City Council who wants to reduce all main roads in the city centre to 20mph, including through routes like the A28 and A257, and I think I was the only one who gave a polite and considered objection (both mentioning delivery vehicles and pointing out that reduced speed limits also make buses slower) that wasn't just a pro-car rant. Ritchie333 ^{(talk) (cont)} 11:49, 14 September 2022 (UTC)

gate is also an old synonym for "road" (many old roads at least from Scotland down to the Midlands are called -gate), but I'm not sure whether that is relevant here. —Kusma (talk) 09:32, 8 September 2022 (UTC)

Is that actually true (that "gate" is a synonym for "road")? If you trace the routes of roads with the 'gate' suffix, they tend to be roads that led to the old city gates when cities were walled. I suspect that in most cases they were originally called something like "Trongate Street" and lost the 'street' at some point. You can see it particularly well with the roads leading into the old City of London, where the introduction of road signs has frozen street names in aspic at a moment where some of them had lost the 'street' suffix but some hadn't yet—thus "Bishopsgate" and "Aldersgate Street" running parallel. Most of the other cases probably derive from the era when toll roads still existed, and thus there was a literal physical gate to enter the road.

The difference between a bus lane and a bus gate is length. A bus lane is a lane that's either permanently bus-only or bus-only at particular times; a bus gate is a line across the road through which only buses are allowed to pass, so it serves the function of a gate (i.e., a barrier) keeping other traffic out.

On the OP about general mood, my 2c would be that the 'eternal September' meme may be a cliche but is based on experience. Online communities, particularly those with strong connections to academia, do historically go to pieces each September; in most of the Western world that's when new students traditionally start college and when recent graduates start new jobs, so each year there's a sudden influx of people who either think they know everything or who don't understand the rules and social norms, and get on everybody else's nerves. Plus of course the US and UK—the two biggest contributing groups here—have had a summer of constant economic bad news and in the case of the UK a near-complete political breakdown, and what people experience in real life inevitably filters through to their other activities. ‑ Iridescent 10:18, 19 September 2022 (UTC)

Please vote in the 2022 Wikimedia Foundation Inc. Board of Trustees election

Hello hello. I hope this message finds you well.

The Wikimedia Foundation Inc. Board of Trustees election ends soon, please vote. Some of the candidates are worthy of support. --MZMcBride (talk) 14:25, 5 September 2022 (UTC)

@MZMcBride I've already voted for the one person I know. What can I say? Politics is personal. I dream of horses (Contribs) (Talk) 18:19, 5 September 2022 (UTC)

It's hard not to read "Some of the candidates are worthy of support" as "The candidates aren't uniformly awful this year!" It's a nice change, in any case. —Cryptic 18:27, 5 September 2022 (UTC)

I've voted only for the two candidates who have made a commitment to address the volunteers' needs for better technical support. The current board has recently told Wikipedia that technical support is none of the board's business and if the volunteers are not satisfied, they should do the software engineering themselves for free. Hence NPP's massive appeal to the WMF for some proper engagement. If you haven't seen it yet, you're all invited to support it - only if you want to of course. Kudpung กุดผึ้ง (talk) 23:28, 5 September 2022 (UTC)

Perhaps I am being quite dense, but does m:Wikimedia Foundation elections/2022/Candidates actually link to the candidate statements anywhere? CMD (talk) 23:53, 5 September 2022 (UTC)

They're on the individual candidates' pages linked from #Candidates list. —Cryptic 00:01, 6 September 2022 (UTC)

Ah, thanks. I assumed those were userpage links and that the statements would be after the text saying "candidates will respond". CMD (talk) 00:08, 6 September 2022 (UTC)

I missed taking part in this, but I somehow doubt anything I did would have made the slightest difference. As far as I can tell we could elect a pigeon to the WMF board and it wouldn't make the slightest difference. ‑ Iridescent 10:18, 19 September 2022 (UTC)

Way of displaying multiple images in TFA blurb either randomised or as slideshow

Dank suggested to me you might remember when we have done the above. This regards the impending TFA rerun of Elizabeth II Monday next as a way of satisfying both those who prefer a formal image and those who want an informal one. Many thanks. Wehwalt (talk) 13:55, 12 September 2022 (UTC)

{{Random item}} works fine for logged-in users. I don't know offhand how often it will change for the cached views non-logged-in users get. —Cryptic 15:29, 12 September 2022 (UTC)

Actually, hm, it doesn't seem to update very often - or at all - even for logged-in unless the page is purged. (Maybe someone clicked the purge link on the template documentation page between my first two views?) —Cryptic 15:39, 12 September 2022 (UTC)

I played with it and it looks like you are right. Other possibilities?--Wehwalt (talk) 16:00, 12 September 2022 (UTC)

One possibility would be to create a video file that looks like a slideshow. In other words, select the still images that you want to include, then create a new image file in a format that supports video, in which the first image is shown for however many seconds you want, and then the second image is shown, and so forth, and then it loops back to start over again. Help:Creation and usage of media files and the pages it links to has some information on what kinds of file formats to use; my impression is that WebM might be a good choice. I realize that this might take some effort, and I'm afraid that I'm not at all skillful in this area, although I'm pretty sure that Village Pump Technical could help. (It's also possible that a limitation might be that the reader will have to click on the image to start the video running.) --Tryptofish (talk) 19:10, 12 September 2022 (UTC)

I looked at Commons:Commons:Video, and particularly Commons:Commons:Video#Looped animation, and perhaps a GIF video would be a better file type. --Tryptofish (talk) 19:26, 12 September 2022 (UTC)

@Wehwalt, super-quick driveby reply but the script from the TFA of Middle Ages should be what you're looking for. All these 'ramdomise' scripts only update when the page is purged, but on the main page that's not an issue since it gets so many readers it's purged hundreds of times a minute. A video clip or animated gif on the main page is a really, really bad idea for technical reasons. ‑ Iridescent 2 03:45, 13 September 2022 (UTC)

I have a talent for really, really bad technical ideas. --Tryptofish (talk) 17:58, 13 September 2022 (UTC)

Thanks. Is doing it that way materially different in user experience than using {{random item}} for the images to be displayed?--Wehwalt (talk) 18:03, 13 September 2022 (UTC)

A bit too late obviously, but I don't think there's any material difference; they both rely on using internal Wikipedia statistics that change rapidly and unpredictably to generate a pseudorandom number.

To answer Tryptofish, the reasons we don't use animated gifs on the Main Page are fairly boring.

The first is an artefact of how browser software handles image scaling. For still images the MediaWiki software scales it to the display size and then delivers it to the reader's browser, meaning only a tiny data packet needs to be sent; for animated images, the whole thing is sent full-size to the reader's browser with instructions to scale it to the correct size, meaning a much larger data packet needs to be sent. This not only means that readers with slow connections will just see a blank space while the image is downloaded and processed, it puts a load of unnecessary strain on the servers when an animated gif is included on a page that gets tens of millions of views.

The second is psychological. One of the main reasons we don't change the design of the Main Page even though everyone agrees it wasn;t fit for purpose in 2004 let alone now, is that readers are used to it behaving in particular ways. One of those ways is "if I'm interested in one of the images, I can click on it to expand it and to find out where it came from". If we use an animated slideshow, clicking the link will take the reader to the video loop rather than to the photo in question, which will just unnecessarily confuse and annoy readers. ‑ Iridescent 10:17, 19 September 2022 (UTC)

Content Assessment

Let me know if this sounds silly or unfeasible... I had an idea to kind of combine the process at Peer Review with A-Class and B-Class to create an entirely new process (this would utilize the existing PR resource rather than increase the burden). This would make A-class an official process, which fixes the issue of differing requirements and the resulting lack of trust in the article ratings. We also already have built in assessment criteria for B-class in talkpage templates currently and a B-class review icon at Template:Icon. This would also allow for a couple of stepping stones... One between GA and FA and the other between C (possibly start articles as well) and GA. I believe we would have three types of reviews, including: All-purpose (current PR), B-Class, and A-Class. The B-class review would be the least strict and would only require the involvement of one editor (more are obviously welcome to join in). A person would list the article for review given that it is at least a start-class article. The reviewer would then review the article in its entirety against the 6 B-class criteria and provide advice to the editor to help them get the article to B-class. There would be no set time period for the improvement of the article as long as improvements are continuing to take place and both parties wish to continue. It would then be promoted to B-class if the reviewer believes the criteria have been satisfied. This would essentially be a less-strict review that would work to improve an article up to B-class. I think it would be helpful for newer editors to get them used to writing without expecting them to put in GA-level work immediately as is often the case of the current PR. The A-class review would require two or more editors to review. A person would list the article as an A-class candidate. Other editors would then review the article against the A-class criteria (formal criteria would be somewhere between GA and FA). The process would be divided into two parts. The first stage would be initial reviews, kind of a workshop per se, where no support or opposes would be declared given that the article is reasonably close to A-class (obviously exceptions would apply). The reviewers would leave comments and work with the editor to improve the article. This workshop would last around 10-14 days or longer if it is constructive. If half of the involved parties (minimum either the two reviewers or one reviewer and the nominator) signal they are ready to move on to the next stage, a coordinator would open a subsection where people could either support or oppose the candidacy with their rationale. Once a sufficient period of time had passed, the coordinator would judge whether or not consensus exists for promotion to A-class. The general purpose review would be similar to how PR is currently, but would focus on articles that are already at B-class or A-class and hadn't been reviewed for a while. What are your thoughts on this and changes we could make to content quality assessment? Noah^Talk 18:32, 19 September 2022 (UTC)

Another change I think that could be worth looking at is having a required user right to promote an article above B-class in order to prevent articles that aren't worthy of a particular status from being promoted. Obviously this would be an issue for GA and A class. GA has honestly been a joke in quite a few cases and many articles are promoted that shouldn't be due to inexperienced reviewers. I think a reviewer should have to request assistance if they haven't been granted the user right, which would be given to those who have written multiple quality works and/or conducted thorough, quality reviews. I believe it could be accomplished by transcluding article class parameters from another page that has a special page protection. This would serve as a safeguard to ensure that articles aren't promoted when they don't meet the criteria. Given that there would be numerous people with this right, the workload on any individual person checking the work of reviewers would be low, unlike that of FAC Coordinators. What would you think about this? Noah^Talk 23:26, 19 September 2022 (UTC)

