Wikipedia:Village pump (policy): Difference between revisions

Browse history interactively

← Previous edit Next edit →

Content deleted Content added

VisualWikitext

Inline

Revision as of 15:34, 20 February 2021

Table of contents
First discussion
End of page
New post

Shortcuts

The policy section of the village pump is used to discuss proposed policies and guidelines and changes to existing policies and guidelines.

If you want to propose something new that is not a policy or guideline, use Village pump (proposals).
If you have a question about how to apply an existing policy or guideline, try one of the many Wikipedia:Noticeboards.
This is not the place to resolve disputes over how a policy should be implemented. Please see Wikipedia:Dispute resolution for how to proceed in such cases.

Please see this FAQ page for a list of frequently rejected or ignored proposals. Discussions are automatically archived after remaining inactive for two weeks.

« Archives, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202

Lacuna in SPS:BP policy - think it needs an update urgently to address think tanks, advocacy organizations, academic group projects

I believe the self-published source policy regarding biography of living person articles - WP:BLPSPS - needs clarification urgently. The context is that I'm involved in a relentless tug of war about the inclusion on the Douglas Murray (author) page of an academic group research project titled the Bridge Initative[1] which is meant to address Islamophobia in the public space and is maintained by Georgetown University on a BLP article. It does not seem to be contested that the Bridge Team[2], to whom the articles are credited, is highly distinguished, including professors John Esposito Farid Hafez and Susan L. Douglass, the human rights lawyer and commentator Arsalan Iftikhar and a host of others, nor that it has been cited by other RS's[3][4][5][6][7][8][9] Nobody has been able to distinguish Bridge in evidentiary terms from advocacy groups like the Anti-Defamation League and the Southern Poverty Law Center. However, several people have pointed to sections in editorial policy like this one:

Per WP:USESPS: "Self-published works are those in which the author and publisher are the same."

Self-published material is characterized by the lack of reviewers who are independent of the author (those without a conflict of interest) validating the reliability of contents.

Per WP: V:"Never use self-published sources as third-party sources about living people, even if the author is an expert, well-known professional researcher, or writer".

Bridge articles are written by a team and attributed to them, just as articles written by advocacy groups like the Anti-Defamation League, the Southern Poverty Law Center, the Innocence Project and so on tend to be. Articles on controversial BLP subjects feature articles from these groups as a matter of course: see the pages for Milo Yiannopoulos,[10][11][12][13], Richard B. Spencer [14] and Lauren Southern [15] for example. However, a literal reading of currently policy could exclude them as "self-published sources", and potentially exclude anything that doesn't have a person with the job title of "editor". It needs to be clarified whether the above are "self-published" sources or whether they are acceptable for BLP articles. I don't believe those who wrote this policy sincerely intended that this would be the interpretation and I believe it is contrary to the spirit if not the letter of the existing policy, but believe the policy needs to be urgently updated and clarified to state whether think tanks, advocacy organizations and academic projects are "self-published" and whether they are acceptable for BLP articles. Noteduck (talk) 08:32, 27 January 2021 (UTC)[reply]

Well, WP:SPS would be a problem only for WP:FRINGE think-tanks, not for notable groups of mainstream academics. E.g., Bart Ehrman's blog (full professor, world authority on the Bible, writing under his own name) could be used as WP:RS (not very high quality RS, but RS nonetheless, for less controversial stuff). Tgeorgescu (talk) 09:22, 27 January 2021 (UTC)[reply]

The concern is about content on BLP's specifically. They want our definition of self-published changed so they can add think tanks, advocacy organizations, and academic projects as content on BLP's. --Kyohyi (talk) 14:15, 27 January 2021 (UTC)[reply]

Note, this question is related to a current RSN discussion here [[16]]. Springee (talk) 11:58, 27 January 2021 (UTC)[reply]

Springee and Kyohyi, nice of you to join me. Tgeorgescu, it's worth noting that despite the widespread presence of the SPLC and the ADL on controversial BLP pages, WP: V does state "Never use self-published sources as third-party sources about living people, even if the author is an expert, well-known professional researcher, or writer". This doesn't appear to leave any exemptions for experts no matter what level of renown. I can't imagine those who wrote the policy intended to exclude those sources - or are think tanks, advocacy groups, academic projects etc not "self-published sources"? Am I missing something? Noteduck (talk) 04:59, 28 January 2021 (UTC)[reply]

This is becoming obvious FORUMSHOPing. This question was extensively discussed and you didn't get the answer you wanted. Springee (talk) 11:19, 28 January 2021 (UTC)[reply]

Springee, there is an obvious lacuna (or simply a norm of violating policy) in Wiki's SPS:BPL policy. I'm on this page to get it clarified and amended if possible - if you don't have anything helpful to add then don't add anything Noteduck (talk) 21:00, 28 January 2021 (UTC)[reply]

I disagree: This appears to be a genuine inconsistency in how WP:BLP is written or applied. :::::@Noteduck: Have you placed a notice at some of the places where editors familiar with BLP will see it e.g., WP:BLPN, WT:BLP? ElKevbo (talk) 21:25, 28 January 2021 (UTC)[reply]

I'm fine with opening the discussion but I think it's problematic when an editor is actively involved in a dispute over this exact fact to then open this discussion without informing the original discussion. Given some of Noteduck's other behaviors this certainly looks like simple forum shopping. That said, Noteduck has since notified the other discussion. There is at least one SPS discussion I'm aware of which took place here [[17]]. VP is probably not the best place for this though it would be a good place for a notification. Springee (talk) 22:30, 28 January 2021 (UTC)[reply]

Comment - upon a review of the related discussions, in which I am not INVOLVED, I would like to corroborate that this does seem to me to represent a lacuna. The SPS prohibition for BLP articles - which makes sense in and of itself - seems to me to be too broad. While we should not allow feuds even among highly espected experts to be reflected in WP articles when sourced only to their blogs, it seems to me that at least some of the following should be allowed on BLPs: (1) SPS from acknowledged experts as references for uncontroversial matters of fact; (2) attributed judgements from relevant experts (individuals and groups), sourced to self-published or other sources where editorial control is not fully separate from the author or authors; (3) authoritative judgments using references by respected organizations that are responsible for their own publications. I'm not sure what exactly what the path would be to recognize some or all of the above as valid for BLPs, but I do believe it would benefit the encyclopedia to do so, by enriching the published content without taking any risks of harm to BLP subjects through poorly-sourced claims. Newimpartial (talk) 21:44, 28 January 2021 (UTC)[reply]

Comment WP:USESPS is not a policy nor a guideline, it's not been vetted by the community; its statement "Self-published material is characterized by the lack of reviewers who are independent of the author (those without a conflict of interest) validating the reliability of contents" makes almost no sense, if it's read to make unusable the sources that organizations and universities publish. Publishers do not typically send articles and books out-of-house to be vetted by independent person's unpaid by the publisher, and thus, without a coi (academic journals perhaps are the only publishers who seek those not-employed-independents to review in peer review before publication). -- Alanscottwalker (talk) 22:47, 28 January 2021 (UTC)[reply]

I'm not sure who that freestanding comment was intended to address, but my (immediately preceding) comment was about WP:SPS - which is a policy - and which states, Never use self-published sources as third-party sources about living people, even if the author is an expert, well-known professional researcher, or writer. That was what I for one was criticizing as, ahem, lacking necessary nuance. Newimpartial (talk) 22:52, 28 January 2021 (UTC)[reply]

My comment was not indented because it was not a reply to you, it was a reply to the OP. Alanscottwalker (talk) 23:00, 28 January 2021 (UTC)[reply]

Is this notice board with this opening the property venue for this discussion? This is a reasonable question when taken out of the context on which it was raised. As I noted, it was discussed before but there was no resolution. I would suggest closing this discussion and raising it neutrality on the WP:V or WP:RS talk pages. Springee (talk) 23:29, 28 January 2021 (UTC)[reply]

Alanscottwalker and Newimpartial, thanks for your responses. ElKevbo, I did take it to WT:BLP and it raised a bit of argument and went nowhere - there doesn't seem to be consensus anywhere.[18] There was a BLPN thread which is relevant a few years back in which a few editors, including Kyohyi contested that the SPLC's Hatewatch blog should not be treated as an RS for BLP pages, partially on the grounds that it was an SPS, but they overruled.[19] Nonetheless the current policy still seems very unclear. When it comes to think tanks, advocacy groups and group research projects on BLP articles, in my opinion three propositions are possible:

there is an ongoing pattern of non-adherence to editorial policy on Wiki, evidenced by the frequent use of these groups (especially the SPLC and ADL) as sources on pages related to controversial BLP subjects
there is a lacuna in the SPS policy, and the policy needs to be clarified to make it clear these sources are permissible in at least some instances
think tanks, advocacy groups and research groups (if group projects) are not "self-published sources" for the purposes of Wiki policy

In my opinion the second proposition is most likely correct, and the current definition is contrary to the spirit if not the letter of the policy. What do others think - am I missing anything? Noteduck (talk) 10:04, 29 January 2021 (UTC)[reply]

Considering reliability of sources at the level of publisher causes issues. - Ryk72 ^talk 12:48, 29 January 2021 (UTC)[reply]

I think these concerns about SPLC and ADL are misplaced. First, they often deal with groups, rather than individuals, and groups are not subject to BLP rules. Second, we usually cite a newspaper or magazine article that says that the SPLC or the ADL put the group on a list, rather than citing their own publications directly. If you see someone citing SPLC's own publications to claim something specifically about an individual BLP, then please go visit your favorite web search engine, find an independent news article that says the same thing, and replace the source. WhatamIdoing (talk) 16:53, 10 February 2021 (UTC)[reply]

Comment My take is that a lot of editors don't follow how policy defines a self-published source. I do think our policy partially buries the definition (by putting it in a note, and not in the main body of text), but it does exist on WP: V. Per V "Self-published material is characterized by the lack of independent reviewers (those without a conflict of interest) validating the reliability of content.". Groups which have a specific POV, like think tanks, advocacy organizations, and research groups have an inherent conflict of interest with regards to their POV. This means that any internal review process also has a conflict of interest, and we would need to see that there is a review process that is not beholden to the aims of the group to justify that they are not self-publishing (something akin to an independent editorial board). To make an example that ties into this situation. Georgetown University has the bridge project, and it also has a University press. If the bridge project were to publish it's findings through the university press then it would not be self-published, but the bridge project's content published on university pages that it controls are self-published. --Kyohyi (talk) 13:41, 29 January 2021 (UTC)[reply]

To my mind, the core of the issue here is distinguishing the difference between a source that is being used as a reliable secondary source to support a statement of fact, vs citing that same source as a PRIMARY source for a noteworthy and relevant statement of opinion.

For example: let’s say that the SPLC had concluded that a BLP subject is a “racist”. That conclusion is certainly relevant and noteworthy... and it should be mentioned in the subject’s article. However, that mention should be phrased as BEING an opinion, attributed in text to the SPLC, and not stated as fact.

If our policies are not allowing us to state relevant and noteworthy opinions, when phrased AS opinions, then we need to amend our policies.

If, on the other hand, our policies make it harder to state opinions as if they were fact, then our policies are working as intended. Blueboar (talk) 14:14, 29 January 2021 (UTC)[reply]

Well to me, it seems like we should be following WP: V closer. Within SPS there's a sentence "Exercise caution when using such sources: if the information in question is suitable for inclusion, someone else will probably have published it in independent reliable sources.". When we rely on independent sources to cover content from advocacy groups, think tanks, and academic groups it makes the POV nature of such groups more apparent. Further making it obvious that such things are opinions and should be attributed. --Kyohyi (talk) 14:32, 29 January 2021 (UTC)[reply]

I for one don't accept that "facts" and "opinions" can be distinguished cleanly, at least not in the way that has been suggested here. For one thing, sources on the continuum between pure SPS and independent publication often document "facts", rather than "opinions", in my view. Examples include tabulated data put out on a researcher's blog or microblog, or a first-person account of email correspondence. A statement of these may not be accurate (or verifiable) if it is does not meet independent standards, but it is nevertheless a factual claim rather than an "opinion", and for non-BLP topics we do allow sources self-published by recognized experts to be considered reliable.

On the other hand, I also do not accept inflating the scope of "opinion" to include all judgments so that none of those are considered facts. QAnon is an antisemitic conspiracy theory - to me, that is an objective fact. Quibbling about the underlying judgment required to make that factual statement seems to me, in theory but also in practice, to lead to an absurd degree of FALSEBALANCE and relativism thanks to the dubious assertions that all statements resting on judgments are "opinions" and that opinions are never objectively true (or that they should never be presented as such in Wikivoice). Balderdash. Newimpartial (talk) 14:56, 29 January 2021 (UTC)[reply]

I've felt that a better way to handle this is to look at what others say about the information the source is providing. For example, the SPLC says on their website "Mr X is a racist because they felt Southern Fried Rabbit was funny". Now we have an article about Mr X. Should we include this claim? Well if the NYT said, "the SLPC said Mr X is racist because..." then I think weight has been shown. However, if no RSs have picked up on this claim then I would say no, it doesn't have weight. I guess that isn't a question of reliability though it would certainly have to be presented as the SPLC's opinion. In this way I would be treating the SPLC as a typical opinion source. I think <pb>Newimpartial's comment about the gray scale between fact and opinion is valid. Some facts are very objective (the house is 2 stories tall), some are semi-objective (the house is Georgian style architecture), some are quite subjective (the house is ugly). It's not always easy to decide if a source is reporting facts or reporting on their analysis of the facts (the speech meant X). <pb>Anyway, I think the core question here is where is the SPS line. My feeling aligns with Kyohyi, any time the "editorial staff" is not independent it should be treated as SPS. News organizations are supposed to address that. University projects are different. I'm sure some are very careful but where is the line? How do we decide this is a careful one vs a lab with just one student and one professor publishing their opinion on the lab website? What if this is a bigger lab with more collaboration? They all have the same issue so long as they don't have an independent reviewer. BTW, this discussion should really be moved to a proper forum. Springee (talk) 15:17, 29 January 2021 (UTC)[reply]

Wikipedia's definition of a self-published source is: Self-published material is characterized by the lack of independent reviewers (those without a conflict of interest) validating the reliability of content. It is completely bonkers. A journalist working at a news paper does not have their work subjected to independent reviewers. Editors are not independent reviewers. Even if they were, editors mostly spell and grammar-check whatever the journalists write, they neither have the time nor the knowledge required to fact check the content. So given Wikipedia's rules, every article in every news source ever written is self-published.

