Wikipedia:Village pump (idea lab)/Archive 43

• Village pump
  • Policy
  • Technical
  • Proposals (persistent)
  • Idea lab
  • WMF
  • Miscellaneous

Village pump (idea lab) archive  

This page contains discussions that have been archived from Village pump (idea lab). Please do not edit the contents of this page. If you wish to revive any of these discussions, either start a new thread or use the talk page associated with that topic.

---

< Older discussions · Archives: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62

Topic related help resources

Yesterday I was wondering if anything special happened when you searched for suicide in Wikipedia like it does in many other major websites. I did search and nothing happened beside being shown the article for it. I went to its talkpage to say something about this idea and there I saw that this subject was actually a recurring discussion on that page so much as to warrant its own subpage and a RfC, both mentioned on the talkpage's header. I went down that rabbit hole and read most of what was typed in both links but that made me think:

Why not provide different help resources in regard to different topics? Suicide, albeit an extreme case, doesn't have to be a special case in this regard and that would help alleviate some of the problems that arise from treating that subject differently from others.

What would this mean in practice?

Much like we assume that someone searching for "suicide" has a arguably rather high potential to be having suicidal thoughts, we can assume that someone searching for "depression" has a arguably rather high potential to be having problems with depression. This assumption can be extended for many mental health problems and even beyond that realm (physical problems, matters such as the global pandemic, etc.) I think it is rather safe to assume that trying to provide some help for these people is not a bad choice. What would that help exactly be? I don't know as I'm not a mental health specialist or a all-knowing being but that can be solved on a per-case basis. This discussion in only for the general aspect of paving the way for such a behavior in some specific topics.

We already have the WP:Reference desk which does provide a similar role but it is buried much too deep to be found by users that have such problems. I am aware that there have been debates on the past about its existence. I am also aware about the fact that much to what I'm proposing might interfere with WP:NOT. Things are fine as they are bureaucratically wise and no change is needed. The problem is the influence Wikipedia has on real life problems (especially helped by Google) on factors unrelated to "what an encyclopedia is or is not". At the moment of typing this, if you search in Google for "suicide", the relevant Wikipedia article comes out as the third result and on its literal top you'll find the words Kill Yourself highlighted in blue. (Things are purposely out of context here to try and give a suicidal person's POV. I've already started a discussion about it.)

We already provide the news section even though it is not related to an encyclopedia per se and a lot of accessibility action is being taken to make Wikipedia a welcoming place for people with different visual and/or other disabilities. I believe this proposal falls on the same category.

For people that haven't been annoyed yet in reading all that, you can also read this other now archived short discussion of mine regarding mental health and Wikipedia elements. - Klein Muçi (talk) 11:06, 9 September 2022 (UTC)

For reference, please also refer to this thread on a similar topic at the proposals Village Pump. isaacl (talk) 16:14, 9 September 2022 (UTC)

Isaacl, wow... I'm really surprised about the synchronicity at hand. I usually don't enjoy the rather harsh tones in the VP so I rarely check its content. I wonder if something we might have randomly seen on a global scale these days might have influenced us subconsciously to think about the suicide article/response at the same time.
Thanks for letting me know! — Klein Muçi (talk) 16:33, 9 September 2022 (UTC)

How to give advice in a neutral way across all topics, in a global environment with different resources available to each community, without ending up in endless disputes is challenging. Even if the community felt this was within its scope, I'm not sure how to do it in practice. For medical topics, which all your examples are related to, there's the additional worry of giving advice that isn't tailored to the specific circumstances and context, where a local health professional would be better positioned to make suitable recommendations. isaacl (talk) 16:22, 9 September 2022 (UTC)

Isaacl, I understand what you mean but I still think there are ways we can act on. We usually make the question from the Wikipedia side and we "get stuck" in this kind of discussion. I believe things would be a bit different if the question was made somewhat from the other direction, for example: Should we allow better accessibility for people with mental health issues? Most likely the answer to that question would be "yes". Suicide and depression-related hatnotes would be one factor but consider warnings for strong flashing lights for epilepsy. There can be a lot of change that can be offered in terms of very simple and small details that would improve the overall experience for people suffering from these conditions and hopefully make the whole community more inclusive. (Some kinds of warnings can even be temporary, like those related to COVID, crude example.) We are already thinking of ways to make Wikipedia more accessible to people suffering from blindness, senior citizens, children in schools... This would be just an extension of that initiative.
As for the worry of giving advice that isn't tailored to the specific circumstances and context, we already have m:Wiki Project Med, no? The mentioned cases wouldn't include instructions such as "drink this medicine" or "do this kind of intervention". We're talking just about refactoring the pages with small changes in layout without altering the core content in such a way as to make them more friendly for some groups of people suffering from certain disabilities. A similar practice is already happening. If you see the 4th question in Talk:Muhammad that's more or less what I'm proposing seen in action, albeit maybe for different reasons. The third question there further provides "accessibility" in the same direction if so needed. — Klein Muçi (talk) 22:27, 9 September 2022 (UTC)