@Hurricane Noah, I was going to write a long reply, but I instead want to ask: What do you think the actual purpose of those ratings is? WhatamIdoing (talk) 00:42, 21 September 2022 (UTC)

@WhatamIdoing:The ratings purpose is to show the development/quality level of an article to inform others of how reliable and complete the content may be and if more work is needed to improve it. If we have articles either being raised prematurely to higher ratings (occurs most often with B and GA) or a lack of trust in ratings (A-Class), then they are pointless. It would be deceitful to editors and the reading audience to list articles at higher ratings when they really shouldn't be. Many editors have been ostracized for promoting GANs prematurely. GA can be a joke unless you have a decent reviewer. We should have some safeguards in place like we do at FAC, albeit less strict. A-Class is supposed to be a rating with a quality level between GA and FA, but it's clearly broken considering many wikiprojects don't use it. It's about time we fix the issues we have with the system. Remember that B+ class used to be a thing and was deprecated. In the case of A-Class, it shouldn't share the same fate, but instead be given a purpose that restores trust in its rating. Noah^Talk 01:21, 21 September 2022 (UTC)

@Hurricane Noah, article ratings were created by the Wikipedia:Version 1.0 Editorial Team for the purpose of deciding which articles to put on a CD for offline use. The goal was to have a bunch of the really nice articles, plus some articles that various groups of editors (called "WikiProjects") said felt important to them, plus some articles that were really popular. They eventually ended up with a sort of $quality\times pageviews\times importance$ formula, and then set the cutoff at whatever would fit on the disks (originally, 2K articles; eventually 40–50K articles).

Hope springs eternal, but the 1.0 team is basically defunct. The last release was more than a decade ago. We maintain the ratings out of habit, and a few groups glance at them occasionally (e.g., WP:MED tries to make sure that top- and high-importance articles aren't stubs), but they are really unimportant now. Also, there's an automated system for estimating article quality, which gives editors an objective reference if they think another editor has either over- or under-rated the article. But – and I say this as an editor who has assessed more than ten thousand articles – messing with quality ratings is really not an important use of anyone's time. If you want to do something more useful, I suggest pulling old "refimprove" tags out of articles that contain more refs than they did a decade ago. WhatamIdoing (talk) 01:46, 21 September 2022 (UTC)

It may not be important in your opinion, but I honestly disagree. The way we use the ratings has changed over time and we have issues to iron out that could be useful. One example would be to standardize and emphasize the usage of Needed-Class (instead of having it as non-standard) so notable topics that lack articles will be categorized together and editors can find and create articles for them more easily. This would be easier than maintaining needed article lists at each individual wikiproject. Noah^Talk 02:00, 21 September 2022 (UTC)

Who exactly are "we", and how does that group of editors "use the ratings"? WhatamIdoing (talk) 02:08, 21 September 2022 (UTC)

We meaning this EN wikipedia community. I can say that many people in the weather project use the ratings quite frequently. The most useful thing for me to do would be to draft an official proposal and get feedback on it. Noah^Talk 02:19, 21 September 2022 (UTC)

How do the participants in Wikipedia:WikiProject Weather actually use the ratings?

Imagine that all of the ratings for WPWEATHER were already 100% perfectly accurate and fully updated. Now tell me: How do you use the ratings? WhatamIdoing (talk) 02:29, 21 September 2022 (UTC)

We use them to determine what needs to be worked on and set project goals. Noah^Talk 02:38, 21 September 2022 (UTC)

It is great that you find a use. My experience is different. I have helped set up most of the assessments for WikiProject Germany 15 years ago. I never use them, and I am not aware of any work going on in WikiProject Germany for which they are beneficial. In theory we could look for important articles of poor quality, but in practice we lack the manpower to do anything about them. (There are far too many articles that fall under the project to be manageable). When I write new articles and wait for them to be assessed, the result is rarely more instructive than a random draw from the set {Start, C, B}, even for articles that soon after pass GA without many changes. —Kusma (talk) 14:57, 21 September 2022 (UTC)

 ↑
That. I can confirm that When I write new articles and wait for them to be assessed, the result is rarely more instructive than a random draw from the set {Start, C, B}, even for articles that soon after pass GA without many changes mirrors my experience, including for articles that have gone straight on to pass FAC with no or minimal changes.  ‑ Iridescent 03:57, 22 September 2022 (UTC)

WAID, re article ratings were created by the Wikipedia:Version 1.0 Editorial Team for the purpose of deciding which articles to put on a CD for offline use I don't believe WP:1.0 had anything to do with the FA designation, at least as I understand the history of the FA project. And I'm not sure about GA, as I know nothing of the history there. Just an insertion that doesn't affect the rest of the discussion ... carry on :) SandyGeorgia (Talk) 03:05, 21 September 2022 (UTC)

FA, GA, and Stub ratings were taken from pre-existing projects. The rest were invented. C-class was added much later. WhatamIdoing (talk) 16:32, 21 September 2022 (UTC)

Quality ratings

WhatamIdoing, Hurricane Noah: This might be a controversial opinion, but maybe we need to do away with universal standards for ratings, at least below "good article," and let Wikiprojects come up with their own standards. If Version 1.0 wanted to come back, there's plenty of good and featured articles to choose from to put on an USB stick. When the community evaluates an article for DYK, GA, and FA, it'll provide an opportunity to glance behind a Wikiprojects curtain, anyway. I dream of horses (Contribs) (Talk) 02:53, 21 September 2022 (UTC)

@I dream of horses: That wouldn't really work well because if everyone had different ratings then nobody would understand what they all meant. It would be complicated and people would have to compare and contrast the ratings of different wikiprojects. Noah^Talk 02:56, 21 September 2022 (UTC)

@Hurricane Noah You have a point. Then again, given how inactive most Wikiprojects are, maybe it's pointless to change much. I dream of horses (Contribs) (Talk) 02:57, 21 September 2022 (UTC)

Any proposal to link importance categories with user rights will make it no distance, as the trend has to slowly deprecate these ratings per WhatamIdoing. Wikiprojects are just too inactive, and the existing GA/FA processes are already understaffed. However, by the same token, if a specific Wikiproject wants to set about with an internal system that actually benefits them, there wouldn't be much opposition. This would for example be the utility of A-class, as an internal review by a group of people who presumably are familiar with the topic, before a FAC that invites scrutiny from those less familiar with the topic. CMD (talk) 03:06, 21 September 2022 (UTC)

A-class is supposed to be superior to GA, but many people do not trust the rating because it is done at the project level rather than for the whole site. Our options are pretty much either fix A-class by centralizing the process (using existing resources) and standardizing the requirements or deprecate it as a standard class at the wiki-wide level while allowing local projects to keep it if they wish. Noah^Talk 03:17, 21 September 2022 (UTC)

I couldn't name a single editor who distrusts (proper) A-class ratings. MILHIST is the only WikiProject that uses them, and their process was excellent, the last time I looked into it. If you mean that it's too easy for some newbie or vandal to change the ratings to whatever they want – well, that's the problem with storing them in a plain wikitext page. About once every year or so, I double-check WPMED's higher importance ratings, and I fairly often find that some inexperienced (or hopeful) person has decided that something the group doesn't care about at all is suddenly top-importance or high-importance. But that's not because editors don't trust the rating system; it's because nobody should trust information that can be changed by any passing editor. WhatamIdoing (talk) 03:27, 21 September 2022 (UTC)

The A-class is used outside of MILHIST as a rating, but Im unaware of any other project that has a great assessment process. That's why it isn't generally trusted for other areas. I believe we have room to make improvements to our systems and I will work on something concrete so feedback can eventually be collected. Disregard the user right since it likely wouldn't work anyways. Noah^Talk 03:34, 21 September 2022 (UTC)

A-Class outside of Military History is rare, and likely the result of drive-by tagging without actual assessment (see Talk:York Minster for an article that is classified as A, B and C-class). —Kusma (talk) 15:12, 21 September 2022 (UTC)

Only if it's a newer rating. Several WikiProjects had an A-class project ~15 years ago, and those ratings aren't necessarily updated later. WhatamIdoing (talk) 16:31, 21 September 2022 (UTC)

Yes, these exist, for example Talk:Mega Man 2. But looking at how many of the subcats of Category:A-Class articles are empty or contain only Milhist confirms that they are very rare. —Kusma (talk) 16:45, 21 September 2022 (UTC)

Streamlining quality ratings

For what it's worth, I favor only three ratings for articles: Start, Good, and Excellent. Start would be any article not checked for accuracy and other deficiencies; Good would be an article certified by a reviewer to be accurate and reasonably complete; Excellent would correspond to the FA rating with all the bells and whistles of an FA. The top priorities of the rating system should be accuracy and neutrality.

The length of an article should not be the major factor in the rating. An article of 300 words might be adequate for some subjects. 10,000 words may be needed for a few topics. Some articles might be deemed too long to get the highest ratings. Smallchief (talk)

With regards quality assessment, I agree 100% with Smallchief. To quote myself IMO the whole setup needs a complete rethink, and we should seriously consider scrapping the importance scale altogether and replacing the ridiculously unintuitive S–S–C–B–G–A–F quality scale with "inadequate–adequate–excellent".. I'm in complete agreement with what WAID says above. The micro-distinctions like Start→C-class made sense of our needing to decide what warranted inclusion in Printed Wikipedia, CD-ROM Wikipedia, and Internet-in-a-Box Wikipedia, but now those concerns are (effectively) irrelevant, it just seems to be a major time sink (add up all the time editors spend assessing and reassessing and it stacks up quickly). Wikipedia is there for the readers, not the editors, and readers don't care about the internal micro-gradations.

With regards importance assessment, I make no bones of the face that I think it (along with the ludicrous WP:Vital articles) should be deprecated altogether. As I've always said, the idea of a "high-importance article" is missing the purpose Wikipedia serves for most readers, which is to act as a backstop covering those topics which don't get good coverage elsewhere. That is, the highest importance articles are those for which Wikipedia is the top Google search result, not the articles on topics which get the most readers and still less the articles which some group of people have arbitrarily decided are "high importance".