But for argument's sake, let's say that an "editorial process" at a news paper means that the source is not self-published. Then we have the following bizarre situation: whatever Amnesty publishes on their website is "self-published" but the exact same content published in the organization's monthly members' magazine is not self-published because the magazine has editors! ImTheIP (talk) 11:13, 30 January 2021 (UTC)[reply]

Indeed, you are right that would be bonkers and bizarre. And, it's only not bonkers, if it's applied as a type or example of publishing, although it would be limited to a very small subset of what Wikipedia universally sees as not self-published, that is, it has peer review as the example. But almost no publishers do peer review (and not just in news) and Wikipedia definitely does not require that, at all. In addition, the language of "characterized" also suggests that the sentence is not complete as an example; elsewhere in that footnote quoting the University of Chicago it says "any Internet site that does not have a specific publisher or sponsoring body should be treated as unpublished or self-published work." That's what Wikipedia basically does. It was also suggested before your comment that we look for an "editorial board", which also makes no sense, no editorial board, at practically any publication reviews any piece before it is published.

Also, construing policy as making Wikipedians accept that which is demonstrably not true, ie, that there is no publisher separate from the author is bad encyclopedia writing. When a reporter writes for The New York Times, they are not self published, When an academic writes for Georgetown University, they are not self published, When a researcher writes for ADL, they are not self published. Not only do those publishers (New York Times Company, Georgetown University, ADL) regularly have the publisher rights, they always have the publisher liabilities. -- Alanscottwalker (talk) 13:32, 30 January 2021 (UTC)[reply]

I agree that the sentence, although it is not, and does not claim to be, a "definition", needs to be fixed.

However, @ImTheIP is wrong about his claim that newspaper journalists and editors aren't independent. If you write a story for the local newspaper, and the editor rejects it, then that has no effect on the editor's paycheck, right? If we adopt the standard that anyone working together in a publication venture is self-published, then everything is self-published: every book, every journal article, every newspaper. (Also, newspapers actually do pay independent reviewers on occasion.)

The bigger problem that I see with this sentence is that it says the reviewer is "validating the reliability of the content". I probably wrote that, and it's maybe true in a technical, Wikipedia-centric sense (the presence of an editor who is not paid by the author is what causes the source to meet the standard that WP:V calls "meaningful editorial oversight" and that WP:RS calls "independent editorial oversight and peer review"), but it's not helpful for understanding the concept of a self-published source. WhatamIdoing (talk) 16:49, 10 February 2021 (UTC)[reply]

I've attempted to improve this by changing that sentence to this:

One characteristic of self-published material is lack of reviewers who are independent of the author (those who are not hired and fired by the author, and whose employment does not depend upon agreeing with the author).

I've also added a short table of examples. WhatamIdoing (talk) 17:58, 10 February 2021 (UTC)[reply]

Bumping this discussion Alanscottwalker, ImTheIP Springee Newimpartial Blueboar Kyohyi etc. We all seem to agree that the currently policy is hopelessly vague and unhelpful - how to go about changing or clarifying it? Noteduck (talk) 23:37, 5 February 2021 (UTC)[reply]

I don't agree that the current policy is either hopelessly vague or unhelpful. I think that you need to stop for a moment and think about what's being asked. Move out of the (pseudo-)academic realm for a moment. Imagine that Apple Inc. – a place with far more lawyers than any think tank, and a more tightly controlled website than most organizations in the world – decides that it wants to say something about Bill Gates of Microsoft fame on its website. The website's contents are written by Apple and published by Apple. Apple is not a traditional publisher, and they're not going to the trouble of sending their content to a traditional publisher. Should editors be allowed to cite Apple's own website to talk about a living person?

And if your answer there is no, then why would you want a smaller organization to be able to do what you ban Apple from doing? WhatamIdoing (talk) 16:58, 10 February 2021 (UTC)[reply]

To follow up on WhatamIdoing's example: Apple's website is a reliable source for what it is saying on topic X, but it isn't (absent other evidence) a reliable source to establish that its statements on topic X have sufficient due weight to be included in a Wikipedia article on topic X. The same goes for think tanks: there needs to be some third-party evidence to support that the think tank's views are sufficiently significant to be included in an article. isaacl (talk) 23:07, 10 February 2021 (UTC)[reply]

This is my feeling as well. Springee (talk) 23:40, 10 February 2021 (UTC)[reply]

So is WP:COSMETICBOT not policy anymore?

I guess this is the right place to put this, since it's a policy-related question. My watchlist has been completely bombarded with seemingly cosmetic bot edits lately, to the point that it's really rendering part of the idea behind a watchlist useless, as it makes it very hard to keep track of edits I would actually want to monitor. For instance, this edit (it's to an article since-deleted by PROD, so only admins can see it, but the edit is by Monkbot, is marked as cosmetic, and simply consists of changing the parameter |accessdate in {{cite gnis}} to |access-date). I watch all articles that I PROD, to see if the PROD gets contested and I may need to follow up with AFD. Monkbot made that same edit to a large number of articles I had PRODded and watched, making it much more difficult to keep track of these PRODs. So what is clearly a meaningless cosmetic edit (accessdate vs access-date isn't an error and has no impact on output) is creating the sort of situation that we have (had?) WP:COSMETICBOT for - the large numbers of automated make it needlessly difficult to actually keep track of changes, and why are articles up for deletion the sudden targets for cosmetic edits anyway? Or stuff like this or this. What are those really accomplishing. And just not showing bot edits in my watchlist isn't an option I'm comfortable with either, because of stuff like this, where a bot went around making link changing edits where there was a valid editorial reason for the human-chosen links.

Sorry for the TL;DR rant, but I was pretty sure that COSMETICBOT was still policy, but wound up being very confused and annoyed when all of a sudden 30-odd percent of the changes appearing on my watchlist were seemingly-cosmetic bot edits. Even the ones working on Category:CS1 maint: ref=harv are essentially cosmetic edits, given that that category says up at the top The CS1 maint: ref=harv message is not an error message., so removing |ref=harv isn't really a particularly useful edit. I'm seriously considering adding {{bots|deny=monkbot 18}} to pages I'm actively watching so I can actually monitor useful changes to those articles without everything being gummed up. Hog Farm _Talk 06:07, 31 January 2021 (UTC)[reply]

Hi, Hog Farm. There is a (quite long) discussion about this at Wikipedia:Bots/Noticeboard#Monkbot 18. — The Earwig ^talk 08:24, 31 January 2021 (UTC)[reply]

The short answer is: The bot has been fixing problems that don't currently appear to make a difference, but which will make a very visible difference (broken refs all over the page) in the fairly near future. WhatamIdoing (talk) 17:20, 10 February 2021 (UTC)[reply]

Proposal to elevate the contents of MOS:ACCESS as policy.

Currently, the entirety of the Manual of Style is considered guideline. However, MOS:ACCESS infringements impact the usability of Wikipedia, and its accessibility. I propose that MOS:ACCESS be elevated above the level of guidelines, and more stringently enforced in mainspace and in wikispace.

There has been a recent proliferation of articles infringing MOS:ACCESS, particularly among COVID-related articles with statistics charts. Needless to say, access to information about an active pandemic should not be obstructed due to poor adherence to the Manual of Style. This troubling trend of flouting MOS:ACCESS should be nipped in the bud, before it becomes so widespread that fixing it is nearly impossible and Wikipedia's famed level of accessibility becomes permanently degraded.70.52.144.5 (talk) 16:05, 31 January 2021 (UTC)[reply]

Practically everything in MOS should be a guideline, because, well, they’re guidelines. Elevating MOS:ACCESS to policy won’t fix the particular issue you’re referring to anyway. Suggest discussing those graphs on their relevant talk; I suspect the limitations are a mixture of technical and needing someone who’s willing to do the work. ProcrastinatingReader (talk) 16:09, 31 January 2021 (UTC)[reply]

No technical limitations. Not ALL of the statistics charts on those pages infringe MOS:ACCESS, just a subset. The way the non-infringing subset are done is therefore the way they all should be done, to adhere to MOS:ACCESS. That makes the continued infringements of MOS:ACCESS, which have now drawn comment from at least 3 users, willful ones. Ridiculously, one of those users was actually asked to fix it on one of the affected pages themselves by uploading images and editing the page, despite their being an IP user and thus unable to upload images, and that particular page being semi-protected so not editable by IP users. Someone else would therefore have to do it, yet no-one would, even though requests for (non-vandalizing) edits to semi-protected pages from IP users are supposed to be honored.70.52.144.5 (talk) 17:52, 1 February 2021 (UTC)[reply]

The "someone else" may have to a be a future you. Register an account, wait 10 days, fix it. - Ryk72 ^talk 23:24, 1 February 2021 (UTC)[reply]

Ridiculous. Your response obviates the whole purpose of semi-protected edit requests. Furthermore, those COVID pages were fine up until January 8. Someone else's edits broke MOS:ACCESS (and particularly MOS:PRECOLLAPSE). Why should I be expected to clean up their mess? I don't even have the relevant expertise. I have little familiarity with the markup used around here. I can fix a typo or add a bit of information here and there, but it would take me weeks to learn enough to be able to fix what whoever broke it could fix in five minutes. If this was a simple matter of personal preference I might simply give up and let the more knowledgeable editor have it their way, but this is not a simple matter of personal preference. It's a matter of a flagrant violation of an important guideline. What that other editor did is WRONG and should not be allowed to stand. What I can't understand is why I can't seem to find anybody who agrees with me! If there was a broad consensus that, say, MOS:PRECOLLAPSE was obsolete, presumably it wouldn't even be there on the MOS:ACCESS page anymore. Since it is still there, there must be a large silent majority who agree with me. So why will none of them speak up?!70.52.144.5 (talk) 16:11, 2 February 2021 (UTC)[reply]

The MOS as policy would essentially require prior knowledge of the MOS before contributing to Wikipedia. That is to say, not writing in a MOS-compliant way could be seen as disruptive. Go ahead and stringently enforce the guideline that it is. Barring an exceptional reason, reverts done to MOS-compliant edits will be viewed as disruptive. Primergrey (talk)

That's not what a policy is. People violating policies or guidelines they don't know about has always what we've had IAR and BITE for. --Izno (talk) 17:48, 31 January 2021 (UTC)[reply]

Regardless of what a policy is or is not, the OP certainly seems to think a "promotion" to policy would make MOS-compliant editing more "enforceable". Primergrey (talk) 20:20, 1 February 2021 (UTC)[reply]

It won't, though. This is a common misconception about Wikipedia:The difference between policies, guidelines and essays. WhatamIdoing (talk) 17:22, 10 February 2021 (UTC)[reply]

But then again, it might. As the page you linked to is "a supplemental page, which is an even more ambiguous group...supplemental pages generally have a limited status during deliberations as they have not been thoroughly vetted by the community." Primergrey (talk) 17:58, 10 February 2021 (UTC)[reply]

No, it won't. Changing the label at the top of the page does not magically result in people reading it and changing their behavior. Even in a long-standing, high-profile policy, it often takes many editors two years to notice that the best-practice has been changed. (When was the last time you read all the policies? Never? Right. Neither did anyone else, so of course you don't know what changed in them last month.) WhatamIdoing (talk) 18:02, 10 February 2021 (UTC)[reply]

"No, it won't." Armed with that level of prescience, you ought to just tell everyone what result this thread will yield, and save them all some time. I'm talking about perceptions, which, given the blurriness of things, are pretty much all anyone has to go on. (And I have read every policy and guideline over the course of my time here. Less clarity was imparted than I had expected.) Primergrey (talk) 18:16, 10 February 2021 (UTC)[reply]

Over the span of many years, a few editors (including myself) have probably read all, or nearly all, of the policies. But if you're not doing that frequently, then you won't know what's changed.

It does not take prescience to know that changing a policy does not magically equip editors with these necessary and complex skills, or to know that people whose current personal opinion falls into the "benign neglect" range will not suddenly become advocates for adding accessibility features to data tables. Approximately 148,000 registered editors have made at least one edit in the last month. Wikipedia:Manual of Style/Accessibility/Data tables tutorial gets five page views a day. That means that for every person who looks up the right way to do it once, 999 other editors aren't doing it at all this month (and that assumes that nobody ever reads the page more than once, which isn't realistic). Slapping a policy tag at the top of the page will not change that. WhatamIdoing (talk) 18:35, 10 February 2021 (UTC)[reply]

More forum shopping. See the links I supplied in yesterday's attempted RFC: WT:WPACCESS#Since many editors treat Manual of Style content as mere ignorable guidelines, MOS:ACCESS content should be clearly made non-ignorable policy. --Izno (talk) 17:48, 31 January 2021 (UTC)[reply]

Not forum shopping. Escalation. And quit stalking me. Your opinion, that MOS:ACCESS should simply be thrown overboard because it's suddenly less convenient to apply, has been made adequately clear, and it is clearly incorrect. — Preceding unsigned comment added by 70.52.144.5 (talk) 17:59, 1 February 2021 (UTC)[reply]

MOS:ACCESS is explicitly a guideline because "it is best treated with common sense, and occasional exceptions may apply.". Not liking the answer you've been given every time you've asked will not and cannot change that. Thryduulf (talk) 21:46, 31 January 2021 (UTC)[reply]

Concur with every named user above who has replied to the anonymous IP address editor. --Coolcaesar (talk) 18:44, 1 February 2021 (UTC)[reply]

MOS:PRECOLLAPSE should probably be regarded as obsolete. CSS and JavaScript are fairly universal. Turning off CSS requires special plugins in many browsers nowadays. Hawkeye7 (discuss) 05:34, 3 February 2021 (UTC)[reply]

It's actually still a problem. Dropdowns/collapses are busted on the mobile site, for example. They either hide totally, or auto-expand, one or the other depending on the class. Creates an awful UI experience when infoboxes, for example, use collapses with a mass of content in it. On mobiles, that infobox appears under the first paragraph of the lead, auto-expands, and then becomes a very long scroll. ProcrastinatingReader (talk) 05:46, 3 February 2021 (UTC)[reply]

(sigh) Is it really necessary to remind everybody of https://www.mediawiki.org/wiki/No-JavaScript_notes again? Up to 7% of the user base can't see many of the statistics charts on the affected COVID articles. There are devices with browsers that don't support Javascript. There are security reasons to disable or whitelist it. Anyone arguing that MOS:PRECOLLAPSE is obsolete is arguing that people should be forced to let Wikipedia run code on their machines or no charts for you. Code that "anyone can edit"! That code will become a very tempting target for bad actors to modify, and probably already is. Basically, when a user lets a site run Javascript, any attacker there has already achieved the hardest step: getting remote code execution. They're now just one or two local privilege escalations away from cracking root ... 70.52.144.5 (talk) 19:01, 3 February 2021 (UTC)[reply]

Not everyone can edit global site JS, only the WMF/devs and intadmins. ProcrastinatingReader (talk) 19:09, 3 February 2021 (UTC)[reply]

So, there's a small barrier to access, not unlike the case with editing a semi-protected article. 70.52.144.5 (talk) 18:37, 4 February 2021 (UTC)[reply]

The rest is paranoia with no basis in reality, there is no known bug allowing gaining root access by executing JavaScript in any popular browser. ProcrastinatingReader (talk) 19:09, 3 February 2021 (UTC)[reply]

Poppycock. Ten seconds of googling sufficed to find these recent incidents:

https://www.trendmicro.com/en_ca/research/20/h/cve-2020-1380-analysis-of-recently-fixed-ie-zero-day.html "It is a use-after-free bug in Internet Explorer's JavaScript engine, jscript9.dll."
https://threatpost.com/2-zero-day-bugs-google-chrome/161160/ "... component of Chrome that handles JavaScript and WebAssembly"
https://nakedsecurity.sophos.com/2020/01/09/browser-zero-day-update-your-firefox-right-now/ "Simply put, a JIT compiler takes JavaScript source code, as you'll find in most web pages these days, and converts it to executable computer code"

That's the trifecta, an exploitable Javascript-engine zero-day in each of IE, Chrome, and Firefox, the top 3 browsers outside of Apple's expensive little gated community. All three of them during 2020, so recent. If these happened yesterday, another one could crop up tomorrow. 70.52.144.5 (talk) 18:37, 4 February 2021 (UTC)[reply]

The fact is that you’re going to have a very tough time on the internet in 2021 without enabling JS, a suboptimal browsing experience on many popular sites and others not allowing access entirely (eg React/SPAs/etc). ProcrastinatingReader (talk) 19:09, 3 February 2021 (UTC)[reply]

That is true, and it is also very, very bad. And it is mostly for ulterior motives. A lot of those sites are serving static text and images as the bulk or even the entirety of what they provide to the end user, which does not technically require any client-side scripting. (Rather like an encyclopedia site, come to think of it.) But anyone who's blocking their scripts is blocking their ads, so they are motivated to force people to allow their scripts. A good and wise site operator would just fall back to good old-fashioned 400x40 JPEG ads but way too many site operators are greedy jerks instead. Of course, Wikipedia has no such motives, since it doesn't appear to be ad supported, which makes the concerted effort certain editors are making to force Javascript down users' throats after all these years all the more baffling, as there does not seem to be a motive for doing so. Yet there is clearly a huge amount of resistance from certain quarters to any attempt to keep Wikipedia functional for non-JS users, to the point of some of the pro-forcing-JS-use faction feeling the need to try to stifle any debate of the topic at all, and to follow members of the anti-forcing-JS-use faction around the site to pester them. It's very odd.

P.S.: Simply deleting everything I wrote yesterday because you disagree with it is NOT a mature way to debate an issue. Please don't do that again. 70.52.144.5 (talk) 18:37, 4 February 2021 (UTC)[reply]

I am sympathetic because the casual discrimination against the disabled that crops up when people ignore and/or argue against *some* elements of MOS:ACCESS is frustrating. Fortunately the WMF, in one of its rare occasions of doing something correctly, has a nondiscrimination resolution which supersedes local policies. The problem with MOS:ACCESS is it handles two different types of accessibility issues. Those related to technical accessessibility (eg browsers that lack CSS/JS support) and those that address disability based access (colourblindness, sight-impaired etc). The former should certainly not be a policy, the latter probably should given it crops up often enough that editors ignore it. Only in death does duty end (talk) 13:40, 5 February 2021 (UTC)[reply]

The distinction you're trying to make, apparently with the intention of carving out an exception that will give the pro-shoving-JS-down-everyone's-throats faction license to operate, does not really exist. Device capabilities and disability are not independent -- consider things like screen reader software or text magnification features. These things (and lots more) are designed to work with plain HTML and, in some cases, images and other commonplace material. Replacing any of these with scripts that attempt to do anything fancier, without appropriate no-script backups, is likely to screw them up. This is part of a more general adage: the more you overthink the plumbing, the easier it is to stop up the drain. 70.52.144.5 (talk) 19:02, 5 February 2021 (UTC)[reply]

Anyone can edit? - I am finding it increasingly difficult.

As an older editor (both in age and in “years editing WP”), I am becoming increasingly concerned by the amount of automated doo-hickies (templates, phabricators, T1234 thingys etc. etc.) one is expected to understand in order to edit articles. I know I am not being very specific in raising this concern (that is because I am not very tech savvy, so I don’t even know enough to complain accurately)... essentially It just feels like we are shifting from “The encyclopedia that anyone can edit” to “The encyclopedia that those with enough technical expertise can edit”. Please discuss. Blueboar (talk) 18:27, 2 February 2021 (UTC)[reply]

@Blueboar:, I am probably of the same generation and have edited about as long as you, and I don't share that experience. I don't pay any attention to anything tracked in pahbricator and I don't even know what "T1234 thingys" are and I edit in about the same way as I always have. I do use templates but those aren't really much of a burden if I remember to check the documentation ~~before~~ once I screw something up before I press "Publish changes". I'm not trying to deny your lived experience but I don't feel that the technical savvy needed to edit has changed greatly in the last decade or so. I'm sorry that's not of more help. Eggishorn (talk) (contrib) 19:15, 2 February 2021 (UTC)[reply]

(after edit conflict) I don't know if I can say I'm "older" at 62, and I count myself as pretty tech-savvy, having worked in technical areas of IT from 1980. I deliberately avoid the more technical areas of the project (the only village pump page that I don't have watchlisted is the technical one) but am concerned that we seem to have a significant number of editors who seem to think that technical IT expertise has some relation to knowing what should be done in an encyclopedia. My greatest concern is that there are so many editors, including admins, who seem to believe that the reaction to a bot not editing according to consensus should be anything other than immediately closing down the bot until the situation is sorted out, although they would apply this principle to a human editor. Phil Bridger (talk) 19:19, 2 February 2021 (UTC)[reply]

There's an established process for bots and consensus. That process is described in BOTPOL. They aren't treat like human editors because they aren't human editors. That's not to say the process is perfect or not in need of any reforms, though, but it does explain your question about why admins don't block contrary to policy. ProcrastinatingReader (talk) 05:32, 3 February 2021 (UTC)[reply]

I guess it depends on what kind of articles (generic) you like to edit and how much they rely on templates or other more advanced markup. Most editors will appreciate it if you at least use some citation templates, but you can probably get by with bare references as long as you don't mind others changing them later. If you're thinking about maintenance tasks like page patrol, yeah, some of those rely on tools or need careful attention to follow all requisite steps. isaacl (talk) 22:35, 2 February 2021 (UTC)[reply]

I agree that the amount of automation has steadily increased over time, but the purpose has been to reduce the amount of technical expertise required, so we can spend more time on the article writing. My experience with the kids in university classes has been that they are less technically savvy than the older generation. They don't look at how a web page is constructed any more. (Why would you? It's probably all full of CSS.) Nonetheless, we also have an educational mission, so in that spirit, everyone is obligated to learn. Working with bots, templates and Lua modules isn't so hard. Hawkeye7 (discuss) 05:16, 3 February 2021 (UTC)[reply]

Working with bots, templates and Lua modules isn't so hard.: I believe the average editor can get by without having to worry about bots and Lua. Automated bot edits can be configured to not show on watchlists, and their edits shouldn't be treated any differently than if they were made from a human. Templates might be less avoidable, as most use them for citations, and almost every page has an infobox. Help pages can be spruced up if non-technical editors can identify gaps they are experiencing.—Bagumba (talk) 05:26, 3 February 2021 (UTC)[reply]

Couldn't agree more. You don't need to worry about them. You can use the templates without knowing how they work. Hawkeye7 (discuss) 10:06, 3 February 2021 (UTC)[reply]

Of course way back when I was new, it did take a while to figure out how those {{...}} worked and where the hell to get the proper syntax.—Bagumba (talk) 11:06, 3 February 2021 (UTC)[reply]

The problem isn't that people can use templates and the like without knowing how they work, the problem is that it ahs become much harder for many people to fix issues in templates, infoboxes, ... in a specific article. E.g. modules are a step remote from templates. People used to be able to look at the template name that was used, type that into the infobox, and then see the actual source; now they will see some invocation of "something" without any indication that this is a Lua module, which is then much harder to parse for many people. Similarly, all the stuff that is taken from Wikidata may seem to make life easier, but in reality is a big black box for most editors, and if they do goover there, they have trouble finding out what they are supposed to do (e.g. creating a new reference there is a lot harder). So yes, there have been some developments which have made editing, understanding what happens, fixing (some) errors, ... increasingly difficult for a lot of people. Fram (talk) 11:32, 3 February 2021 (UTC)[reply]
Fram, I think such complication is necessary of any mature organisation or project. Complexities always develop which are inaccessible to some. No single person will be able to comprehend and efficiently operate in every area of the project. Most people don't need to be able to understand Module:Convert, for example. If they have an issue, they can pop it onto the talk page and it's someone else's problem. Using Lua modules is hence as simple as templates, assuming they're well documented and well designed. As for Wikidata, there's an editing pencil in infoboxes but I happen to think it looks a bit ugly.

(re below) I don't think stuff like {{authority control}} is creating an editing complexity, but it is bloat. ProcrastinatingReader (talk) 13:05, 3 February 2021 (UTC)[reply]
- Re. Wikidata, see something like the template CiteQ, which is extremely easy until you need to change anything (or need to create a new reference). AT List of woodcuts by Albrecht Dürer, the current reference 7 reads " Willy Kurth, ed. (1927). Complete Woodcuts of Albrecht Dürer. Foyles. ISBN 0-486-21097-9. OL 18383602M. Wikidata Q101542418." (with the necessary links). It is generated by {{Cite Q|Q101542418}} A 1927 book by Foyles. The title is a link to a Hathitrust version, [20], but this is a 1946 version published by Crown publishers. The OL (Open Library) link[21] goes to a 1927 version, but published by Dover Publications, not by Foyle. And the ISBN is for a 1963 version, published again by Dover. The intention of CiteQ is that you can reuse the reference over articles, languages, ... If I now change the Wikidata version to be about one specific version (so same year, publisher, ID numbers,...), will I break any uses of this reference? E.g. cases where page numbers are added, which aren't necessary the same in all these versions? I don't know, and I have no way of knowing it either as far as I know. I have already changed a few CiteQ uses on that article to standard references, but they are creeping up everywhere. Fram (talk) 13:20, 3 February 2021 (UTC)[reply]

This is exactly the sort of thing that I am talking about... what the hell does “Cite Q|Q101542418” mean? I have no clue how you get the text of the citation that appears in the article from that string of gobbledygook. There is no way I could edit this if I needed to. It is not something I have the skills to figure out. Blueboar (talk) 13:56, 3 February 2021 (UTC)[reply]

I think the corollary to that is that most new editors would find

{{Cite book|editor-last=Kurth|editor-first=Willy|year=1927|title=Complete Woodcuts of Albrecht Dürer|url=https://babel.hathitrust.org/cgi/pt?id=uc1.32106006231002|publisher=[[Foyles]]|isbn=0-486-21097-9|ol=18383602M}}

just as impenetrable and be just as lost about how to edit it. There's no way around having some structured data in our articles (citations for example), but tools like VisualEditor and Wikidata/{{Cite Q}} are an attempt to make it so the average editor doesn't have to work with it directly, and so reduce the technical barrier to editing that you complained about above. They're not perfect and I certainly agree that we need much more approachable interfaces for editing Wikidata (e.g. one just for bibliographic entries), but it seems a step in the right direction. – Joe (talk) 14:34, 3 February 2021 (UTC)[reply]

How so? If in this example, you wanted to change "Foyles" to "Dover publications" or whatever, you actually see Foyles in the code you posted, and can simply replace it. Not so in Q12512121564. I have gone back to the same example article, List of woodcuts by Albrecht Dürer, and opened it in Visual Editor (which is still so sloooow). The reference I discussed above is defined inside a table. So I go to the table, and try clicking on the [7] indicating the source. No luck, I open the table instead. Some very precise double clicking seems to do the trick eventually, but even then I can't do anything. Now, I want to copy that reference to elsewhere in the same article, but make a new version with a specific page number. Oops, I can't do this in VE (copying a reference is just anoher instance of the same ref, without the option of making a new ref from it), and shouldn't do this in Wikidata (or do we want a specific Qnumber for every page in a book?). So I try adding a new reference instead. Someone told me that simply adding an ISBN is sufficient, but the results are hardly satisfactory, with lots of double comma's in the reference so produced (e.g. "Others: Kurth, Willy, 1881-1963,, Dodgson, Campbell, 1867-1948,, Welsh, Silvia M.,). So, once again, lazy editing tools producing unsatisfactory results but being hailed as superior anyway. I'll stick to wikitext and continue opposing both VE and Wikidata (as used on Enwiki, not as a separate entity). Fram (talk) 15:14, 3 February 2021 (UTC)[reply]