Article assessment also has a WP:BITE side-effect that isn't acknowledged enough. Quite often it's the first thing new editors see of Wikipedia's back-office functions. I don't think the WMF has ever done any research on this particular aspect of editor retention, but we know empirically that quite a lot of new editors disappear after they create a new article for the first time. I find it hard to believe that at least some of those cases aren't the result of them working hard to do the best they can, and the only feedback they get on their efforts being an impersonal "This has been rated of low importance". (To quote Newyorkbrad in 2015, "Almost my first, newbish substantial contributions to Wikipedia was the biographies of the federal judge in Puerto Rico under Woodrow Wilson (Peter J. Hamilton). A few months after I wrote it, a wikiproject came along and tagged my article "start class and low importance." Those were both accurate ratings, but it was fortunate that I was already invested in the project by that point. I've often wondered whether, if my initial contribution had been tagged "start class and low importance" the day I wrote it, I would have wanted to write any more.". I agree with this entirely.) Sure, we need quality standards and we need a mechanism to tell newer editors "it is important that you comply with our quality stamdards", but I definitely don't believe the existing massive article assessment meat-grinder is the best way to go about it. ‑ Iridescent 03:47, 22 September 2022 (UTC)

A simple way to slightly improve on the biteyness without any measurable impact on anything else would be to just change all "low-importance" to "mid-importance". The low-mid distinction serves only to bite newbies and has no known practical use in projects looking after more than a hundred articles. Dropping the importance ratings altogether would probably also have far more benefits than downsides. —Kusma (talk) 08:48, 22 September 2022 (UTC)

The importance parameter is meant to indicate importance to a particular group of editors, e.g., Eleanor Roosevelt is important to people who care about women's history but not so important to people who care about New York. It was the invention of the 1.0 team, and the purpose was to make sure that small groups (e.g., people interested in a small country) could mark an article as important to them even if the article wasn't popular in terms of page views, and that would boost the chances that an article on a fairly basic subject, like History of Andorra, would be included in the set, despite getting fewer than 100 page views per day.

The easiest way to reduce the offense, I think, is to rename |importance= to |priority=, which is the alternate, and under-used, parameter name in the banner template (and, I think, technically incompatible, at least back in the day, instead of being a normal alias for the same thing). This would require some re-jiggering of templates and categories plus a major bot run.

Maybe the bot could also add stub-class assessments while it was at it. The ORES ratings for stubs are quite reliable. WhatamIdoing (talk) 16:41, 22 September 2022 (UTC)

Yes, that is the intent of the importance/priority parameters. My point is that I am not aware of any evidence that their existence is actually helpful to Wikipedia editors. —Kusma (talk) 17:53, 22 September 2022 (UTC)

Well, it was helpful back in the day, but presumably you mean helpful to someone editing during the present decade, rather than assembling a list of articles in 2008.

WPMED did use the importance parameters some years back as part of an effort to choose the ~100 most common/significant medical conditions (e.g., Asthma, Cancer, Tuberculosis), and then to improve those past the Start-class stage. This could have been done just as easily merely by making a list on a wiki page, however.

I use it indirectly now as a way of finding potentially promising editors (using RecentChangesLinked), but that's really dependent on a list posted to a wiki page, rather than using the rating directly. The importance/priority ratings could be deleted without affecting what I'm doing. WhatamIdoing (talk) 18:54, 22 September 2022 (UTC)

Some years back, some of the student programs tried to use article quality assessments. The idea was that you find an article, expand it, and then editors would notice your efforts and change the assessment. The problem was that you can improve an article in really important ways without the assessment rating changing at all. The Wikipedia:WikiProject Military history/Assessment/B-Class checklist would probably have been more useful to them than a cursory "Still looks like C-class to me" kind of review. (Now these projects have their own checklists, so they don't try that any longer.) WhatamIdoing (talk) 05:02, 24 September 2022 (UTC)

"Inadequate-class" and bot assessment

I think inadequate would also be BITEY since it implies substandard, poor work. I would recommend "Average", which largely reflects how the articles fall (most are below C). I think we should abandon the bottom part of the scale (remove start and stub entirely), but keep more assessments towards the top (rename of course) in order to help newcomers more easily move their work between the major steps. It might be motivating to a newcomer to see their work move up a tiny bit on the scale and change ratings. I will be working on a plan to help us move forward. Noah^Talk 12:14, 22 September 2022 (UTC)

The median assessed article is a stub. If you want to be more precise, of the pages assessed as articles (e.g., not lists, not redirects, not unassessed, not dabs, etc.), then:

• almost 56% are stubs,
• 34% are Start-class,
• 7% are C-class,
• 2.5% are B-class,
• 0.6% are (currently listed) GAs (former GAs are generally B-class, but aren't tracked),
• 0.1% are FAs (ditto for former FAs), and
• 0.03% are A-class.

(Numbers do not sum to 100% perfectly due to rounding.) Just a hair under 90% of articles are stubs or starts. WhatamIdoing (talk) 16:36, 22 September 2022 (UTC)

It should be noted that the ratings reflect the state of the article when it was last assessed, which often is ten years ago. —Kusma (talk) 17:57, 22 September 2022 (UTC)

Yes. In addition to wishing for an ORES-using bot to add |class=stub to tens of thousands of unassessed articles, I also wish for it to update |class=redirect and |class=disambiguation, and to produce a list, for manual review, of anything that is two classes off in the rating (e.g., a stub that should be marked C or vice versa). There is no need for human review on some things, and we should have a bot doing that work. WhatamIdoing (talk) 18:57, 22 September 2022 (UTC)

(You presumably already know this, but it's worth putting in black and white.) There are good reasons to be very, very cautions about any kind of mass bot assessment run. When you're assesing 6,826,961 articles, then even if ORES is 99% accurate—which it isn't—then that means 68300 incorrect assessments; 68300 people potentially hurt and upset that their magnum opus has been flagged as 'start class'; 68300 people sneering that some obvious hoax article or piece of blatant spam has been rated as well-written and high quality; 68300 people whining to anyone who'll listen about how the WMF isn't just figuratively but literally run by impersonal drones more interested in the process than the product. A bot run of every page on the project—which by definition is going to flash up repeatedly on every single editor's watchlist—is a very good way to haemorrhage goodwill very quickly if not done perfectly, and the WMF does not exactly have a stellar record for getting things right on the first attempt.

I do question the point of the whole concept of 'stub' and 'start'. Back in 2006 they made sense, when the notability bar was higher and it could reasonably be assumed that when a page was only two paragraphs long, it was because it was incomplete. Nowadays we have (literally) hundreds of thousands of pages like Bray Cove Halt railway station where policy or custom dictates we have a stand-alone page, but there's literally no way it could ever be expanded. A useful article quality matrix nowadays would focus much more on "what is the potential that this article could be improved?", and that's something that can't be coded. If we're going to move to an ORES-based model, then to me the classes would be 'very high', 'high' and 'standard' (or in the old money, FA/GA/everything else), with ORES scanning the 99% for those things which are high enough quality to justify a human spending their time reviewing them. ‑ Iridescent 04:10, 24 September 2022 (UTC)

I've spent some time with the ORES classification work. The only three classes I would trust for a bot are the three that I named. In particular, ORES tends to (in my opinion) slightly over-rate articles that contain large numbers of references and/or lists. (I'll call them biggish stubs, but ORES will call them smallish Start-class articles.)

Because the margins between the classes are porous, I've found it more effective to ask for a two-class discrepancy if I want to manually review things. The difference between a high-end Start-class article and a low-end C-class article is not important, but if the talk page says stub and ORES says C, then it's never yet been a stub when I look at it. WhatamIdoing (talk) 04:56, 24 September 2022 (UTC)

I've probably said this here before, but when there was a de-stubbing contest some while back, I found about 30% or more of the articles I looked at for it weren't stubs at all. Johnbod (talk) 05:22, 24 September 2022 (UTC)

You inspired me to click on Special:RandomInCategory/Stub-Class medicine articles 10 times. I found one redirect, six that ORES rated as Stubs, two that it called Starts (one I agreed with; the other is on the border between Stub and Start), and one that it called C-class, which I weakly agreed with. That seems to align with your experience pretty closely. WhatamIdoing (talk) 05:39, 24 September 2022 (UTC)

I would not oppose ORES based bot runs, but I think a complete rethink of the system is a better use of resources. —Kusma (talk) 11:00, 24 September 2022 (UTC)

Impacts on editing

Going to the next level, if we were to hone in on the three-tier classification (excellent-good-start, roughly equivalent to FA-GA-everything else), what would be the impact on editor behavior? Would it inspire more editors to help clean up or clean out the GA/FAs that aren't (at least a third of FAs and probably 2/3 of GAs)? GA quality has not kept up without Geometryguy and Mally, for example, while a few valiant others (eg Ealdgyth) hang in there. Would it inspire more editors to work more articles towards FA/GA? I suspect we'd see more editors wanting to move more articles to the easier GA but not more to FA, and since GA classification is spotty enough already, depending on the reviewer, would GA be overwhelmed? I don't think FA processing would be hit. If more people engaged FAR, that would be grand.

Thinking of my own area of editing, I'd be most happy to see all of the B-class medical articles (that aren't) sent to that lower group, as we'd have a clearer picture of how bad or dated most medical content is, and I suspect we'd see more editors push for GA there, meaning a smaller set of inflated ratings.

A (side effect) use of article assessment is here (scroll up; that editor was disagreeing on medical content with all those other medical editors and appealing to authority ... oh, and that editor is now community banned I see).

If I look at my own top edited articles, the three-tier scheme would send the huge former FA Hugo Chavez start class, which makes no sense, so it seems another name would be needed for the "everything else". Other than that, there would be no impact on my editing, as I won't put articles I've written through GA. Venezuelan presidential crisis (who cares) stays at the lower tier, so does the no-longer-a-thing Asperger, and there's nothing there worthy of working up to FA class.