Have you tried the VisualEditor recently? I find it works very well for writing and working with simple templates (citations, infoboxes, etc.) – Joe (talk) 11:45, 3 February 2021 (UTC)[reply]
Thank you, but shudder, no. I loathe that thing. Citations are for me much easier with the wikitext editor. And it doesn't address any of the issues I discussed. Fram (talk) 12:39, 3 February 2021 (UTC)[reply]
I was responding to Blueboar, not you directly, sorry for the threading confusion. – Joe (talk) 13:33, 3 February 2021 (UTC)[reply]
No problem. Fram (talk) 13:45, 3 February 2021 (UTC)[reply]
I would agree with Fram. I too prefer to edit using the wikitext editor (I like to see the full text of a citation written out when I go into edit mode). Blueboar (talk) 14:07, 3 February 2021 (UTC)[reply]
Blueboar, It's my hypothesis that a number of editors, especially those who had mastered the intricacies of adding references with the wiki text editor, checked out an early version of visual editor and were very turned off. I sympathize with that view, because I tried it early, found it lacking, and didn't use it much for a while. However, I came back to it, and find that many of the bugs have been worked out. While I still find the need to use both editors, I almost exclusively use VE for adding references and it is much better on the occasional time I need to add a table. I understand that once you've mastered the ability to add a reference manually, VE doesn't count as much of an improvement, but the ability to drop in an ISBN and spit out a properly formatted reference is a godsend. S Philbrick (Talk) 14:52, 3 February 2021 (UTC)[reply]
It's obviously feature creep and software bloat. It's a sign of the times that the billionth edit was to add an {{authority control}} template to a non-notable stub using AWB. That was just busywork – gaming the system to boost edit count. Editors who enjoy grinding naturally like using tools to amplify their activity. Writing carefully researched and cited encyclopedia text is much harder to automate and so it's not done. Another typical symptom of the problem is using an automated tool like Twinkle to drop a tag on an article rather than actually fixing the issue. Such tools tend to bias activity towards brute-force fixes like deletion rather than activity that is difficult to automate such as writing and editing. Andrew🐉(talk) 12:28, 3 February 2021 (UTC)[reply]
- I don't mind the {{authority control}} because who cares, at least it adds new information of some kind, but other than that I agree completely. I'm fine with people enjoying repetitive and simple tasks but you can do that with anti-vandalism or learning a couple of simple references you can find and add (like adding IGN reviews to video game articles). Or, even better, here's a repetitive and labour-intensive task that is a full decade backlogged. But you're not helping anyone if you slap 1000 {{more citations needed}} templates on articles and don't add a single new citation. — Bilorv (talk) 01:03, 6 February 2021 (UTC)[reply]
Reading some of the comments has me thinking... perhaps part of the problem is that that I am very TEXT oriented, and WP is becoming more and more DATA oriented. Using templates and automation does make it easier to update DATA in articles, but it makes it very difficult to write and edit TEXT. Blueboar (talk) 14:13, 3 February 2021 (UTC)[reply]
- I'm not sure what policies/guidelines you are trying to discuss here Blueboar, but WP:BOLD and WP:IAR have a long, strong history. I've told others before if you want to write about something but don't know all the nuances of the templates/markup/etc - then just write it! Use references, if you don't format the references in a way that others like they can just go refactor them. If someone says something like "you should use template:x instead of a bare reference" take it or leave it! If someone makes some template-heavy-reference that is a pain to deal with, ignore it and just put in a reference you actually used however you want. Don't edit war with someone who wants to come behind you and update the markup - move on to something else you actually care about! We need good faith editors making constructive articles and article improvements as much or more than we need the references to be in a shiny markup if it is in the way of brilliant prose. — xaosflux ^Talk 15:33, 3 February 2021 (UTC)[reply]
  - That is great advice for when someone is adding new info or creating new citations ... my difficulty is figuring out how to edit/amend existing info and citations. The more we rely on templates and coding to generate text (whether it be the text of an infobox, the text of a citation, or the running text of an article), the harder it is to fix/amend that text when something needs fixing/amending. Blueboar (talk) 18:27, 3 February 2021 (UTC)[reply]

@Blueboar:, is there an example that you can point to of a cite or template that you feel is too obtuse? The concerns you express are feeling a bit abstract and nebulous at this point. I can think of examples where I've had trouble (e.g., I have come to despise the opacity of Wikidata inclusions) but I don't know if you are referring to similar things. Eggishorn (talk) (contrib) 18:46, 3 February 2021 (UTC)[reply]

Sorry to not be more specific. I suppose this is more a generalized sense that I am increasingly unable to contribute as fully as I once could... rather than something inspired by a specific example. Wikidata is certainly contributing to this, but it goes beyond that. I will try to find other examples.

I raised my concern here because I was not sure where else to discuss it (and because I know this page has lots of people watching it). I don’t even know if there IS a remedy... I just wanted other editors to know how I felt. Something has changed in the way WP operates, and I am saddened by it. Blueboar (talk) 19:33, 3 February 2021 (UTC)[reply]

I think I know what you mean -- for me, it does take time sometimes to figure out how to add to a template/table etc so it renders the information it should, and with more different templates/tables etc. sometimes I just move-on and the improvement is does not get done, at least by me. -- Alanscottwalker (talk) 19:41, 3 February 2021 (UTC)[reply]

If you were learning how to edit templates, you would start with something simple, and work towards more complex patterns. That's how a textbook would be organised. But for the editor who comes from editing an article and wanting to change something, then you might well be jumping in the deep end. The templates, like the articles, have tended to become more detailed over time, due to precisely this process. There is quite a few You might be confronted by one of those templates that makes you recoil. It took some time, but I managed to get the rocket engine template to call the nuclear reactor one as a module. Persistence pays off. Hawkeye7 (discuss) 00:52, 4 February 2021 (UTC)[reply]
Blueboar makes some valid points, in my view – and I'm a retired computer science academic, competent in both the template language and Lua (I converted most of the WP:Automated taxobox system).
- One problem is that automation = obscurity. Consider citations. A citation written out as text is easily comprehensible and editable. Given that the parameter names are meaningful, a citation written out inside a citation template is also reasonably easily comprehensible and editable. A citation created by {{Cite Q}} is neither easily comprehensible nor easily editable. It's particularly problematic in my view when templates here use data from elsewhere, particularly from Wikidata, which has to be edited in a very different way. There should be much more community discussion before such templates come into use.
- Another problem is the complexity of templates/modules, as they are expanded to do more tasks, and do them better. Increasing modularity is good in software terms (or so I used to teach), but creates issues for editors here – one Lua module now often services multiple templates, making changes to it considerably more difficult to make safely. All this results in a shrinking of the pool of editors who are willing or able to edit templates/modules. I'm not sure whether there is a solution to this problem, but we should be aware of it. I think that in some areas we are dangerously dependent on a small number of editors.

Peter coxhead (talk) 10:39, 4 February 2021 (UTC)[reply]

Agreed with Blueboar. It's worrying how many of our most frequent editors began editing in 2010 or earlier. Because these editors will not all continue in decades to come. We need a new generation of Wikipedians and the ones we're recruiting are too technical-oriented (or politics-oriented). The learning curve is just so steep to be able to improve content with no technical background. To you, ten changes in ten years just means you have to learn one thing a year. To a new user, this means another ten things to add to an existing twenty before they can just fix the bloody date in article prose from "February 4" (a mistake) to "February 5". Or more accurately, it means that the user can't fix the date, gives up and never tries to edit again. New users don't get a voice in our conversations, because they're not here. Just the people who are so familiar with our existing workflows that they have absolutely no idea what the new user experience is like. We also forget that most people are reading on mobile and can only edit on mobile, particularly those in Africa, Asia or South America who are only just getting access to the internet. It will soon be harder than the Labours of Heracles to change "February 4" to "February 5" on a smartphone when the content is transcluded from the table on another article, which invokes a template, which invokes Lua, which invokes Wikidata, which wants you to type in flawless mm/dd/yyyy format without explaining this. Can you imagine stumbling through this process when you've never even heard of the Wikipedia namespace or found WP:5P yet? And don't forget the reference (which has to be cited in a completely different way) or you'll be reverted immediately by a long-term editor who can't distinguish your edit from subtle vandalism. — Bilorv (talk) 01:03, 6 February 2021 (UTC)[reply]
The main thing I do on Wikipedia is vandalism fighting, specifically of longstanding vandalism. Two frequent pathways to vandalism becoming longstanding are vandalism to references (at least in my experience, a lot of the "fixed vandalism from 2010" comes from people inserting obvious vandalism into refs) and vandalism grandfathered in from other articles. Things like CiteQ that pull copy from Wikidata seem like a ticking time bomb for both. (After seeing this entry it took me literally 5 minutes to find (and revert) several instances of obvious Wikidata vandalism that were there for some time.)- I know Wikidata has a counter-vandalism project but I'm not sure how active it is compared to here. Gnomingstuff (talk) 04:52, 8 February 2021 (UTC)[reply]
- Concerns over Wikidata vandalism (which I share) have been the main reason the community has opposed Wikidata integration in most forms. It's worrying to see things like CiteQ bypassing our filters by being introduced without, so far as I can see, consensus and widespread discussion outside of the techno-centred communities which are much more likely to favour it. There was a quite negative closure at Wikipedia:Templates for discussion/Log/2017 September 15#Template:Cite Q in 2017 but since then, what's changed since that makes CiteQ acceptable for transclusion on, at present, 44,000 pages (mostly articles)? — Bilorv (talk) 23:22, 8 February 2021 (UTC)[reply]

PROD and articles that share a title with articles that were previously deleted at AFD

Ok, so we know that articles that have been previously discussed at AFD are not eligible for PROD. But what if it's a different article created at the same title as an article that had previously been deleted at AFD?
Two scenarios:
Scenario A
Someone creates a bio of a singer named John Smith. The singer John Smith is found to be non-notable and the article is deleted. Time passes. Someone creates a bio of an actor named John Smith, this is a totally different subject but since they share the same name, if someone tries to PROD the John Smith article, it gets thrown into Category:Proposed deletions needing attention as an "D: Article which has undergone an articles for deletion Discussion".
Since this is a totally different article about a totally different subject, one would think it would be Prodable. This article has not been previously discussed at AFD, but since it's in that category, someone will come along and remove it.
Scenario B
There is an unsourced essay like article that is deleted at AFD. Some time passes. Someone comes along and creates a new article about the same subject but it is a totally different article. This new article is not subject to WP:G4 deletion because it is not a "sufficiently identical copy" of the deleted page.
Scenario B differs from Scenario A in that although it is a different article it is about the same subject. This article has not been previously discussed at AFD, but this subject has. Assuming Scenario A should be proddable, should Scenario B be proddable?

For the record, this discussion was inspired by Garden real estate, which fits scenario B, and which I prodded but which was deprodded by Spiderone. However, I would like this discussion to be about the policy more generally, and if we need to clarify the policy any.

So, should articles which fit Scenario A be subject to PROD? If so, should articles which fit Scenario B also be subject to PROD? If the answer to both is yes, how can we amend the Wikipedia:Proposed deletion policy to clarify this? ~ ONUnicorn^{(Talk|Contribs)}problem solving 21:56, 3 February 2021 (UTC)[reply]

Discussion

Scenario A should definitely be PRODable as it should be treated as a totally different article. Scenario B is a tricky one. I think if someone voted 'keep' in the original AfD then that would count as some amount of opposition to deletion of the topic, which would make it ineligible for any non-controversial deletion, which PROD is. If it was deleted unanimously at the old AfD and someone then created an article again on the same topic but with totally different content, then I'm not sure. In the case of Garden real estate, I removed the PROD in good faith on the basis that I believed that it being recreated so soon after its previous deletion by User:Willow4 could be perceived as 'contesting the deletion' and hence make it ineligible. Spiderone^{(Talk to Spider)} 22:04, 3 February 2021 (UTC)[reply]
I have had two articles prod, one I saw and removed, which I was repremarnded for by another editor only for several others come in and point out the article fulfilled notablility, and another which was deleted before I saw the prod, and when I challenged the editor who deleted admitted that the prod was then wrong and so the article was recreated. My issue is the Prod is really a pointless system, as several afds that I have viewed have shown, and the examples given by User:ONUnicorn has provided. I think Prod should be dropped and AFD the only way to go. Let a concensus of editors put the case for and against, and if it is bad normally a mass of delete, and that way we won't get these silly incidents that just poke fun at how silly prod is. Davidstewartharvey (talk) 22:07, 3 February 2021 (UTC)[reply]
If any article was worthy of a PROD, it's Garden real estate. It's existed for over thirteen years with precisely zero sources, it's mentioned anyway in Niche real estate, and most of its basic content is copied from here. I would just redirect it to Niche real estate and have done with it. To get back to the point though, yes - anything in Category A should be able to be PRODded. If I was going to do it myself I'd PROD it anyway, remove the category, and edit the template to inform the potential de-prodder that the previous AfD was not about this subject, but that's quite a lot of hassle that most people wouldn't think of doing. Black Kite (talk) 22:20, 3 February 2021 (UTC)[reply]
In case A, if a PROD tag was removed when it's clearly for no reason other than the old article, it may be restored (whether the old timestamp may he used, or it needs a new timestamp, depends on how long the tag was gone). 217.132.229.147 (talk) 23:35, 3 February 2021 (UTC)[reply]
In Scenario B, that shouldn't be PROD-able. At that point, I would want an explicit consensus to be determined if the article has a new iteration on the same topic. Scenario A is PROD-able as everyone else said. –MJL ‐Talk‐^☖ 05:41, 7 February 2021 (UTC)[reply]
Agree that scenario A is a completely different article and is okay to PROD. Also agree that B is a bit more complicated and that complicatedness, to me, means that PROD is not suitable. Unless it's an easy CSD for defamation or something, what's the harm in putting it through AFD? If it gets deleted again, the name can be WP:SALTed to prevent a third occurrence. Matt Deres (talk) 14:26, 12 February 2021 (UTC)[reply]
I think that in order to address a tiny sliver of cases, you're looking to complicate the rules for PROD, which exists specifically to be simple - no article done by PROD could not be deleted in some other manner. The benefit to be gained by addressing these edge cases is not worth making a new series of tests to make part of the process and thus discourage its use. --Nat Gertler (talk) 15:16, 12 February 2021 (UTC)[reply]

Notability and stand-alone lists of sports accomplishments

I stumbled across a notability issue recently in stand-alone lists of sports accomplishments, specifically cricket statistics. See, for example, Wikipedia:Articles for deletion/List of international cricket five-wicket hauls by Lance Gibbs. There has apparently been a great deal of debate recently at Wikipedia talk:WikiProject Cricket which does not appear to have resulted in a consensus. Rather, to the contrary, it appears to have generated enough dissension and hard feelings to provoke editing restrictions and for editors to leave editing that project or retire completely. I hope that by bringing this to the wider community, some clarity may be achieved. Time will tell if that hope is justified or not, I suppose.