My question is whether moving to a three-tier system would result in overwhelm anywhere in the Project. SandyGeorgia (Talk) 12:16, 24 September 2022 (UTC)

@SandyGeorgia: Please see here for what I have been working on. I don't think a three tier system would exactly be friendly to the newcomers since it is removing assessment criteria and levels between the low end and the high end. It's a work in progress right now. Noah^Talk 12:49, 24 September 2022 (UTC)

If you wanted to encourage editors to improve articles, then it might make more sense to identify specific things to do, instead of a whole-article assessment. Imagine that instead of assigning an overall class, editors could assess categories:

• completeness of the content (missing stuff–mostly there–has everything)
• writing (needs work–appropriate and understandable–good)
• sourcing (none–needs work–good enough–great)

The ratings would thereby give a hint about how to improve the article.

We could perhaps even merge some of the more boring maintenance banners into this.

This, of course, assumes that we believe editors are motivated by things like ratings and scoreboards. I'm not sure that's true for most content contributors. One of the difficulties in having these discussions is that the editors who talk about editing actually do relatively little content creation, so we're not very good at predicting what matters to the people who add content. We hear a lot of "as a patroller/AFD denizen/rule maker/general busybody, I think other editors should..." but not much "as a person whose primary (or even sole) activity is adding paragraphs of missing information to articles, I..." The end result is that we have don't always have rules made by the editors, for the editors; we have rules made by the busybodies (e.g., me – I am the all-time top editor of WT:EL, WT:PG, VPM, and VPIL, and among the top 10 for many other back-room pages), for the benefit of gnomes and patrollers, (sometimes) against the content contributors and readers. This is "just how it works", but maybe not "how it would ideally work". WhatamIdoing (talk) 04:12, 25 September 2022 (UTC)

Regarding I suspect we'd see more editors wanting to move more articles to the easier GA but not more to FA, I agree. The two are qualitatively different, however, and it's not just a case of one being more difficult. FA is the only part of the internal assessment scales that's reader-facing, owing to the permalinks to Wikipedia:Featured articles and Wikipedia:About Today's featured article on the Main Page. GA, on the other hand, is effectively an internal process (yes we have the green dot, but literally no reader knows what it means or even notices it). An easier GA, as a halfway stage between the general sludgepile and FA, wouldn't necessarily be a bad thing.

I do agree with WAID that the assumption that most editors are motivated by ratings is likely a misconception. Of the editors who explicitly write (as opposed to those who review, correct, format etc) I'm certain the motivation of most is improving coverage of whatever their preferred topic happens to be, and at absolute most the assessment processes are just a mechanism for reassuring them that they're doing it right. FWIW, to the best of my knowledge Wikipedia:WikiProject Visual arts got rid of importance ratings there was literally no impact on editor activity and I'm not sure a single reader even noticed (there were certainly no "I'm confused, why can't I see an importance rating on this article?" comments anywhere). ‑ Iridescent 04:57, 25 September 2022 (UTC)

the motivation of most is improving coverage of whatever their preferred topic happens to be <pipes up to confirm> Yes, I've been inactive for most of this year but that's mostly because a) the stack of work every December to update articles is beginning to scare me and b) I ran out of topics to write about. Jo-Jo Eumerus (talk) 07:32, 25 September 2022 (UTC)

As a content creator maybe I'm typical of some. I don't care much about ratings -- except that I get pissed off when a drive-by reviewer looks at a recent creation of mine in which I have a certain pride and labels it "start" or even "stub." A few times I have complained, and usually found satisfaction. On the other hand I've never nominated an article of mine for GA or FA. It's too much trouble to go through that process. I doubt that I would change my mind with a new reviewing system. So, my advice to reviewers is don't insult the author -- spend more than 30 seconds on your review -- and justify your rating with a comment on the talk page. That comment might be as simple as "good start, but need more information"; or "good info, but too many typos"; or "you need better references to verify what you have said"; etc. It would be helpful to content creators to know how the reviewer came up with his rating. That additional burden on the reviewer would slow down the review process, but I don't care much whether an article I've worked on is promptly reviewed or not. In fact, I don't care whether they are ever reviewed. Better no review than one that is cursory. Smallchief (talk) 10:39, 25 September 2022 (UTC)

The greatest thing about the GA process is that you usually get a reader for your obscure article plus a decent amount of feedback. The green plus is a nice extra. Review processes without feedback are useless to me. —Kusma (talk) 11:40, 25 September 2022 (UTC)

This is my main use for DYK. It's nice to know someone else has looked at your work and found it passable. The review processes turn the solo activity of editing into something a bit more social. CMD (talk) 15:28, 25 September 2022 (UTC)

WAID's idea is intriguing; I doubt we have enough editors to make it work. Distinct, I think, from what both of you are getting at, I submit to the FA process as a way of showing what excellent content in similar topics should be. (When I wrote TS a gazillion years ago, there wasn't a single decent neuropsych article; after I wrote it, we got a proliferation of same from other editors.) An FA can be a roadmap to improvement. I don't much concern myself (obviously) with the 90% of the Project that is mostly unfixable crap that I hope no one is reading, since in the medical realm it can be dangerous. I'm interested in what impact a new scheme would have on encouraging other editors to produce better content, and FAs (when maintained) provide a guide to new editors. If one FA can guide 100 other editors to improved content at any level of assessment, it's worth the effort. Hence, my focus on how any proposed assessment scheme would impact editor motivation and content review. SandyGeorgia (Talk) 13:31, 25 September 2022 (UTC)

@SandyGeorgia I don't much concern myself (obviously) with the 90% of the Project that is mostly unfixable crap that I hope no one is reading, since in the medical realm it can be dangerous. I'm going to burst your bubble rather bluntly, but please know I completely sympathize. I wouldn't want to edit war with Wannabe Doctor Randy from Boise, either.

No one should count on [eople not reading inaccurate information, particularly if they lack 'common' sense. Hence, why there are (albeit rather hidden away) disclaimers about how one should see their general physician if they're concerned about their health. I dream of horses (Contribs) (Talk) 16:34, 25 September 2022 (UTC)

Yep, I tried (quite unsuccessfully) to get a more prominent dislaimer on our medical content. As my sister-in-law remarked when I explained the difference between my Wikipedia activity and other editors: "Louisa May Alcott is not malaria". I can't fix all the medical articles; I can try to show what they should be, and hope that inspires some editors towards cleanup. SandyGeorgia (Talk) 17:15, 25 September 2022 (UTC)

"I don't much concern myself (obviously) with the 90% of the Project that is mostly unfixable crap that I hope no one is reading" That's the crux of the whole problem with Wikipedia at the moment. If you step away from editing or admin areas and just try to use the encyclopedia to find out stuff, you can easily come across an unreferenced article that reads like a personal essay of the subject or contains completely speculative claims. To test this, I predicted The Searchers (band) would have these issues before I went to read the article, and I was not disappointed - except that the "more citations needed" tag was only placed two years ago. This is, incidentally, why I think FAC is something of a fool's errand unless it's on a very obscure topic that not many people would ever want to edit; you'll be forever fighting people who don't know or care about any quality standards on Wikipedia, and probably eventually lose patience and snap .... as Eric knows all too well. Ritchie333 ^{(talk) (cont)} 22:19, 26 September 2022 (UTC)

Ritchie333 we did it at J. K. Rowling quite a few months ago, and it's holding well because the FA star gives an article more watchers hence more clout to deal with disruption, vandalism, POV pushing and random dumbassery. On the other hand, if I look at the medical articles that have been defeatured (all because their main editors left), they just get worse and worse and worse as time passes, because nobody's watching them and no one cares. Constantly trying to keep crap out of what was featured content (like Alzheimer's disease and Parkinson's disease) IMO makes one equally likely to snap, because you have no help as you do with most FAs. Yes, FAC may be a fool's errand, but when some new editor is trying to do something stupid at Alzheimer's or Parkinson's, I can point them at dementia with Lewy bodies to explain how it should be done. Anyway, back to why bother with content assessment because it is 95% crap, I still think we need FAs if for nothing more than an example for other editors on similar topics (recognizing though that even at least a third of FAs aren't). I'm coming around to Iri's three-tier idea ... but wonder how it would impact most editors (we know how it would impact articles: it would be more honest, as we'd be acknowledging that 95% of the project is random junk). SandyGeorgia (Talk) 22:52, 26 September 2022 (UTC)

What constitutes 'editing', & which topics are yet to be covered?

I agree with SandyGeorgia's concern that a more content-oriented/thought-requiring assessment system would be time consuming.

This has also made me wonder about whether the tiny fraction of editors whom we choose to label "the community" at the English Wikipedia does as much content creation as we did when we were new editors. I've always been more interested what happens in the back end, but I have an impression that other experienced editors started off creating content and now edit articles in fairly superficial ways (e.g., adding maintenance banners or fixing typos instead of adding paragraphs and sources. Also, why is adding or removing a maintenance banner still "article editing", when the whole thing should be handled as proper metadata, as proposed about 15 years ago?). I've heard others hypothesize that this drift happens to some fraction of editors, but I don't think there has been any research on whether it's a real effect. It's not really a case of "those who can, do; those who can't, coach", because we can, but I suspect that inexperienced editors are more focused on content creation than those of us who spend our free hours arguing on the back pages. WhatamIdoing (talk) 05:30, 28 September 2022 (UTC)

The drift is two-way. There's certainly a contingent of editors who come here to write about their pet topic, and then once the work there is done drift into behind-the-scenes tinkering; and there's a small but disproportionately influential cohort who start off as 'true' content editors, get some kind of special status (RFA, a job at the WMF, coordinator at one of the still-active projects…) and end up getting sidetracked in to that. I'd argue that in terms both of raw numbers and of impact, these groups are more than counterbalanced by the new editors who come here to correct a minor error or fix a typo, gradually do more and more minor maintenance tasks, and only after a few months develop the confidence (and the skills) to start making significant contributions.

To be honest I don't think trying to measure it would be a particularly productive exercise. Wikipedia editors aren't professionals, and the impact of reality probably has a hugely more significant impact than any number of changes to the editing interface, outreach programs, or tweaks to the assessment scheme could possibly have. Wikipedia has now existed for long enough that those editors who started off as schoolkids goofing around occasionally have become students with a lot of spare time, good access to source materials, and a strong practical reason for a space where they can practice their writing skills and have their efforts critiqued by others without real-world jeopardy; the students have become junior employees with little spare time but a keen interest in keeping involved in a community that distracts them from a grim real life; the junior employees have become married with kids and have virtually no time at all; the newlyweds with kids have seen the kids grow up and leave home, suddenly giving extended chunks of free time in the evenings; the employees who only had evenings and weekends free have become retirees with huge chunks of free time; the retirees who have always been a largely unacknowledged driving force behind Wikipedia have started to succumb to physical and mental deterioration and are no longer to take as active a role as they once did. ‑ Iridescent 11:23, 28 September 2022 (UTC)

LOL. Your comment on the deterioration of the oldsters reminded me of the lyrics of a Toby Keith song. "I ain't as good as I once was, but I'm as good once as I ever was." I'll let you imagine what that lyric refers to. Smallchief (talk) 12:42, 28 September 2022 (UTC)

What I'm about to say about that drift isn't particularly novel, but our needs for the creation of new content have changed over time. For the last several years, our needs for improving bad articles into better ones has come to far outweigh our need for new articles (although new ones will always be needed). The fact that experienced editors have trended towards article maintenance and improvement, from new article creation, is simply a natural progression. --Tryptofish (talk) 18:57, 28 September 2022 (UTC)

I'm not convinced that we don't need new articles. Thousands of notable new films, albums, and books are published each year, in addition to the tens of thousands of older works by artists outside English-dominant countries that we're missing. Each field could presumably say similar things: more than a hundred notable drug candidates enter clinical trials each year, thousands of new professional athletes each year, dozens of notable new vehicles each year, and so forth.

Have a look at pages like Wikipedia:Featured articles in other languages/French or Wikipedia:Featured articles in other languages/German. There are hundreds of subjects that we're missing, but other Wikipedias found worth turning into FAs. WhatamIdoing (talk) 02:01, 29 September 2022 (UTC)

He's not saying we don't need new articles, just not as many as we used to. Never mind the new published/released crap, or young sportspeople, many basic encyclopaedic areas are still extremely poorly covered, just because we happen never to have had strong content editors interested in them. Economic history, Engineering (non-electronic), the decorative arts and so on. As I've been saying for over a decade, "Our “top level” topic articles are generally weak, especially on abstract concepts. Our articles on specific things, whether people, films, ships or species are often very good, and our coverage is usually very wide, even if many articles are pretty short. So the more general and abstract the article subject, the worse our articles tend to be. This is the reverse of the pattern generally found in published reference works. It also means that some of our weakest work within a subject area gets the highest number of viewers. This is not good." My own completely new articles on "old" subjects in 2022 are: Wilderness (garden history), Garden room, Woodland garden, Italian Renaissance sculpture, Jacopo da Trezzo, The Eight Great Events in the Life of Buddha, Life of Buddha in art. That French list is only showing a few French FAs with no English articles, & the ones it does show are pretty dodgy - instead of "Birka et Hovgården" for 2 Viking sites, we have Birka and Hovgården. One shouldn't expect to find the equivalent of Château du Petit Trianon at Castle of Petit Trianon, as any fule kno). We have a couple of screensfull on Petit Trianon. Not "hundreds" of missing articles, I think, maybe a dozen or two (a park in Rennes etc). Maybe more in German - they seem rather strong on cat medical matters - vetinary stuff is another weakness on en:wp. But I'm sure about half of them are just under other names. Johnbod (talk) 02:19, 29 September 2022 (UTC)

I saw w:fr:Renaissance florentine dans les arts figuratifs (~Art in Florence during the Renaissance; Art in Florence is a redirect to Florentine painting) in the list of FAs at the French Wikipedia, and thought of your Italian Renaissance sculpture. We are missing many scholarly subjects.

I think it's easy for a conversation about the relative weakness of articles on general subjects to turn into bashing the work of editors who would rather write about pop culture or commercial subjects or otherwise trying to pull up the ladder, so that today's newcomers don't get to experience the same freedom that brought us into the movement. I plan to die someday. I want another generation behind me. If they want to write about whatever passes for music these days, that's no skin off my nose. WhatamIdoing (talk) 03:17, 29 September 2022 (UTC)

Number of English Wikipedia articles

"We no longer get as many new articles" is a myth. There was (as you'd expect) a spike during the mid-2000's Wikipedia boom, but other than that the net article creation rate has barely wavered; the "total number of Wikipedia articles" line forms the kind of near-perfect diagonal about which statisticians dream.

If anything the de facto creation rate for new articles has risen. Our increased tolerance for stubs means someone setting out to write about a new topic is more likely to be replacing a one-line microstub than a redlink, and replacing a 20-word stub with a 10,000-word article doesn't affect the article count but for all practical purposes is a new article.

(This is to repeat a tune I've played before, but even on the pop-culture topics which Wikipedia is caricatured as over-representing, there are gaping holes, particularly when it comes to topics that dropped out of popularity before the internet age. For some variety from my usual Enid Blyton bibliography example, take Robert Westall; he was a hugely acclaimed author about whom it's reasonable to assume every book he wrote was the subject of prolific commentary, but because he died in 1993 hardly any of his books have articles. Even things like List of works by Vincent van Gogh and Charles Dickens bibliography are a sea of stubs and redlinks.)  ‑ Iridescent 13:25, 29 September 2022 (UTC)

Different editing styles & improving others' work

• Interesting discussion. I often self-assess depending on how long it took me to write the article and how deep into the sourcing I bothered to go; it's quite easy to decide whether the article could readily be improved by a quick internet search and a bit more perseverence (start), or would need a subject expert and a proper academic library (B). I think there's a great deal of merit to encouraging editors to improve microstubs to decent stubs, and stubs to decent start/C, and if ratings help to provide motivation for that, they seem worth it. (If anything I'd finegrain stub classification further; the difference between a couple of sentences and one not-very-reliable source, and something of 1200–1500 characters with 6 or 7 sources, most of which are at least passable, and possibly a relevant image, in terms of use to the reader, is huge.) It's my impression, though, that hardly anyone is bothering to work in that fashion these days. There are folk that create shortish articles (and sometimes get bullied for doing it too much), folk that work hard to get FAs/GAs, and folk that slap templates on things but rarely write a word, but hardly any drive to actually improve other people's work. I agree that having one's content assessed as stub/start & low feels like a slap in the face; a while back I got assessed as C for something where I'd spent weeks trawling the internet and bothering folk at the Resources Requests board to copy me bits of oop books, and was just a teensy bit put out. ORES ratings are pretty slapdash; it always marks mine down because I've said a calculated FU to the reference citation templates mob. Espresso Addict (talk) 01:11, 2 October 2022 (UTC)

  Regarding hardly any drive to actually improve other people's work, in my experience this is generally the hardest thing to do on Wikipedia. Unless the article is either a microstub or in a truly sorry state, improving something that already exists runs a non-negligible chance of antagonizing whoever wrote it and provoking an irate response. Plus, if the existing article is long enough to have a defined structure, it means working within the structure that already exists even if one doesn't consider that structure appropriate (this is a particular issue with biographies, where we generally inherit a strict linear chronological structure even if that's not the best structure for a full-fledged FA-length article, but it also comes up routinely with things like military campaigns). Add on to that that WP:CITEVAR obliges authors to stick with whatever goofy citation format the original author used, even when it's some kind of eccentric format not used anywhere else on the project.

  In my experience, it's usually easier to rewrite the article from scratch in a sandbox, overwrite the existing article, and incorporate any parts of the early article worth saving into the new article. This better serves both the editor and the reader, but decidedly does not represent the Wikipedia vision of collaboration and constant improvement. ‑ Iridescent 16:51, 2 October 2022 (UTC)

Thanks, Iridescent, it's always interesting to hear a different perspective. I don't recall ever experiencing pushback when rewriting articles, but many of my expansions would fall into microstub/truly sorry state, as I tend to find them at the G13, A7, prod, AfD & copyvio queues, or in the unsourced pages heap. I was thinking more of the drive to reference completely unsourced pages and the drive to destub 50k articles, neither of which is exactly oversubscribed, particularly the former, where two minutes' work with the Wikipedia Library will at least move them to a less-urgent maintenance category. Espresso Addict (talk) 00:02, 3 October 2022 (UTC)

The microstubs and unreferenced articles are to me the exception that proves the rule, as their creation is itself a violation of policy so the creators can't reasonably complain. When there's a stable and reasonably long existing article, improving it sometimes provokes howls of protest. (If you really want to see WP:OWN in its purest form, make a substantial change to something that already has GA/FA status.) ‑ Iridescent 05:12, 3 October 2022 (UTC)

I would, however, conversely say that the majority of people who make edits to GAs and FAs don't know or care about the GA or FA criteria. Ritchie333 ^{(talk) (cont)} 14:19, 3 October 2022 (UTC)

I tend only to do relatively minor edits to GA/FAs, though sometimes wade in with TFA when there are error reports. I think one would need to be a real expert to make significant changes when there's a (weak for GAs) consensus for the approved version, and if the approved version is so poor, perhaps it should be demoted first? Though I can see that annoying the original creator a great deal! Espresso Addict (talk) 22:53, 3 October 2022 (UTC)

Reluctance to make changes as a barrier to recruitment

It's good to be reminded of other people's experiences. I often recommend that new editors try to add a sentence, but the Wiki Ed folks tell me that some students find it incredibly stressful to meddle with "someone else's" article, so they would rather start a new article from scratch. I think this is a much more difficult task, but perhaps difficulty is another thing that's in the eye of the beholder... WhatamIdoing (talk) 19:17, 3 October 2022 (UTC)

I often find when editing or expanding a stub or a poor article that it's easier to start from zero and write the article as if I'm creating it and then integrate what, if anything, is worth saving from the original. I've had very few adverse comments from the creators of the articles I have demolished and recreated.Smallchief (talk) 20:33, 3 October 2022 (UTC)

(EC) I've experimented with various uncontentious early edits. The best session was probably one a couple of us ran about a decade ago with some wikimedia donors. We had a list of articles that had UK grid reference and no image, and we showed people how to search commons and add an image. At that time about half the articles on that list could be illustrated from commons. I don't recall any pushback at all to illustrating such articles. ϢereSpielChequers 20:36, 3 October 2022 (UTC)

[Re WhatamIdoing] I wonder if the type of person who likes wading into someone else's work and adding a sentence is going to make a productive long-term Wikipedia editor, while the type of person who finds it too stressful is less likely to convert to regular editing? Though I suppose there are plenty of editors who churn out high-quality articles but rarely interact with other people's work. Espresso Addict (talk) 23:02, 3 October 2022 (UTC)

Maybe?