Relevant standard

All appear to agree that the relevant standard is WP:NLIST:

Notability guidelines also apply to the creation of stand-alone lists and tables. Notability of lists (whether titled as "List of Xs" or "Xs") is based on the group. One accepted reason why a list topic is considered notable is if it has been discussed as a group or set by independent reliable sources, per the above guidelines; notable list topics are appropriate for a stand-alone list.

The issue appears to be in how to interpret this standard in the light of the available coverage for cricket centuries and five-wicket "hauls". An apparent WP:LOCALCONSENSUS that 25 centuries or some indeterminate number of hauls by one player is justification for a stand-alone list has been mentioned multiple times in recent AfD discussions but multiple discussions at WP Cricket do not seem to have pointed to any place where such a consensus was first formed (e.g.:Wikipedia_talk:WikiProject_Cricket/Archive_84#Steve_Smith_stats Wikipedia_talk:WikiProject_Cricket/Archive_85#Unanswered_question_since_2015 Wikipedia_talk:WikiProject_Cricket/Archive_85#What_to_list_and_not_list Wikipedia_talk:WikiProject_Cricket/Archive_88#Lists_of_International_Centuries Wikipedia_talk:WikiProject_Cricket/Archive_86#Discussion_on_List_of_fifers/centuries_by_ground:_keep,_delete_or_merge_by_country?. The closest appears to be this discussion which resulted in a consensus that a list of either of these accomplishments for a particular venue requires ten or more entries but is not the same question.

Notability in practice

The actual application of the relevant standard in terms of demonstrable outcomes at AfD is extremely variable. AfD discussions of "List of international cricket [five-wicket hauls/centuries] [by/at] [X]" articles that were closed between Sep 1, 2020 and now shows the following distribution:

Keep: 2
Delete: 62 (including 2 large multi-AfD's)
Merge:9
Redirect: 7
Draft:0
No consensus: 0

There are currently fourteen such AfD discussions open, all started by Störm. This may be an attempt to establish an consensus through WP:OUTCOMES that has so far eluded the Wikiproject. When the Multi-AfD's are included, then deletion is the overwhelmingly most common outcome (87%). When only single-article AfD's are considered, this shows that recent practice is appx 41% in favor of merging such articles and 32% to redirecting. The disparity in these results argues in favor of a discussion forming an explicit consensus.

Proposal

A clear standard for this type of list article is apparently lacking. Either the Cricket Wikiproject or a Notability Guideline should include something similar to the following:

"List of" articles for ~~cricket~~ sports accomplishments such as centuries or ~~fifers~~ home runs are only considered notable if there are independent, reliable sources that significantly discuss the accomplishment in question for an individual or a venue as a group or set. Such sources should be more than mere statistical listings and the data should be put in context with referenced explanations, if necessary.

This text is an attempt to create one such clear standard for this type of article that complies with the Core Content Policies, the General Notability Guideline, and other relevant standards. Please state a preference for the proposed standard for "List of five-wicket hauls/centuries" type articles or propose an alternative. While this particular proposal is occasioned by a cricket-specific issue, there may also be reason to think that such a guidance that applies more widely in sports may be advisable. Thank you for your time and attention. Eggishorn (talk) (contrib) 20:49, 5 February 2021 (UTC)[reply]

Modified per below. Eggishorn (talk) (contrib) 00:10, 6 February 2021 (UTC)[reply]

Notability discussion

I appreciate the idea here, because these controversial cricket AfDs are becoming a problem, but your proposal is cricket-specific. Most editors don't care about cricket, and if you make a solely cricket-related proposal at a general noticeboard, most of the participants are going to be the same editors that are grinding against each other already. It may be that we can come up with an actual proposal to cover sports statistics better, but it feels like this is just another attempt to litigate the same cricket notability standards that nobody seems to actually agree upon. We already have a notability criterion for standalone lists, and to be honest I think that mostly covers these cricket stat lists in most of the way you're proposing - have reliable sources discussed "Johnny Cricketplayer's five-wicket hauls" as a group or set? If not, probably not a good standalone list. ~ mazca ^talk 23:24, 5 February 2021 (UTC)[reply]

@Mazca:, I brought it to this board specifically because the previous participants appear to have been unwilling or unable to come to some agreement on the issue. I acknowledge your point about broader applicability, though, and mentioned it above. I've made a slight adjustment to make that clearer. Thanks for the feedback. Eggishorn (talk) (contrib) 00:13, 6 February 2021 (UTC)[reply]

@Eggishorn: I think that minor adjustment is a significant improvement, yeah, as I think the emphasis of any attempt to gain general consensus here needs to remain general to avoid being bogged down in per-sport exceptions. I don't think there's a good argument for arbitrary run thresholds, etc, in cricket granting notability beyond that general principle of "being discussed as a group or set by reliable sources". I absolutely agree with the idea of deciding global thresholds for sport statistics - we just need to make sure it's not some specific solution for cricket. ~ mazca ^talk 02:44, 6 February 2021 (UTC)[reply]

Considering at a broader level beyond cricket, I think these types of records at the individual player level may be far too much, per WP:NOT#STATS - even if there is good coverage of this from sources. While certainly the five-wicket haul seems like a notable achievement that is documented and thus a list like List of cricketers by number of international five-wicket hauls, documenting down to the specific player/each individual point of achievement seems far too much. I would say the same for an article like List of milestone home runs by Barry Bonds (while obviously List of Major League Baseball career home run leaders is a fair list to keep). --Masem (t) 23:31, 5 February 2021 (UTC)[reply]
I have been involved of several of the AFDs surrounding cricket. There seems to be a lot of anger amongst those who to delete, and those who want to keep. The main bone with deletists is that list of stats are WP:NOSTATS, that these lists are not not encyclopedic and Wikipedia should not have statistics that can be found elsewhere, or that these lists are not encyclopedic as there is not enough prose and to much stat. The keeps are saying this is a vendetta after a failed RFC change to WP:NCRIC. My own personal view has been to merge these into the main article, and I have voted as such on 9 that I have seen, which all ended in the 9 merge, and I have posted to others that have been raised that this was the case. The lists are not exhaustive lists, they are for international 5 wicket hauls or centuries only, which are important milestones in cricketers careers and are revered by fans alike. Just look at the coverage surrounding Joe Root in Sri Lanka and India. This is why delete is wrong. My rational is that who is going to search for a List of International Five Wicket Hauls? I wouldn't, I would just go to the Wikipedia page for that player. My other rational is that it's not a ball by ball record for their whole career, just what is regarded as being the highest accolade in cricket. That's my viewpoint on the subject.

My beef is the wording "Notability guidelines also apply to the creation of stand-alone lists and tables. Notability of lists (whether titled as "List of Xs" or "Xs") is based on the group. One accepted reason why a list topic is considered notable is if it has been discussed as a group". The group is the issue. We have Editors who believe they are the only people who have the right to decide what is notable for Wikipedia, they forget the second part "or set by independent reliable sources, per the above guidelines; notable list topics are appropriate for a stand-alone list." They also believe that they should set out what should go on a list. So for example, a nation wide company is not allowed in a list or national retailers because they don't have a page on Wikipedia. Sorry for the rant.

Anyway, the problem I see is that if we have a separate subject specific list for cricket, where will this lead us? It will go mad and every editor with bones to pick will be requesting there own list requirements for each subject Area! Keep it as it is. — Preceding unsigned comment added by Davidstewartharvey (talk • contribs) 00:10, February 6, 2021 (UTC)

@Davidstewartharvey:, I think I have miscommunicated and I need to clarify some things in response.

As I said, I stumbled across this because I troll through AfD's that are not closed occasionally and I haven't been involved previously. I think that it is because of the animosity that you mention that this needs to be handled outside the Wikiproject.
Although this is proposed because of what I found through the cricket AfD's it is not about cricket articles. It is about list articles for stats in any sports. That's why I titled the section "Notability and stand-alone lists of sports accomplishments" and not ..."cricket accomplishments"
Thanks again to Mazca for prompting a clarification to the actual proposal to make the above point clearer.
The wording you object to is not my wording. It is a direct quote from the long-standing notability standard that should already have been followed in all AfD discussions of these list articles. The fact that it has not been followed in such discussions is the reason I am trying to propose a clarification.

I hope that clarifies some things. Eggishorn (talk) (contrib) 16:00, 6 February 2021 (UTC)[reply]

POVFORK? Deletion of these types of lists can happen in sports (Wikipedia:Articles for deletion/List of 40-plus point games by Michael Jordan). The main issue, generically and not sports-specifc, is a grouping that seems to meet WP:LISTN, as it's currently written, when most of the items themselves are not notable. In sports, it's often a list for a player of all games where they met a single-game threshold. However, the games themselves are generally not notable. This is different than when we have a list of players that met a milestone, where most of the entires are players who are notable. IMO, these types of lists of non-notable entries are WP:POVFORKs. It's not notable enough to warrant the bloat in the bio, so a standalone of the stats listing is created instead. Lists of non-notable entries is not unique to sports. There's plenty of lists like List of killings by law enforcement officers in the United States, January 2020, a grouping of individually non-notable entries.—Bagumba (talk) 16:22, 6 February 2021 (UTC)[reply]

Something else to consider... a sports statistics list may not be notable enough for a stand alone article, but may well be appropriate to include as part of a related article. For example, a list of the top cricket players by number of centuries might well be appropriate in the article on Century (cricket). And it might be appropriate for the Michael Jordan article to include a list of his highest scoring games. Blueboar (talk) 17:05, 6 February 2021 (UTC)[reply]
For NBA bios, an FA would typically mention the top games in prose with proper context, and not repeat them with an embedded stats list.—Bagumba (talk) 01:55, 7 February 2021 (UTC)[reply]

So, this seems a part of the standard "Expand->split->merge" cycle that we get all the time:

An article grows to where it is too big, so it needs to be split up by topic. This is standard practice.
The split off article gets someone who complains that it should never have been split in the first place, and it gets deleted.
The information has to go somewhere, so it gets put back in the main article.
Rinse and repeat

Either the information belongs at Wikipedia, and if so it shouldn't really matter where it goes as long as it doesn't make articles too big or too small, OR it fails WP:V or WP:UNDUE and as such shouldn't be at Wikipedia at all. Without saying whether or not the information does belong, if it actually does, and it's too much to put in the parent article, I see no problem with a split-off list article. If it doesn't belong at Wikipedia at all, the point is moot. --Jayron 32 18:51, 10 February 2021 (UTC)[reply]

Exactly. I typically think AFD is a poor place to resolve these because that forum is biased towards binary keep/delete results, and encourages judging articles in isolation when obviously these are part of a larger whole, notwithstanding the formatting into a separate page. Resolving whether a split should be maintained requires understanding of all of the interrelated content and pages, and depends on some familiarity with the subject matter to determine what is relevant and what level of detail is valuable to a reader on different subtopics. And there's also WP:ATD. postdlf (talk) 19:29, 10 February 2021 (UTC)[reply]

@Postdlf:, unfortunately, it appears that AfD is exactly where this issue is being "resolved" because the relevant Wikiproject reached a stalemate. Do you agree that the proposal should be made an explanatory supplement or something similar? Thanks. Eggishorn (talk) (contrib) 15:45, 12 February 2021 (UTC)[reply]

No, it seems like a subject-matter specific legislation of LISTN, which is already far too often misread as necessary rather than merely sufficient. If there's a "stalemate" then apparently there's no consensus in that Wikiproject against such lists. I also don't see why it's worth wasting anyone's time trying to get rid of them. postdlf (talk) 21:17, 12 February 2021 (UTC)[reply]

MEDLEAD

More input is needed at Wikipedia talk:Manual of Style/Medicine-related articles#MEDLEAD. Crossroads ^-talk- 06:49, 9 February 2021 (UTC)[reply]

The role of ArbCom relative to sourcing content in articles

In September 2019, the Arbitration Committee in good faith made the following decision:

5) The sourcing expectations applied to the article Collaboration in German-occupied Poland are expanded and adapted to cover all articles on the topic of Polish history during World War II (1933-45), including the Holocaust in Poland. Only high quality sources may be used, specifically peer-reviewed scholarly journals, academically focused books by reputable publishers, and/or articles published by reputable institutions. English-language sources are preferred over non-English ones when available and of equal quality and relevance. Editors repeatedly failing to meet this standard may be topic-banned as an arbitration enforcement action.

While ArbCom's intentions are honorable regarding that particular topic, and it was clearly an effort to stop disruption in that topic area, I'm of the mind that the remedy is an overreach of ArbCom's scope, and is noncompliant with policy for the arbitration process, specifically that the Committee does not rule on content. With that in mind, see WP:CONTEXTMATTERS which states: The reliability of a source depends on context. Also keep in mind that AE/DS authorizes individual admins to take unilateral action at their sole discretion against an editor, and as some active editors who edit in controversial topic areas will attest, the complexity of DS can quickly turn into a nightmare. An AE/DS action by an admin cannot be overturned by another admin. That opens the door to WP:POV creep because a single admin is making the determination as to whether or not the context of cited sources are reliable for inclusion of content in an article, and no matter how we spin it, that is unequivocally a content issue, not a behavioral issue on behalf of the editor; therefore, the ArbCom ruling is noncompliant per the following:

The arbitration process is not a vehicle for creating new policy by fiat. The Committee's decisions may interpret existing policy and guidelines, recognise and call attention to standards of user conduct, or create procedures through which policy and guidelines may be enforced. The Committee does not rule on content, but may propose means by which community resolution of a content dispute can be facilitated.