One of the profs speaking at Wikimania said that one student was so anxious about editing Wikipedia that they submitted an assignment privately, sort of as a "here's how I would have improved the article, if the whole idea weren't terrifying" piece. It's clearly not everyone's idea of fun. WhatamIdoing (talk) 23:51, 3 October 2022 (UTC)

That's sad. I must admit, I've spent a decade and a half trying to convince cajole guilt various of my acquaintance to edit -- all highly educated computer-literate types, often with expertise that's underrepresented in our editor base, several of whom actually have articles -- and I'm not aware of a single person who actually made an edit. In fact, the best I've done to promote newbies editing, afaik, is an (as I recall, rather minor) error in an article about a grade-I-listed building that the owner of the property felt they had to wade in to correct. They were extremely aggrieved, and certainly not converted to the editing experience. Espresso Addict (talk) 00:11, 4 October 2022 (UTC)

My experience tallies with both of yours, in that even those people who would have something useful to add are usually very uncomfortable with the idea of editing. Plus, recent changes patrol tends to be quite enthusiastic, so there's a good chance that any given editor's first edit will be reverted, and anecdotally people whose first edit is reverted don't tend to make a second.

There's also an issue we don't talk about enough but which is definitely there—even though we have the "anyone can edit" link on the main page, most people don't understand what that means. When it comes to people who aren't regulars on user-generated sites, in my experience close to 100% of people I know aren't aware of what that actually means. Even among people who deal with websites or information for a living, I find that if people are even aware the "edit" button is there, they assume either that it's only to be used by those people who've been given 'editor' status, or that it takes them to the page where they can submit suggested edits to Wikipedia's panel of professional editors. (I suspect—and this is pure speculation—that part of the reason for this misconception is that the combination of Recent Changes Patrol, New Pages Patrol, and Cluebot is in some ways too efficient. Because there's such a high probability that somebody's first edit will be inappropriate in some way, and will be reverted within seconds, there's a correspondingly high chance that when the new editor refreshes the page, they won't see whatever edit they made as having gone live, and will assume that some kind of vetting process has held it up.) ‑ Iridescent 04:45, 8 October 2022 (UTC)

Indeed. One the few people to whom I've owned up to being an administrator here assumed that I worked like a proper commissioning editor for an academic publisher, commissioning articles that fit within an outline I'd generated! I think the lack of real-time response is one of the major demerits of punting non-autoconfirmed editors to the Articles for Creation system: if a new editor does manage to submit their precious text successfully (and judging by the number of blanks, many don't), they expect it to go live soon, not get a formulaic rejection full of gobbledegook up to 3 or 4 months later. Espresso Addict (talk) 07:21, 8 October 2022 (UTC)

Obvious declines get declined pretty quickly. Like, within a few hours. It's the more complex cases, and accepted drafts, that wait 3-4 months. I dream of horses (Contribs) (Talk) 07:26, 8 October 2022 (UTC)

Which unfortunately has the effect that the better class of newbie waits longer than those who pen obvious junk. Not the fault of the reviewers, but the entire system is not favourable to welcoming potentially productive newcomers. Espresso Addict (talk) 10:37, 8 October 2022 (UTC)

Yes, that. Quite a lot of Wikipedia's processes have the unfortunate side-effect of penalizing competence; we're much better at addressing things done incorrectly than we are at recognizing things done correctly. To me, the most glaring example of this is the use of block logs as a metric—an editor with 100,000 edits who loses their temper five times goes down as "that editor who was blocked five times" rather than "that editor who made 99,995 positive contributions".

We're also much better at pointing out to newcomers what they've done wrong, than we are at congratulating new editors on what they've done right, and are terrible at telling new editors that they're on the right track and giving constructive feedback as to how they can improve. To the 99.99% of the world who don't speak Wikipedia jargon, the set of article creation instruction pages to which we direct new editors (see right) might as well be written in Latin.

(It's not just the article creation instructions, for which one could at least make the—IMO spurious—argument that we want to discourage people with no experience from jumping straight in. Try getting any reasonably intelligent person you know to read Wikipedia:Contributing to Wikipedia—which is basically the foundational text of all our "instructions for newcomers" pages—and see if they come away with any impression other than "Wikipedia is a weird bureaucracy run by obsessives enforcing a bunch of arbitrary and impossible-to-follow rules".) ‑ Iridescent 16:23, 8 October 2022 (UTC)

Well, it's conveying that important learning effectively then! Johnbod (talk) 16:29, 8 October 2022 (UTC)

That page also leads me to File:Evaluating Wikipedia brochure.pdf, which is credited to "Wikimedia Foundation" so I assume is official. If so, it's something of an insight into how detached our insect overlords are becoming from the peons. (Signs of bad quality: The discussion page is filled with hostile dialogue. If you have the time or the knowledge, please consider correcting the problems yourself by clicking Edit at the top of the article. in particular seems like a recipe for inviting well-intentioned newcomers to jump headfirst into snake-pits.) ‑ Iridescent 16:48, 8 October 2022 (UTC)

Looking at the history, that's from the part of the WMF that became Wiki Education Foundation after Sue Gardner's m:Narrowing focus project, so it would be the 12-year-old views of the people running student editing projects.

I don't imagine that the English Wikipedia would welcome professionally/top-down written documentation on how to edit. Smaller communities probably would, but here, I imagine that we'd be upset about how those so-called experts took out the thing that I like quote to to get my way in disputes. So we'd start off by saying everything can be re-written except the bits that I wrote (which must be treated as holy writ), plus the stuff you wrote (which are lucid and obviously helpful), and then the stuff he wrote (which isn't bad, but it's a little focused on discouraging pop culture and business-related content), and then the stuff they wrote (just to be fair) ...and then we've reënacted XKCD on standards in our basic documentation again. WhatamIdoing (talk) 21:27, 8 October 2022 (UTC)

I'm sure you're right that the community would instinctively reject anything that looks like top-down imposition of instructions; even if the instructions were perfect, there are understandable reasons not to want to set 'the WMF has the power to impose rules' as a precedent.

If I were a WMF staffer tasked with creating a modern-day version of Wikipedia: The Missing Manual, I'd:

1. Post at places like Wikipedia:Help Project and the Village Pumps asking for suggestions;
2. Email people who'd set up accounts and edited for a month or so but then stopped, asking why they'd stopped and if there was any particular difficulty that drove them away;
3. Take a random sample of newly-registered accounts asking them to volunteer to be tracked and to periodically give feedback on their experiences joining Wikipedia;
4. Collate the results of 1, 2 & 3 and use it as the basis for a set of instructions;
5. Post the result of 4 on-wiki, again widely publicising it and asking for feedback;
6. Rewrite it based on any constructive feedback received, and at this point bring in a professional writer with no experience of Wikipedia or similar sites, to ensure it all makes sense to those outside the Wikisphere;
7. Hold an RFC as to whether to give it 'official documentation' status.

Since the only people likely to participate in the RFC would be those who'd participated in the document's creation the result would likely be a foregone conclusion, but community consent is important to avoid the perception of orders being imposed by the WMF. If for any reason the RFC led to rejection, then the document could be rewritten based on whatever the reasons for opposition were, and re-submitted once it addressed those concerns. ‑ Iridescent 05:11, 9 October 2022 (UTC)

If I were the WMF staffer charged with such a project (which is very unlikely, but not completely impossible), I'd tell them that the obvious solution is to beg User:John Broughton to write a new edition. Before we could reach that point, though, we'd have to figure out whether it was meant to be universal (which would suggest writing it at MediaWiki.org and getting translations) or if it was meant to be only applicable to the English Wikipedia.

2 was done some years back. One result was that editors we thought were "inactive" still thought of themselves as being active. A less-fun result is that many felt rejected, excluded, and harassed by what experienced editors would call standard operating procedures (like reverting unsourced content about living people).

3 is done by the Growth team as part of the Newcomer Homepage work.

7 would need clarification: What does it mean for something to be "official documentation"? Does that mean that all ~650 welcome templates here link to it instead of other pages? Or just that it gets a fancy tag at the top, and everyone ignores it after that? WhatamIdoing (talk) 16:49, 9 October 2022 (UTC)

[Re Iridescent]: The thanks button is a boon; it provides a positive metric to balance the block log. I do try my absolute best, when I've got my admin hat on, to AGF as hard as I can & explain patiently to newbies how to improve their articles in my own words, but it's time-consuming, often not productive of any further edits, and even on the rare occasions when the newbie tries to follow my advice, they rarely get it right enough to pass AfC/NPP. I despair sometimes. Women in Red will help out with any article within their purview, but for anything else, the newbie is swimming with sharks. Espresso Addict (talk) 23:00, 8 October 2022 (UTC)

I agree with the concerns above, but I also feel like there's a no-win aspect to this. I'm frequently annoyed to see "Welcome to Wikipedia" messages posted mechanically onto the talk pages of thoroughly not-here accounts, so that the welcome is followed immediately by sequential warnings and a block, just because the welcomer didn't bother to check whether the new account was just here to disrupt (but hey, if you keep copy-pasting welcomes onto new user talk pages, your edit count rises rapidly!). In no way do I mean that to take away from the significance of the problem of making good editors feel unwelcome, but I'm just saying that it's hard to solve this problem. --Tryptofish (talk) 23:07, 8 October 2022 (UTC)

@Espresso Addict, my impression is that getting out of AFC often requires the articles to be better than two-thirds of what's already in the mainspace. I have asked at Wikipedia:Request a query#Number of refs in the median article to see if we can get more information (RAQ may not be the best place, as it sounds like Quarry can't count the number of refs directly). In particular, I'm curious whether most current articles have two or three refs, but we require multiple times that to get out of AFC. Nobody wants to be the editor who approved a "bad" article, so our minimum requirements ratchet ever higher... WhatamIdoing (talk) 23:31, 8 October 2022 (UTC)

WhatamIdoing: Indeed! It's not the fault of the reviewers, the system Just Doesn't Work. I make myself very unpopular there by occasionally blundering in and trying to rescue something, but I've found spending prolonged time there so depressing, I have to take a wikibreak to recover. Espresso Addict (talk) 23:44, 8 October 2022 (UTC)

To be clear my comments are a dig at the process, not at either the reviewers trying to follow the process nor the people who in good faith designed the process. The ratchet effect mentioned—in which nobody wants to risk being the one who let the defamatory edit/the vandal editor/the fake reference go live—means that over time our processes get steadily more exclusionary.