This decision directly effects content on so many levels that the remedy creates an even bigger problem then it resolves as it fails to take into account the critically important aspects of sourcing that unequivocally makes it a content issue, not a conduct issue. It also doesn't factor-in the associated problems resulting from political biases in RS, or the fact that a single admin who may be biased, unknowingly or otherwise, is making the decision as to what content can be included based on their sole discretion/interpretation of the cited source(s). Any editor who might be citing sources in proper context that disagree with a particular POV, may inadvertently become the target of a indef t-ban or block under this AE remedy. WP's ideological bias is no secret so in an effort to preserve WP's reputation as a neutral encyclopedia, we probably should at least try to keep ArbCom within its scope. As to my overall position and concerns, see this BLP comment and the quote directly below it.

In the survey below, either Agree that the above mentioned decision is out of ArbCom's scope and noncompliant with policy for the reasons stated above, or Disagree. Atsme 💬 📧 16:07, 9 February 2021 (UTC)[reply]

Survey (Arbcom and sourcing)

There is no clear distinction between creating policy and interpreting policy. Sometimes it help to take uncodified and unwritten expectations and put them into formal writing, But for many of the perennial issues at WP, the unwritten guidelines are too flexible to be effectively reduced to simple sentences. So when we do, it still leaves the same problems, except that we now call them interpretations. The most obvious analogy to me is the American Constitution and especially its Bill of Rights, and how key phrases in it have never been changed, but have been reinterpreted repeatedly to fit the prevailing ideology. DGG ( talk ) 20:15, 9 February 2021 (UTC)[reply]

Apologies, DGG but is your comment to agree or disagree? Atsme 💬 📧 12:01, 10 February 2021 (UTC)[reply]

as I try to explain a little later, agree or disagree doess not do justice to the actual complexities. DGG ( talk ) 06:11, 18 February 2021 (UTC)[reply]

Not a productive use of anyone's time See below for explanation. Eggishorn (talk) (contrib) 20:17, 9 February 2021 (UTC)[reply]
Seems fine to me. There was a behavior that needed addressing, and the quoted ArbCom decision addresses that behavior and does not address content; they are not making any statements about what the articles should or should not say, merely addressing the problem of people misusing source material and devising a reasonable method of stopping that problem. --Jayron 32 18:44, 10 February 2021 (UTC)[reply]
I'm opposed to the camel's nose under the tent expansion of ArbCom remit into editorial issues until such time as the community assesses whether we have to drop our opposition to a centralized editorial board, and if so, think through whether a centralized editorial board ought to be a new group comprised of editorial experts or dumped into ArbCom's lap. (I elaborated on this in a post in the discussion section.)--S Philbrick (Talk) 19:31, 10 February 2021 (UTC)[reply]
I'm with Jayron32 on this: this isn't a content decision. The policy on disruptive editing includes citing unencyclopedic sources as a form of disruptive editing, and warns that this may be enforced with blocks. ArbCom isn't dictating what an article can or can't say; they're delegating their banning authority to help better enforce the community's policies about content. Vahurzpu (talk) 05:12, 13 February 2021 (UTC)[reply]
While I don't think this specific restriction has worked as well as intended, it was within arbcom's remit to make it. Thryduulf (talk) 12:23, 18 February 2021 (UTC)[reply]

Discussion (Arbcom and sourcing)

See, I always thought that a lot of the Ideological bias on Wikipedia and complaints about it is actually the base rate fallacy - because of our sourcing requirements we are automatically biased in favour of topics with reliable sources as opposed to, say, hearsay and libel. And that's a desirable thing, really. Jo-Jo Eumerus (talk) 16:20, 9 February 2021 (UTC)[reply]

Jo-Jo, it would definitely be desirable if that were truly the case, but it isn't. Even Jimbo acknowledged that we have a problem. The sources at our disposal today are not quite like the neutral sources of yesteryear, as Ted Koppell explained in numerous interviews. It's clearly a POV issue but that aside, we don't have any type of qualifiers in place for admins who focus on AE and who will be judging the quality of sources. We already know that qualifiers will never happen because of anonymity, although something along the line of MEDRS would be nice, but again, topics involving POV politics are not easy to qualify. Atsme 💬 📧 16:36, 9 February 2021 (UTC)[reply]

"Even Jimbo acknowledged" doesn't mean that Jimbo is correct. WhatamIdoing (talk) 18:25, 10 February 2021 (UTC)[reply]

The "ArbCom does not rule on content" meme is one that should be definitively retired as having outlived its purpose. To the extent that there ever was a clear line between content and behavior, an iffy division at its very best, that line has been thoroughly obliterated in the last 5-7 years. If an editor advocated for or uses a poor source to slant an article, is that a behavior issue within ArbCom's purview or a content issue outside the realm of cases ArbCom is authorized to handle? The plain truth is: it is both. This case offers an excellent demonstration of that evolution, in fact. At a time when Holocaust minimization and denialism has reached epidemic proportions that include governmentally-organized revisionism, it is ludicrous to contemplate any ruling on such a case that did not address sourcing concerns. ArbCom's ruling here does not require a 16-months-delayed attempted review. This thread appears to do little more than express, well, I'm not quite sure what it is supposed to express: Frustration? Disappointment? Disapproval? It is unclear what practical result this would accomplish. Does the OP wish to overturn ArbCom's decision? Get ArbCom to modify their decision? Wag their finger at ArbCom? If it is the former two possibilities then this is the wrong place for this thread and that goal needs to be much more clearly stated. If it is the latter possibility then there is no point to this. Eggishorn (talk) (contrib) 20:17, 9 February 2021 (UTC)[reply]
Eggishorn, I see it differently. New editors are often surprised to learn that there is no central editorial board. This is deliberate. In fact, I think it's fair to say that in the early days of Wikipedia we eschewed having either a central editorial board or a central "behavioral" board. The hope and expectation was that all such issues could be addressed without the need for centralized authority. I think Jimbo reluctantly came to the conclusion that behavioral issues needed a centralized board and that's why he created Arbcom (while taking care to make sure the creation of it did not mean it reported to him, he started the ball rolling and let the community decide how to organize the group). While the community reluctantly accepted the need for a behavioral board, it still strongly opposed a centralized editorial board which is why ArbCom is supposed to be limited to behavioral issues not content issues.

However, while some editors are making and can make strong arguments for the need for a centralized editorial board, it isn't at all obvious to me that we should simply broaden the remit of ArbCom. I think the community should revisit the decision to have no centralized editorial committee, but if the conclusion is that we should have an editorial committee, it is far from obvious that we should simply expand the role of ArbCom, if no other reasons than that the skill set to assess editorial issues and the skill set to assess behavioral issues are not identical. Someone will argue that it's overly bureaucratic to create yet another group, but if we decide to have centralized editorial control we've crossed that bridge and we ought to work out the details of whether the centralized editorial committee obviously ought to be ArbCom, or group of people selected for the editorial expertise. S Philbrick (Talk) 19:28, 10 February 2021 (UTC)[reply]

@Sphilbrick:, actually, I don't think we're saying things that are really all that different. I'm aware of the history of ArbCom's creation and I've always felt that the content/behavior dichotomy is a false one. The earliest ArbCom case i was able to find from the time period soon after Jimbo passed some of his authority to it was a clear case of interpersonal conflict and a whole heapin' helpin' of behavior issues but they all stemmed from content issues. From the very beginning the two areas have blended together in varying proportions depending on the particulars of each ArbCom case. I'm not arguing for an increase of ArbCom's remit; I'm saying that this discussion was premised on a clear discrimination between two realms that exist along a spectrum and the limitation that ArbCom supposedly stepped over never really existed. I hope that helps clarify. Eggishorn (talk) (contrib) 19:47, 10 February 2021 (UTC)[reply]

The simplest change in the general scope of arb com, which is closer to reality, is to say that arb com does not rule directly on content. That's what actually happens, and we might as well say so. What I think the op might realistically hope for is that the decision say High quality sources are strongly preferred, particularly peer-reviewed scholarly journals, academically focused books by reputable publishers, and/or articles published by reputable institutions I do not think it unrealistic for this to be suggested as an amendment. Any statement here about sourcing using absolute words like "only" is incompatible with the variety of the real world. DGG ( talk ) 00:06, 10 February 2021 (UTC)[reply]
@DGG: that just reiterates WP:BESTSOURCES and is not a remedy (more like a principle). "Should", "strongly preferred" and similar wordings which are not "must", do not help the problem. The issue is people who are so compelled by their POV they bludgeon people and disrupt talk pages by pushing crappy sourcing. Your amendment does not fix that issue, but the current ArbCom decision does. As such, I'm in favour of sourcing restrictions, having seen the damage and sheer waste of productive editor time SPAs and POV pushers cause on some controversial articles. I agree that with Eggishorn and yourself that the line between conduct and content is blurry, but I think violations of our existing polices, or the intentions behind them, is a behavioural issue. ProcrastinatingReader (talk) 20:49, 17 February 2021 (UTC)[reply]

one can equally disrupt talk pages -- and articles -- by contrived or biased objections to what is adequate sourcing. I think the history of AP this past year makes it evident that disputing sourcing has become the primary way of disputing content, and that this can be done in a helpful or in a unhelpful way. In another light, everything is a behavioral issue in a sense, because the act of making an edit is human behavior, so if one tries hard enough, one can get anything under that umbrella. Whether something is in fact a violation of our existing content guidelines is not necessarily a question for arb com. It can be a factual question of the verifiability of information. And, come to think of it, one can use aggressive and even NPA behavior to try to make what are in fact constructive edits. And this very situation is what causes dilemmas for arb com; my service there has shown be the great difficulty of deciding how to handle such issues, and has taught me that not everything I had thought obvious and clear about our practices really was all that simple. DGG ( talk ) 06:09, 18 February 2021 (UTC)[reply]

I was the only arb who voted against that remedy, precisely because I thought it went too far towards ruling on content. And I still think that sourcing restrictions imposed by ArbCom, AE, ANI, or anything except a content guideline with broad consensus (e.g. WP:MEDRS) are a bad idea. That said, I don't agree that it was in breach of WP:ARBPOL. The committee has the latitude to propose means by which community resolution of a content dispute can be facilitated and ultimately, the interpretation of ARBPOL and whether a specific remedy is within its scope is up to ArbCom itself. That is an important principle without which the committee cannot fulfil its core purpose as a final binding decision-maker. No discussion here can tell ArbCom how to interpret ARBPOL or retrospectively overturn a remedy. If we want to clarify that sourcing restrictions are not within its scope going forward, there needs to be a formal amendment to ARBPOL. – Joe (talk) 13:48, 18 February 2021 (UTC)[reply]

Knockout brackets in sports events

In tennis, snooker, and other events commonly decided in a knockout format, it is common on wikipedia for people to enter the flag of the country of the winner before the winner is known. For example, here you can see a Czech flag entered in round 4, event though the 3rd round match between Karolina Pliskova and Karolina Muchova hasn't taken place yet (as I write this). I've tried explaining that it's confusing, it looks unprofessional, no reputable sports publication does it, it makes about as much sense as entering "Karolina" in the next round, it takes no account of possible double disqualifications, or both players being sick or withdrawing or otherwise unable to play etc... but still people do it, and revert it whenever I raise these points, sometimes using "other stuff exists" type arguments. (It's often IP's who do this). Usually the issue is resolved within a few days when the match actually takes place, meantime the article looks like crap.

Maybe the MOS needs a section on how to report knockout results from sports events? Maybe I'm like a grammar pedant who reacts with horror to "different than", but those anticipatory flags irk me. MaxBrowne2 (talk) 09:15, 12 February 2021 (UTC)[reply]

I agree that that example is not good. To be honest, it looks almost like a loading error, where the name has not populated for some reason. If this is indeed not just a one-off, I'd support deprecating that kind of thing. Matt Deres (talk) 20:38, 12 February 2021 (UTC)[reply]

Look at that article's history and you can see what happens if you raise the issue, they just revert without even putting up an argument or even an edit summary. No BRD, nothing. I'm tempted to add "Karolina" to the player's name for round 4 but that would be just a little too WP:POINTy. MaxBrowne2 (talk) 00:13, 13 February 2021 (UTC)[reply]

I don't know whether it's written down but adding known flags is the accepted practice for tennis. It must be exceptionally rare that nobody advances from a scheduled match. I followed tennis for many years and don't recall it ever happening. There were a few tournaments where the final was never played due to rain delays but that is irrelevant here. I support adding the flag but definitely not a partial name, and I have never seen it done. Most publications don't use flags or write nationalities in draws so the issue is not relevant for them. PrimeHunter (talk) 00:59, 13 February 2021 (UTC)[reply]

Sounds like Appeal to tradition to me, a fancier name for "other stuff exists". MaxBrowne2 (talk) 01:04, 13 February 2021 (UTC)[reply]

If the "other stuff" is not selected examples but a well-established practice then it's how Wikipedia works, and often the basis for making it an official guideline. PrimeHunter (talk) 01:10, 13 February 2021 (UTC)[reply]

Obviously it would be ridiculous to enter "Karolina" before the result is known. That's the point. By the way Karolina won. I think she's from the Czech Republic. MaxBrowne2 (talk) 02:40, 13 February 2021 (UTC)[reply]

I don't think it needs a MOS section specific to this issue. WP:CRYSTAL is sufficient reason to exclude the flag. - Ryk72 ^talk 02:43, 13 February 2021 (UTC)[reply]

I agree with the OP that we shouldn't be pre-populating flags in this type of situation. If people generally do this, they should stop. power~enwiki (π, ν) 02:49, 13 February 2021 (UTC)[reply]

The television coverage tends to pre-populate the flags when showing the draws as well. Regardless, it's not a big issue. It's never more than a few days between matches. Sportsfan77777 (talk) 02:57, 13 February 2021 (UTC)[reply]

I also agree flags shouldn't be pre-populated, we don't need to add it to MOS because it's already specified in CRYSTAL, and if it's been done in the past it should stop. Levivich ^harass/_hound 04:37, 13 February 2021 (UTC)[reply]
WP:CRYSTAL includes: "Individual scheduled or expected future events should be included only if the event is notable and almost certain to take place". It is almost certain that if two players are set to meet then one of them will advance. We don't need a published reliable source to make that "prediction". It's not certain the advancing player actually plays the next match but if they withdraw with injury before that match then their name and flag will remain in the draw for the match. PrimeHunter (talk) 09:32, 13 February 2021 (UTC)[reply]
CRYSTAL then goes on to say "Avoid predicted sports team line-ups, which are inherently unverifiable and speculative." Levivich ^harass/_hound 17:40, 14 February 2021 (UTC)[reply]
A predicted sports team line-up is not a prediction about which teams will play but about which players will be selected for a match by a team. Such speculation is hardly comparable to predicting that one of the two players will advance from a tennis match and not change nationality before the next round. PrimeHunter (talk) 22:26, 17 February 2021 (UTC)[reply]

I'd argue this also has some crossover with WP:LIVESCORES which is also against consensus, but a difficult thing to actually get the community to fulfil. Best Wishes, Lee Vilenski ^{(talk • contribs)} 17:46, 14 February 2021 (UTC)[reply]

I personally think that results should be presented for a complete match, and not a partial result. Filling in the country before the match is done is a partial result. Additionally, it doesn't provide any info that readers can't infer for themselves. isaacl (talk) 23:10, 17 February 2021 (UTC)[reply]

I think this is much ado about nothing. Would I populate the flag early.... no. Would I put in the scores before that match is complete...no. But the news does both and it's really really difficult to police that sort of thing when an hour later it will fix itself. Fyunck(click) (talk) 20:05, 18 February 2021 (UTC)[reply]
An hour later is not really a big deal (although it still shouldn't be done). Sometimes though matches can be days or even weeks later - that is a big deal and should be reverted if done. We are an encyclopaedia not a newspaper, television broadcaster or other sports reporter. Thryduulf (talk) 21:33, 19 February 2021 (UTC)[reply]

Wikidata linking

Why EPWP users do not linking their articles to Wikidata or even worse, they creating new items instead of linking to already existing items? Eurohunter (talk) 12:48, 14 February 2021 (UTC)[reply]

Because Wikidata has issues, and does not mesh with WP policies and guidelines. Blueboar (talk) 13:13, 14 February 2021 (UTC)[reply]

@Blueboar: There is no any problem to add new item or chceck if item already exists and connect. Eurohunter (talk) 13:47, 14 February 2021 (UTC)[reply]

Linking is done from Wikidata, so I think you are asking why English Wikipedia editors don't also spend their time editing Wikidata. There are probably several reasons, but the main one is probably because they are busy editing English Wikipedia. Most people here only edit this wiki. For most people, Wikidata is a wholly different concept from an encyclopaedia. Personally, I find the site unfathomable. It's like it was work designed for bots (which sometimes appear to be struggling). We do have an information page: Wikipedia:Wikidata which you're welcome to improve. The advice basically amounts to "know that an article exists in Swahili". That's not very credible for a variety of reasons. -- zzuuzz ^(talk) 14:40, 14 February 2021 (UTC)[reply]

@Zzuuzz: But previously you used interwiki so? It's just 2-3 clicks so what is the problem? Eurohunter (talk) 14:43, 14 February 2021 (UTC)[reply]

All sorts of people used to add interwiki links, often editors from other wikis (or usually, bots from other wikis) and not the main authors. In other words, it was done as if it was done from Wikidata. Finding articles in other languages is not always an easy task. And again referring to the information page, "if you cannot find a concept under a certain name, it still may exist under another name. Hint: use unique identifiers from related databases to confirm that a concept does or does not exist". Most people would be like, "I'll just write the article". -- zzuuzz ^(talk) 14:56, 14 February 2021 (UTC)[reply]

@Eurohunter: Wikidata certainly has issues, but I disagree with some of the others above that it's a "wholly different concept". Once it addresses its vandalism/referencing challenges, it will be a fantastic resource to centralize and automatically update certain types of information. Regarding the specific issue you raised, I'd suggest raising it at wikidata:Wikidata:Project Chat, where you'll get engagement from people much more likely to know how to solve the problem. If you find a specific instance of a duplicate, see wikidata:Help:Merging. {{u|Sdkb}} ^talk 21:10, 14 February 2021 (UTC)[reply]

Discussion at Wikipedia:WikiProject Council/Proposals/Edit requests

You are invited to join the discussion at Wikipedia:WikiProject Council/Proposals/Edit requests. P,TO 19104 (talk) (contribs) 14:18, 14 February 2021 (UTC)Template:Z48[reply]

Disambiguation cross-references

In the disambiguation page Pitt-Rivers, there is a list of people who happen to be from one family. Some names recur (EG George). I recently edited it to change "* Michael his son" into "Michael George's son". Is there a preferred way of identifying WHICH George in such a page? -- SGBailey (talk) 09:35, 16 February 2021 (UTC)[reply]

Maybe append the (birth-death) dates to the name? — GhostInTheMachine ^{talk to me} 12:13, 16 February 2021 (UTC)[reply]

I've done that at Pitt-Rivers. What do you think? -- SGBailey (talk) 12:49, 17 February 2021 (UTC)[reply]

That helps with the confusion, but I would not use the sub tags — GhostInTheMachine ^{talk to me} 22:59, 17 February 2021 (UTC)[reply]

Clarification to Template:POV

At this article talk I and Julian Brandon came to conflict over wording of Template:POV#When_to_remove. I think it would be appropriate to change this wording from

This template is not meant to be a permanent resident on any article. You may remove this template whenever any one of the following is true:

There is consensus on the talkpage or the NPOV Noticeboard that the issue has been resolved.
It is not clear what the neutrality issue is, and no satisfactory explanation has been given.
In the absence of any discussion, or if the discussion has become dormant.

To this:

This template is not meant to be a permanent resident on any article. You may remove this template whenever any one of the following is true:

There is consensus on the talkpage or the NPOV Noticeboard that the issue has been resolved.
It is not clear what the neutrality issue is, and either none or only an unsatisfactory explanation has been given.
In the absence of any discussion, or if the discussion has become dormant.

The template should not be removed if there is an active ongoing discussion in which "the issues are resolved" consensus has not been reached.

Please confirm and if you think it is an OK change then please implement it in the template page. --Gryllida (talk) 22:40, 17 February 2021 (UTC)[reply]

What problem would be solved by this change? Arguing with an SPA about an article on the president of an authoritarian regime is never going to be easy and I do not see how an adjustment to the description of the POV tag would help. Re the proposal, the original looks good to me—it is simple and clear. I don't know how to handle an SPA with unlimited time but forcing a tag on the article is not the solution. Johnuniq (talk) 23:00, 17 February 2021 (UTC)[reply]

At what stage would it be appropriate to force a tag onto the article? It currently is of rather poor quality. Gryllida (talk) 23:14, 17 February 2021 (UTC)[reply]

I don't think it is ever appropriate to force a tag into an article. Bear in mind that while it might be desirable to have rule that a good editor can force a tag when arguing with an SPA, such rules won't fly and there cannot be a situation where an SPA can force a tag by persistence. Asking for help at noticeboards is all we can do. Johnuniq (talk) 23:20, 17 February 2021 (UTC)[reply]

Okay, good to know. Thanks for the response. Gryllida (talk) 23:29, 17 February 2021 (UTC)[reply]

Bad idea. Would just be a means of permanently tagging an article that a POV pusher doesn't like. Hawkeye7 (discuss) 23:22, 17 February 2021 (UTC)[reply]

Hi Hawkeye7. Thank you for your reply. Good correction to my initial interpretation. I didn't do this before and just picked it up in the helpme queue, would rather get to the finish of it somehow. With the tag off to be fair it might require a bit more editing. Again, appreciate the tip. Gryllida (talk) 23:46, 17 February 2021 (UTC)[reply]

Isn't a lack of an explanation also an unsatisfactory explanation? --Izno (talk) 01:24, 18 February 2021 (UTC)[reply]

Hi Izno. In this case, the other contributor argued that satisfactory explanation was given, but they also instantly resolved all issues listed in this explanation. This is adorable and brilliant. Their fix, however, lacked some depth; some issues remained; and I can't afford to reply to them instantly to detail. That leaves the article in a bad state, without a tag, for however long it takes someone to either re-articulate or fix the remaining issues. For a BLP I find this a bit disturbing and my first thought, initially, was that the tag got to stay. (For this particular incident, I've made a draft version in my sandbox, which has some of these issues fixed at the cost of breaking the article flow, and is not publishable; then I invited the other contributor to finish it up; so far, no response.) Gryllida (talk) 02:56, 18 February 2021 (UTC)[reply]

Remove the second condition altogether.

if no clear reason exists, people are going to hash it out on the talk page. The tag should remain up until people figure out it was a drive-by tag. Then, the first condition becomes true (consensus exists that the article meets NPOV), and someone should remove the tag.
if neither a clear reason nor a discussion exists (very few interested parties), the last condition is true, and someone should remove the tag.

If the tag exists to notify readers of an active discussion about an article's neutrality, the first and last bullet points suffice, and the middle unneeded. Rotideypoc41352 (talk · contribs) 05:55, 18 February 2021 (UTC)[reply]

WP:VANISH, WP:CLEANSTART, and privacy concerns

I am in a public-facing field where harassment and doxing is common, and have witnessed it happening to many of my colleagues in the past several years. As such, I've tried to do basic identity hygiene, closing old accounts, attempting to get taken off people search, and otherwise removing any information that can be linked to my identity. (I realize this sounds paranoid, and I'm sorry I can't be more specific.)

Unfortunately Wikipedia presents a problem in this regard. I made my original account in late high school/early college, when I was not in such a public-facing field. I know for a fact that the username, though not my name, is traceable to my real-world identity and, with some work, vice versa, and given that the account has somewhere in the tens of thousands of edits, I'm sure there are plenty of breadcrumbs there to other past usernames, communities, and identifiable information in the wrong hands. In an ideal world for me, the account would be nuked from orbit or at least renamed to something unique to Wikipedia.

Nevertheless, if I am reading them right, WP:VANISH and WP:CLEANSTART seem to be mutually exclusive. I don't recall being involved in any vandalism, blocks, major conflicts, or anything that would contribute to a negative reputation, and to my knowledge was in good standing. (The "I don't recall" and "to my knowledge" isn't me trying to evade anything, just that I don't remember everything that happened ~15 years ago as a high school junior.) I have no intention of using the old account, as that'd defeat the entire purpose, and as my username implies I am mostly interested in behind-the-scenes improvement, rather than any kind of subject matter-specific editing. (Which is also less traceable to my identity.) But nevertheless the options seem to be leave Wikipedia altogether, or live with the fact that my old info is just going to sit there indefinitely like a ticking time bomb.

So, I was wondering if the mutually-exclusive part of the policy could be revisited, at least for people whose clean start is for privacy, not reputational issues. It seems relevant that Wikipedia is over 15 years old at this point, the policy was created towards the beginning, and the average person's life circumstances change a lot more in 15 years than in 5 or so. It also seems relevant that online harassment is quite a different beast in 2021 than it was in 2007 and has grown to an extent that few people predicted at the time. Gnomingstuff (talk) 10:37, 18 February 2021 (UTC)[reply]

Wikipedia:Clean start does appear to apply for privacy concerns. You do not have to have a problematic record to use it. Graeme Bartlett (talk) 12:03, 18 February 2021 (UTC)[reply]

Clean start can be used for any reason other than evading scrutiny. The reasons listed in the lead section are just examples of the most common situations not a prescriptive list. Thryduulf (talk) 12:20, 18 February 2021 (UTC)[reply]

I did get that part (considering, well, the fact that I'm doing it and all); this is more in reference to the WP:VANISH note "Vanishing is not a way to start over with a fresh account. When you request a courtesy vanishing, it is understood that you will not be returning. If you want to start over, please follow the directions at Clean start instead of (not in addition to) this page. If you make a request to vanish, and then start over with a new account, and are then discovered, the vanishing procedure may be reversed, and your old and new accounts may be linked." Gnomingstuff (talk) 16:37, 18 February 2021 (UTC)[reply]

A 2¢ that I do think there should be an alternative to the two current options, for cases like this or for people who are undergoing active online harassment. —Wingedserif (talk) 23:24, 18 February 2021 (UTC)[reply]

Gnomingstuff What you're getting at has been discussed extensively, and I suggest you read some of the early discussions, especially c2:WikiMindWipeDiscussion, meatball:RightToVanish, and meatball:RightToLeave which are external discussions that influenced our early drafts of VANISH and CLEANSTART. Allowing something like vanishing for editors who want to disavow or hide their previous contributions can actually be counterproductive. For example, the problem with allowing c2:WikiMindWipes is that they have a Streisand effect. An account and its edits used many years ago isn't going to be noticed or easy to find. What will make it easy to find is flooding the recent logs with reverts, renames, and redactions. I say this having been in a similar situation (see questions 8 and 9 in my RFA) and with sympathy. In an ideal world for me, the account would be nuked from orbit or at least renamed to something unique to Wikipedia. Accounts in good standing can request a rename to just about anything they want, so if you simply wish to have the old name changed you can make a request by following the steps at WP:RENAME; Vanishing just uses a particular pattern to indicate that the user has decided to leave us forever and never return. In general though, even vanishing won't resolve the issue of breadcrumbs. Your talk page signatures will still be there. The rename log will still be there. The content of the edits will still be there. Modifying them will just bring hundreds of pages to the top of recent changes and admin action logs all singling out your information. It seems counterintuitive, but abandoning an old account is one of the best ways to ensure it doesn't get found. We have over a billion edits to millions of pages; finding things is hard, even when you know what you're looking for. — Wug·a·po·des 23:04, 18 February 2021 (UTC)[reply]

Apologies, I'd checked the talk pages for both projects and didn't notice anything recent. Some clarifications: When I say tens of thousands of edits, it's with the caveat that the majority of them were cleanup, wikifying, that sort of thing. I doubt anyone here would even remember me at this point; I wasn't super memorable. I'm also not that concerned about somebody finding me from Wikipedia (which, as you mention, is highly unlikely, though probably possible if someone knows enough about me) but the other way around, finding my profile here from my username (which would be easy). Of course it'd take some effort to dig through those tens of thousands of edits, but the kinds of people who harass people online unfortunately overlap with the kind of people who'd do that (or just grep it). Gnomingstuff (talk) 01:11, 19 February 2021 (UTC)[reply]

@SlimVirgin: you might also be interested in this conversation. You brought the VANISH draft over from meta in 2007, and I remember you participating in a related discussion at AN a few months ago. No pressure though if you don't have much to say. — Wug·a·po·des 23:06, 18 February 2021 (UTC)[reply]

I’m likely going to advertise this on the functionaries list since we literally just had a discussion about this topic. My general sense is that it’s better just to abandon an account because if you’re an established user vanishing just brings more attention to it. That’s what I did when I abandoned my account I created as a teenager to create this one. TonyBallioni (talk) 23:09, 18 February 2021 (UTC)[reply]
The gist of the recent functionary discussions Tony mentioned is that vanishing is inconsistently performed and misunderstood by many users. Personally, I would like to see it deprecated entirely. One common misunderstanding is that if you have disclosed personal information with your account, vanishing will make it private again. It won't. It obscures your former username to a limited extent, but anyone familiar with Wikipedia can find it and link it to all your previous contributions in less than 10 seconds.