@WAID, I'm not sure median articles is a particularly useful metric. Five minutes clicking on Special:Random makes it clear that the "56% of articles are stubs" claim is correct (and thus that the median is a one-paragraph stub, probably 'sourced' to a database rather than an actual source), but that doesn't mean Berquist Ridge, Cafe Mascot or Lisa Bertini are models to follow. The very-delayed fallout from the defenestration of LauraHale and Lugnuts, currently taking place at Wikipedia:Arbitration Committee/Requests for comment/Article creation at scale, might shift the dial somewhat, but any cleanup will take years. As things stand, AFC is in a no-win situation. Either we say "most of this project consists of useless articles on barely-notable trivia, take that as a model and add more garbage to the pile" which means new editors are highly likely to see their contributions promptly tagged for deletion; or, (as currently happens) we say "do as we say, not as we do" which means new editors get the impression that Wikipedia is run by a gang of hypocrites who've self-selected themselves into positions of authority and then pulled up the drawbridge behind themselves. Unless and until we grasp the nettle of cleaning up the mess, this is always going to be an issue.

"Number of references" isn't always particularly useful either, unless it's a genuinely contentious topic where one needs to reflect multiple viewpoints. (If anything, a one- or two-sentence stub with half a dozen references is typically a warning flag of a spammer or POV-pusher.) Even at FA level, it's not all that unusual for articles on non-controversial subjects to be sourced exclusively to the works of whichever expert wrote the definitive book or books on the topic—for something like Petter's big-footed mouse or Nico Ditch it would be wilfully perverse to deliberately use a less-reliable source purely for the sake of being able to say "we're using multiple sources". (Some regulars have learned to intentionally mask this by using alternative sources for such things as distances and locations, so as to satisfy "multiple independent sources" whilst relying on the works of the leading expert for anything substantive, but that's very much process-for-the-sake-of-process.) ‑ Iridescent 04:50, 9 October 2022 (UTC)

Johnbod's Law: 5 refs on a line is almost always a sign of trouble. Johnbod (talk) 15:11, 9 October 2022 (UTC)

I've written at length elsewhere about scope creep in NPP, and I think that the enwiki-only PageTriage software is partially responsible. In particular, it is missing a feature that is needed: the ability to see how many other reviewers have already looked at the page.

The process is meant to be:

• Alice looks at a new article, and tags it for deletion as a copyright violation. (Only touch it once – efficiency is important here, because experienced editor time is valuable.)

or:

• Bob looks at it, sees that it's not a disaster, and moves it out the patrollers' collective queue.

What we actually get is:

• Alice looks at a new article. She see it's not a copyright violation, which is her personal area of interest and expertise. There's no way to tell other editors that it's not a copyvio, so she silently moves on.
• Bob looks at the same new article and sees it's not blatant vandalism. He silently moves on.
• Carol looks at the same new article and sees it's got a source. They silently move on.
• David checks the same article for copyright violations again, and then silently moves on.
• Frank checks all of the Wikipedia:Criteria for speedy deletion, and since it doesn't qualify for any of them, he silently moves on.
• Grace looks at the article, doesn't see anything wrong, but silently moves on because she doesn't know much about the subject area, so she doesn't feel comfortable publicly "approving" articles.
• Heidi tags the article with two maintenance tags. She is our sixth reviewer to look at the article, but the first to take any action that can be seen by other reviewers.
• Ivan notices that the article has been in the queue for 30 days, but he doesn't want to edit articles about people, so he silently moves on.
• Judy checks to see that the article contains a source, but her interest is just in getting unsourced articles deleted, and this one has two sources, so she silently moves on.
• Mike is the fourth reviewer to check whether this article is a copyright violation, but it isn't, and there's no way to record that it was checked for copyvios and found to be acceptable, so he silently moves on.
• Niaj, our 11th reviewer, notices that the article has been in the unreviewed queue for 60 days. Niaj checks one of the references, which is an online magazine, but can't decide if this is a good enough source, so they give up and silently move on.
• Olivia runs across the article. She can't think of any reason to get rid of it, but she thinks it's being used for self-promotion. She silently moves on.
• Pat begins to panic about this article being overlooked before Google will index it (currently set by us to 90 days, and dependent entirely on Google's voluntary choice to respect this). Wow, the NPPers must be overloaded, if something like this hasn't been checked by anyone except Heidi in more than two months! Pat pulls up CSD and carefully checks each of the criteria. It doesn't qualify for CSD. The subject is a living person, though, and it doesn't say the person is bad. Pat, who feels proud of having a sixth sense for sniffing out hidden conflicts of interest, decides to add Template:Undisclosed paid. Pat can just tell that nobody would write about this subject unless they were being paid.
• Rupert sees all the maintenance tags on the page and decides to reject the article. Rupert can't see any reliable path to getting the article completely deleted, so Rupert moves the page into Draft: space, where it will either become WP:AFC's problem or be deleted after six months anyway.

Most of the time, these articles just quietly disappear. In rare cases, they lead to drama about why someone would get rid of an article on such an obviously notable subject (e.g., Disney's new CEO [whose article was hidden in draftspace nine minutes after creation, despite a note saying that the editor was still working on it], politicians in developing countries, certain Nobel Prize winners), but mostly, these articles just disappear.

One of the flaws in mw:Extension:PageTriage is that it's set up to have a single "I approve of this article" action. It would IMO make more sense to structure page review according to the CSD requirements: Alice determined that this wasn't a copyvio, and clicks the 'not {{db-copyvio}}' button, so nobody needs to check that page's existence against copyvios again. Bob determined that it didn't qualify for {{db-vandalism}}, and nobody needs to check that page for vandalism ever again. And so forth, through the whole list. And when you get to the end of the CSD list, the page moves out of NPP's queue, because the other stuff (e.g., should we have an article about this album, or should it be merged up to the band's page?) isn't NPP's problem anyway.

NB that I'm not saying that PageTriage should be improved. It might make more sense to scrap it. It certainly could be true that the WMF feels like a piece of software used only at the English Wikipedia might not be their strategic priority for the next year or two. I'm only saying that what we have now is causing wasted, duplicative effort, and that part of the problem is structural, because making one person sign on the dotted line as approving something always results in people refusing to approve something that's okay but not great. WhatamIdoing (talk) 16:07, 9 October 2022 (UTC)

That's not just an issue with the PageTriage extension. Admins have a wearily well-documented issue with the fact that there's no obvious way to flag an editor as non-problematic—thus, someone flagged as having a potential COI who has a perfectly good explanation, but doesn't feel comfortable making the explanation public and thus outing themselves (someone citing a pre-publication book, say) tends to end up blocked as a spammer even if they've privately explained themselves to half-a-dozen different admins. (The most notorious occasion this problem arises is probably with the self-appointed civility cops. Someone can make a talkpage comment which twenty different people see and conclude is non-problematic, but it only takes one admin to misinterpret it and the editor in question gets blocked. EEng is probably the most obvious example of this, but the problem is that most normal people aren't as sanguine as EEng in this situation, and either walk off in disgust or go down the "You want incivility? I'll give you incivility" road and the situation just escalates. ‑ Iridescent 16:50, 9 October 2022 (UTC)

15 years ago...

{{u|Sdkb}} ^talk 05:14, 20 September 2022 (UTC)

Happy Adminship Anniversary!

15 years is frightening. To put that in perspective that's more than 2⁄3 of Wikipedia's existence. (The pedant in me needs to point out that in my case "15 years since RFA" doesn't equate to "15 years of being an admin", owing to desysopping/resysopping). ‑ Iridescent 03:50, 22 September 2022 (UTC)

• Happy 15 my friend. That was a long time ago. Oddly I thought you'd been an admin much longer than I had. Star Mississippi 19:01, 22 September 2022 (UTC)

  Thanks… No, I was in the same batch as you—one of the wave of people who got around to creating accounts in 2006 when the crackdown made it harder to edit as an IP, and passed RFA a year later. We're Wikipedia's equivalent of the baby boom. ‑ Iridescent 04:13, 24 September 2022 (UTC)

There are probably some Wikipedians around now who weren't born when you passed RfA, which is probably a bit of a sobering thought really. Certainly I can remember editing WP as an IP before my eldest son was born, and he's now taller than me. Ritchie333 ^{(talk) (cont)} 22:15, 26 September 2022 (UTC)

Not even probably but certainly, given how many outreach and editor-recruitment drives are aimed at schools.* I'd be surprised if there aren't at least some admins under the age of 15. ‑ Iridescent 05:32, 3 October 2022 (UTC)
*FWIW, I think targeting recruitment at schools is a totally misplaced use of resources. Even Wikipedia:WikiProject AP Biology 2009—usually held up as the high point of outreach—as fas as I know didn't result in even a long-term editor once the school project was over, and in the meantime sucked up a vast amount of the time of other editors hand-holding the participants throughout the process.