We're probably overdue a conversation about how we can help people keep personal information private in this day and age, but we have to remember that Wikipedia is not a social media site, it's a publisher of free content. Retiring a Wikipedia account isn't like asking Facebook to delete your profile and all the information it has about you. It's more like writing a novel, then asking the publisher to remove your name from the cover after it's in shops. They're going to say "sorry no can do, but next time you can use a pen name". That's the space Wikipedia has to work in too. We can probably do better at warning people about the consequences of disclosing personal information under a CC BY-SA License, but we can't unring the bell if they do. – Joe (talk) 08:57, 19 February 2021 (UTC)[reply]

"but we can't unring the bell if they do." Wikipedia (ENWP) and by extension the WMF can unring that bell if they want to. It chooses not to. Material released under a CC BY-SA license may be kept against the wishes of the person who posted it but there is no legal obligation that requires it to be (except those rare cases related to attribution). ENWP cites the CC license as a reason for not deleting personal information, but its a matter of policy and process. And policy and process can be changed. Only in death does duty end (talk) 09:30, 19 February 2021 (UTC)[reply]

No, it can't. Even if it were practical to expunge say, a talk page signature with someone's real name from the history of thousands of pages (it's not), it's impossible for enwp or the WMF to do anything about the hundreds of mirrors, archives, database dumps etc. that have replicated that signature across the web. Copyleft licenses by definition relinquish control over information the moment it is released. – Joe (talk) 09:54, 19 February 2021 (UTC)[reply]

Again yes it can. It chooses not to (in the first case). Its technically trivial for a qualified database engineer to replace that sort of information. The main problem the WMF has is that its technical staff couldnt code their way out of a paper bag. And in the second, that argument is essentially 'other people will keep it, so we might as well too'. Again that is process issue, not a legal requirement. That a license relinquishes control over material does not then dictacte how that material *must* be used. Or even kept. "It would be difficult and we couldnt make other people do it" is not a valid rebuttal to the suggestion that we dont do it. People are generally not going to care if crappy little mirror with maybe 1000 views a month has the potential that their personal info might be seen, when ENWP with its millions of views will guarantee it will be. Only in death does duty end (talk) 10:02, 19 February 2021 (UTC)[reply]

That's a lot of assumptions. Let's take a common story: I'm worried that someone, let's call them my nemesis, could use my Wikipedia contributions, associated with my real name, against me. I vanish this account and persuade someone at the WMF to run a query to remove my name from millions of page versions and log entries replicated across god knows how many database instances (invalidating the attribution chain and edit history of tens of thousands of pages on the way). After that, my nemesis googles "Joe Roe + Wikipedia" and finds, in the first page of results, a mirror like this. Ah, they think, I did have a Wikipedia account. They go to https://en.wikipedia.org/wiki/Special:Contributions/Joe_Roe and find a red link, but no problem, put https://web.archive.org/ in front of that address and you get a permanent archive of all my contributions. So all that work to bend the third pillar and make me disappear was undone in about thirty seconds.

Or another common story. I don't have a specific nemesis, but I regret associating my contributions with my real-life identity and do the same vanishing and transparency-breaking database procedure. Someone sees a talk page comment signed User:Renamed user XYZ and gets curious about who it is. In the first few pages of XYZ's contributions, they see that they're an archaeologist interested in the Near East and statistical computing, who lives in Denmark. Those four diffs alone narrow it down to maybe 2-3 people in the world and again a quick google search will establish it's probably me. Is it "trivial" for a database engineer to identify those sorts of incidentally-disclosed clues to someone's identity? Bearing in mind that two of those diffs are to articles, not talk pages? And one isn't mine, it's another user (appropriately) alluding to personal information I've disclosed elsewhere?

I'm not denying that we can do things (vanishing, oversight) to make it slightly harder to find personal information on Wikipedia, but presenting these kludges as solutions to privacy concerns is dishonest and potentially dangerously misleading. What we need to do is make it crystal clear to people, from the beginning, that everything they do on Wikipedia is in the public sphere. And that in an emergency, they need to take steps to protect themselves beyond what we or the WMF is capable of. – Joe (talk) 11:01, 19 February 2021 (UTC)[reply]

What is dishonest is your continued insistance that it cant be done while relying on 'well it would be kept elsewhere so we might as well not'. It exaggerates the edge cases. Firstly absolutely no talk or user page contribution on ENWP is necessary to be kept at all. And while yes archive sites *may* archive some material, this is not universal. You also appear to have a basic misunderstanding of Attribution. Attribution is required on ENWP for material kept on ENWP. It is not necessary, or in fact required in any form, for ENWP to keep an attribution record chains simply for the purpose of third parties who may use material copied from ENWP, when that material has been removed/deleted from ENWP. The onus on the third party to attribute correctly. If it breaks for them, that is not our problem. The point of removing personal information is not to make it completely impossible to identify someone, it is to make it significantly more difficult to do so and to eliminate the casual dissemination of personal info. The attitude of 'well since a significantly dedicated person with lots of time on their hands could jump through 15 hoops to get it means its pointless' is both lazy thinking, irresponsible, and violates any number of data protection principles in various parts of the world. What we need to do is change the attitude that other peoples personal information is fair game forever just because it was decided in the past that was what ENWP should do. Only in death does duty end (talk) 12:30, 19 February 2021 (UTC)[reply]

I think the focus on attribution misses the point. As brought up in c2:WikiMindWipeDiscussion, while anyone is free to remove content, anyone is free to add content as well. If someone goes around removing their signatures, they will quickly get reverted by the community. We have two options then: convince everyone to not do that or convince admins to start blocking people who revert modifications to talk page archives. Both of those are uphill battles, to say the least, and if there is any controversy we risk bringing the personal info squarely into the spotlight at AN, completely negating the privacy benefit of a mindwipe. Going beyond that, we cannot guarantee privacy once the information has been posted, and pretending to do so is irresponsible if not unethical. Even if we were to allow revision deletion of edits by vanished users, we risk increasing the potential visibility and harm. There are people and robots which watch our deletions. An easy way to get their attention is to have a nicely organized, compact section of log entries that correlates page redactions with a rename. We could have those log entries redacted too, but redacted log entries are even more conspicuous than revision deletions. Even if we did agree to allow this, such operations take time and are error-prone. If anything is missed, the whole process could have been pointless, and even if we are perfect, the time it takes from start to finish is more than enough time for mirrors to preserve the information beyond our control. Now, at this point, we've succeeded in eradicating the editor's personal info from our servers, sure, but we've also painted a huge target on their back as every deletion log watcher begins to wonder why hundreds of log entries were redacted without explanation. They then go to mirrors and find all the info they need and then go post about it on Wikipediocracy for some bad faith actor to find later. Have we succeeded in helping the editor? Did all our work protect anyone from harm? Is this a situation we should encourage anyone worried about their privacy to put themselves into? No on all counts. Let's go one step further and try to fix the fundamental issue: logs. That's the domain of the MediaWiki project, not EnWiki, and getting devs to change the software to eliminate any kind of logging will likely be even harder. The software is built around transparency, and we would need substantial buy-in from that community to have them change their primary principles. Are these edge cases? Maybe, but vanishing is an edge case in the first place. Security systems which ignore edge cases are not secure, because exploiting edge cases is exactly how you defeat security systems. — Wug·a·po·des 22:48, 19 February 2021 (UTC)[reply]

This is kind of where my "it's now been 15 years" remark comes into play; that's a very long time both in terms of developments regarding Internet privacy and in terms of life stages. Editing in high school/college is one; another very plausible situation is that someone might take a job requiring extensive background/security checks, or a job with risk of being fired for off-work activities like grade-school teaching, that was not on their radar over a decade ago. "Everything you do is in the public sphere" is all well and good in hindsight. The concern about edits being conspicuous is valid and something I don't have a good answer to. The only thing that immediately comes to mind is a system that automatically anonymizes usernames after X period of inactivity (which, of course, could be undone should the editor decide to come back), but that doesn't solve the log issue above so much as drown it in noise, and I imagine people might have other objections to it. But at the very least, the current policy ("leave it or leave," essentially) has room to be loosened. Gnomingstuff (talk) 02:41, 20 February 2021 (UTC)[reply]

Re Gnomingstuff's problem, I recommend abandoning the account without any drama (no further edits, no "retired" tag or other farewell message). After a delay, start a new account with no mention of the old account. Or, if your particular situation would be helped by vanishing (renaming) the current account, do that per VANISH. However, the advice above is correct, namely that vanishing is a very flimsy mechanism that is easily undone. Ignore the instructions about connecting your new account to the old. Instead, send an email to Arbcom (see User:Arbitration Committee) briefly explaining the situation and telling them you have created the new account (which you would reveal) without linking it to the old account in order to avoid harassment. Johnuniq (talk) 00:59, 20 February 2021 (UTC)[reply]
@Gnomingstuff: Just to be clear, "courtesy vanishing" is just a rename to some random nonsense, userspace deletion, and possibly the deletion of irrelevant meta discussions about the user themselves, which most people don't have to begin with anyway. It doesn't do anything beyond that. Your signatures will still be there and they will still link to your renamed account. You can still rename your account to some random user name, delete your userspace, blank your talk page, and then do a clean start. You'd be achieving the exact same thing, the only difference is that your rename would have to be "carefully chosen" rather than random characters caused by slamming your hands on the keyboard. ~Swarm~ ^{sting} 01:43, 20 February 2021 (UTC)[reply]
We should have a hygiene-vanish process where an old account gets renamed and a bot searches-and-replaces signatures from the old name to the new name. Some trusted authority who've signed NDAs like arbcom or maybe OS or CU (or T&S) can administer and keep a record for any copyright or other legal-related needs. This won't be 100% deletion because the old name will still be in old revisions and that's unavoidable, but at least it'll make it so searching for the old name won't bring up any results. And that option should be open to any editor in good standing. Levivich ^harass/_hound 08:08, 20 February 2021 (UTC)[reply]
- Umm, no. First, wide-scale editing is disruptive, particularly in archives where "this hasn't been touched" is a good indication of the archive's integrity. More importantly, such changes would draw massive attention with the certain result that dozens of people would notice that User:X was renamed User:Y, and a non-trivial number of those would be sufficiently curious to investigate the background. Finally, if someone ever wonders why they can't find the editor they planned to harass, they can simply look in the bot's contributions for a permalog of all vanished users. Wikipediocracy and other troll sites would quickly set up a system to translate the bot's contributions into a handy table: X was renamed Y on such-and-such date, with a comments section for any gossip they can find. Johnuniq (talk) 09:35, 20 February 2021 (UTC)[reply]
  - The bot can delete its own edits. Heck it doesn't even have to be a bot, this can be done "server-side" and not be logged at all. We can think outside the box. We can prioritize privacy over watchlist disruption. Levivich ^harass/_hound 15:03, 20 February 2021 (UTC)[reply]

Evaluating WP:NEXIST

There is a discussion over at Wikipedia talk:Notability#WP:NEXIST that might be of interest. Thank you CapnZapp (talk) 15:34, 20 February 2021 (UTC)[reply]

@@ Line 412: / Line 412: @@
 **Umm, no. First, wide-scale editing is disruptive, particularly in archives where "this hasn't been touched" is a good indication of the archive's integrity. More importantly, such changes would draw massive attention with the certain result that dozens of people would notice that User:X was renamed User:Y, and a non-trivial number of those would be sufficiently curious to investigate the background. Finally, if someone ever wonders why they can't find the editor they planned to harass, they can simply look in the bot's contributions for a permalog of all vanished users. Wikipediocracy and other troll sites would quickly set up a system to translate the bot's contributions into a handy table: X was renamed Y on such-and-such date, with a comments section for any gossip they can find. [[User:Johnuniq|Johnuniq]] ([[User talk:Johnuniq|talk]]) 09:35, 20 February 2021 (UTC)
 ***The bot can delete its own edits. Heck it doesn't even have to be a bot, this can be done "server-side" and not be logged at all. We can think outside the box. We can prioritize privacy over watchlist disruption. [[User:Levivich|Levivich]]&nbsp;<sup>[[User talk:Levivich|harass]]</sup>/<sub>[[Special:Contribs/Levivich|hound]]</sub> 15:03, 20 February 2021 (UTC)
+== Evaluating [[WP:NEXIST]] ==
+There is a discussion over at [[Wikipedia talk:Notability#WP:NEXIST]] that might be of interest. Thank you [[User:CapnZapp|CapnZapp]] ([[User talk:CapnZapp|talk]]) 15:34, 20 February 2021 (UTC)