I started editing as an older teenager because my older brother used to edit and I was bored. Honestly, I think that's the best way to retain editors; 'recruiting' organically. The first step: raising awareness that yes, you can edit Wikipedia, which sometimes people don't know. I dream of horses (Contribs) (Talk) 05:59, 3 October 2022 (UTC)

I totally agree; the way to recruit and retain editors is to see what the people who've drifted into our orbit want and to try to keep them, rather than to try to decide the kind of people we want and set out to reach them. Good luck convincing the WMF of that, though; for at least the past decade their attitude has been that all the problems could be solved if they could somehow dismiss the community and appoint a new one. ‑ Iridescent 08:39, 3 October 2022 (UTC)

(EC) That event was over a decade ago, I'm pretty sure the focus of outreach has since been undergraduate and above. I worked for the UK chapter for a couple of years (2013-2105) and I don't remember being involved in High school outreach then, and neither of the retired teachers who used to do that or argue for it in the UK are still on Wikipedia. Though even outreach to 14 year olds isn't going to result in 15 year old admins. I could see some logic in outreach to 6th formers if 6th form work requires an understanding of reliable sources, but if I was still involved in outreach and trying to recruit new editors I'd be going to retirement fairs and the university of the third age; places where people are open to getting involved in a new sedentary hobby. As for the idea that any of our current admins might be under 15, we only have twenty who created their account in December 2015 or more recently the rest of us would need to be less than eight years old when we took up the hobby of editing an encyclopaedia. Given that we have some RFA regulars who would oppose on "maturity concerns" if they thought a candidate was under 18; and the de facto requirement to have mastered the art of doing an inline cite to a reliable source; and the general greying of the pedia that has come with requiring sources and not having a decent editing environment on the smartphones and tablets that young people use; I would be surprised if we had more than one or two admins passing RFA before their 18th birthdays in the last half dozen years. We may not have any current admins under 18 years old, and I would be astonished if we currently had one under 16. ϢereSpielChequers 07:05, 3 October 2022 (UTC)

The WMF probably have some statistics somewhere, albeit they'd rely on people telling the truth in surveys. You make a good point (obliquely, at least) in that Wikipedia 2022 has a much steeper learning curve than it did in the days of Acalamari. That said, I'd still be surprised if there weren't at least some admins—or at minimum, people with a high level of community respect who act as de facto admins—under 18. ‑ Iridescent 08:39, 3 October 2022 (UTC)

I agree with you re our de facto admins and am encouraging one to run an RFA when they are legally adult. As to whether any of those twenty will surprise us in the next few years by announcing they've now graduated from university five years after becoming an admin? Time will tell, but there are only twenty of them, I've had a good look at several to the point of being sure they are legally adult. If we'd run the same stats fourteen years ago then every single admin we had had an account new enough that they could have been nine when they started editing. I assume you accept the logic that if someone has had an account here for nine years we can safely assume they are now at least 18? ϢereSpielChequers 09:26, 3 October 2022 (UTC)

@WereSpielChequers Besides losing support at RfA from !voters citing concerns over maturity, is there any reason to wait until one is 18 before running for admin? As far I'm aware, admins aren't responsible for managing money, negotiating and signing contracts, or anything else that requires guardianship over self and property. I dream of horses (Contribs) (Talk) 20:57, 3 October 2022 (UTC)

Articles/drafts in the speedy queue reasonably often contain material that we should not be promoting minors encountering. Espresso Addict (talk) 00:16, 4 October 2022 (UTC)

Espresso Addict I hate to break it to you, but you have a false sense of security. Before a draft or article is deleted, it's still viewable to the public. They're "no-indexed" so search engines can't find them, but humans can still find them (accidentally or not!). One reason why Wikipedia isn't censored is because we can't censor to everyones needs. That's why I'm focused on the legal rights of adults that children don't have, like the right to sign binding contracts, instead of inevitably variable cultural values. I dream of horses (Contribs) (Talk) 00:57, 4 October 2022 (UTC)

Sure, but no-one is actively pointing minors at them. And the hypothetical admin minor would also be able to look through deleted articles, including those deleted as attack pages. Espresso Addict (talk) 01:32, 4 October 2022 (UTC)

@Espresso Addict I have concerns that you'd want the sort of security that would require a level of age verification beyond what most editors would be comfortable with, to confirm that all admins are adults with self-guardianship/conservatorship. I dream of horses (Contribs) (Talk) 03:45, 4 October 2022 (UTC)

Every edit involves a legally binding contract.

I suppose someone should have another go at changing the language in pages like Wikipedia:Drafts#Publishing a draft. It isn't about "publishing" at all, and I worry that some enterprising lawyer will be sending sternly worded letters to an innocent AFC editor about "publishing" libel, since some of them have insisted that they, rather than the original editor, are the ones "publishing" the articles. WhatamIdoing (talk) 03:21, 4 October 2022 (UTC)

@WhatamIdoing Every edit involves a legally binding contract. Hmm. You have a point. There is a partial copyright waiver, at minimum. Perhaps any parent who'd have issues with their child editing Wikipedia would figure out to prevent access to the site, which means the kid wouldn't be an admin; it'd certainly fall on the parent or guardian to do that.

I worry that some enterprising lawyer will be sending sternly worded letters to an innocent AFC editor about "publishing" libel, since some of them have insisted that they, rather than the original editor, are the ones "publishing" the articles. It doesn't make sense to insist you're the one publishing the draft instead of the one who moved it. I might fix it at some point. I dream of horses (Contribs) (Talk) 03:56, 4 October 2022 (UTC)

@I dream of horses, I'd strongly recommend leaving that change to WAID to handle since as a WMF employee, she can run it by legal to confirm any proposed new wording is correct. I have no idea what California law is on the matter, but in at least some jurisdictions "moving the page to a more visible location" would legally be as bad or even worse when it comes to defamation and/or copyright issues—§230 may shield the WMF itself to some extent but won't necessarily shield individual editors particularly if they don't live in US jurisdiction. People have certainly been hauled off to court for things like retweets, and this would seem to be an analogous situation. ‑ Iridescent 04:56, 8 October 2022 (UTC)

Yeah, I wasn't thinking of doing anything, and no one else should do anything unless they know about that sort of thing and work for the WMF. I get that. No worries. I dream of horses (Contribs) (Talk) 05:09, 8 October 2022 (UTC)

If y'all want work-me to talk to Legal, then I can, but they can't always provide a public response, which can make it complicated to get people to go along with it. As the staff person who coordinated the change from the old "Save" button to the modern "Publish changes" button, I do think they want people to know that the publication happens when you click the big blue button, not when (e.g.) someone else later approves it for indexing by a search engine. WhatamIdoing (talk) 21:36, 8 October 2022 (UTC)

@WhatamIdoing: I do think they want people to know that the publication happens when you click the big blue button Making sure people know what happens is a good idea. Of course, that does require people to read; there's still the odd "I thought I was privately saving this" confusing at en-help, for example. As to whether or not you should talk to legal, I'll leave that up to you. I dream of horses (Contribs) (Talk) 02:22, 9 October 2022 (UTC)

• Get back to work slacker. *whipcrack* (also congratulations. My wife has been watching someone in the US who walks around cemetaries and I thought of you.) Only in death does duty end (talk) 08:56, 3 October 2022 (UTC)
• I was 17 when I passed RfA in 2020. @WereSpielChequers, on your ...more than one or two admins passing RFA before their 18th birthdays in the last half dozen years..., I'm aware of at least 4 (counting myself) who passed between 2020-2022. I would guess the actual number is a little higher than that. Moneytrees🏝️(Talk) 04:38, 4 October 2022 (UTC)

  @Moneytrees I remember a conversation on IRC with a couple of oversighters; they told me that "Some children edit better than some adults," then reminded me they often oversight[ed] age disclosures of those 15 and under. I'm not surprised that there are at least 4 [admins] who passed between 2020-2022. I dream of horses (Contribs) (Talk) 05:06, 4 October 2022 (UTC)

  @Moneytrees I'm glad to be proven slightly wrong, clearly my theory about the greying of the pedia is not quite as bad as I thought, and my ability to spot young editors is a bit rubbish. Good thing I'm not a barman. However, my theory that very few of our admins can currently be under 18 still holds up. You created your account in 2018, and it is the tenth newest among our admins, and you are now over 18. So going back to "As for the idea that any of our current admins might be under 15, we only have twenty who created their account in December 2015 or more recently the rest of us would need to be less than eight years old when we took up the hobby of editing an encyclopaedia. Given that we have some RFA regulars who would oppose on "maturity concerns" if they thought a candidate was under 18; and the de facto requirement to have mastered the art of doing an inline cite to a reliable source; and the general greying of the pedia that has come with requiring sources and not having a decent editing environment on the smartphones and tablets that young people use; I would be surprised if we had more than one or two half a dozen admins passing RFA before their 18th birthdays in the last half dozen years. We may not have any current admins under 18 years old, and I would be astonished if we currently had one under 16." I'd be interested to hear what you and the other teens, and recently ceased to be teens think of my comments re mobile editing. Am I correct in thinking that your generation is skewed towards using mobile devices rather than PCs? Are you in a minority of editors who use smartphones, or do most teens have proper PCs in their bedrooms? Also are there any among you who created your accounts before your mid teens? @I dream of horses, my personal test of whether someone is old enough to be an admin is Do your parents and guardians allow you to have a computer password that you can't share with them? others have other concerns. As a crat I accept that age and maturity concerns !votes are valid. As a nominator if I know a potential candidate is on the young side I tend to advise them to wait until they can say they are legal adults, though I might now review that advice. As a voter I consider that the sort of teenagers who take up editing an encyclopaedia as a hobby tend to be more responsible than the oldies like me. ϢereSpielChequers 07:09, 4 October 2022 (UTC)

  @WereSpielChequers Not being able to share a password is honestly a concern for any advanced user right, but particularly for user rights you need to go through a !voting process for. That makes sense; it's a security issue. Thankfully, my parents weren't interested in logging into any of my internet accounts when I started editing Wikipedia, even though they technically had the right to do it; I was a minor at the time.

  Generally speaking, if you get offers to run for adminship from multiple people, you probably have the emotional maturity to go through the process with dignity or the insight to refuse to go through with it, regardless of age. I dream of horses (Contribs) (Talk) 07:21, 4 October 2022 (UTC)

  I wouldn't say Generally speaking, if you get offers to run for adminship from multiple people, you probably have the emotional maturity to go through the process with dignity or the insight to refuse to go through with it is particularly true. I don't think you were around during the time when we were getting RFA candidates more regularly so we had a large sample to work with, but I was there and can say with some confidence that there wasn't a particularly obvious correlation in either direction. If anything, the long-standing and well-regarded editors were the ones who tended to take opposition at RFA as a personal insult and flare out spectacularly, whereas the self-nominations who garnered a stream of "oppose, clearly unsuitable" comments didn't have such high expectations and tended to shrug it off. (Have a wade through the archives of Wikipedia:Unsuccessful adminship candidacies (Chronological) if you want to conduct your own sampling exercide.) ‑ Iridescent 05:06, 8 October 2022 (UTC)