@GreenC, maybe you are interested in taking a look around here as well if you haven't yet. — Klein Muçi (talk) 23:05, 9 September 2022 (UTC)

You spoke about the reference desk so I thought you were suggesting to give advice. If you're suggesting changes to writing style and presentation, I agree there's a path for the community to reach a consensus on related guidance. isaacl (talk) 00:29, 10 September 2022 (UTC)

Overthrow the Administrators & form a Communist State

This is just an idea, but it would be great for Wikipedia to become a Communist State.

Here’s why:

-Article edits evenly distributed -Secret police eliminate vandalism -No classes aside from titular “leaders” -Featured articles all about communism -National pride through flags & anthems -All articles can be edited — Preceding unsigned comment added by Hanojak (talk • contribs) 14:19, 25 August 2022 (UTC)

I think you'll find that North Korea has a bigger proportion of its population involved in administration than any liberal state. Phil Bridger (talk) 14:34, 25 August 2022 (UTC)

Unlike some sites, administrators are just that: they work tirelessly to do our admin, including tasks that we can't trust to just anyone such as deleting articles. Wikipedia is probably as close as you'll get to the communist ideal of from each according to his ability, to each according to his needs. Certes (talk) 15:15, 25 August 2022 (UTC)

@Certes wait, you guys work tirelessly? Maybe a stimulant of choice grant request to the foundation? Nosebagbear (talk) 15:09, 14 September 2022 (UTC)

This idea has been scrapped due to already existing. Hanojak (talk) 15:51, 25 August 2022 (UTC)

• It is already policy that Wikipedia is neither an anarchy nor a democracy and there is no free speech. This doesn't leave many possibilities but the trouble with the communism idea is that it's also policy that it's not a bureaucracy either. Andrew🐉(talk) 20:36, 25 August 2022 (UTC)
• I like this idea. Levivich 19:38, 5 September 2022 (UTC)

Extending the amount of time an RFA is opened

Hi, I would like to know if there have been any proposals calling for extending the time RFAs are opened. I checked the archives to see if there was and I couldn't find any. I would be interested if anyone would support extending the time RFAs are opened from 1 week to 2 weeks. I think that the longer amount of time that the RFA is open will give less stress to the candidate when answering questions. We've had many discussions on Wikipedia that have been open for more than 1 week, so why should RFA be so short? Interstellarity (talk) 21:30, 3 September 2022 (UTC)

Many editors would just be maximally stressed for 14 days instead of 7. —Kusma (talk) 21:38, 3 September 2022 (UTC)

My understanding is that the rationale for seven days was that some editors only log on once or twice a week (say on weekends) so we maximise the chance of editors being able notice the discussion and participate. Allowing for longer would allow for more participation, but there would be a diminishing rate of return, and as Kusma pointed out, this must be balanced against the stress on the candidate. Hawkeye7 (discuss) 00:06, 4 September 2022 (UTC)

• AFAIK there haven't been any such proposals, and IMO such a suggestion would fall on deaf ears. Admins (and some former admins) speak for themselves here on the thrill of going through the process. Kudpung กุดผึ้ง (talk) 02:06, 4 September 2022 (UTC)

IMHO, the problem with RFAs nowadays is too many questions, especially complex hypotheticals about certain edge cases. Years ago it was common to pass RFA/RFB with only the 3 standard questions and maybe 4-5 additional questions. Now it seems even effectively unanimous passes with almost/over WP:200 supports still answer almost 20 questions. Adminship is a toolkit for trusted users, and was always intended not to be a big deal, they aren't supreme court justices. Andre🚐 02:18, 4 September 2022 (UTC)

• I don't think this is a good idea. Full disclosure; my own experience at RfA has prejudiced my opinion rather strongly. -Ad Orientem (talk) 02:25, 4 September 2022 (UTC)
• (Having gone through the process recently...) I agree with the statements above in that this would probaby increase stress, and additionally we may get fewer willing candidates as they may feel like they have to wait until they have a two-week window where they are able to be constantly available and responsive to questions, rather than one week. DanCherek (talk) 05:12, 4 September 2022 (UTC)
• (I also just went through this process) In my opinion, longer RfAs will lead to more optional questions, thus increasing the candidate's stress. I would not want to go through a two-week RfA. Z1720 (talk) 18:38, 4 September 2022 (UTC)
• Disagree. In most cases, consensus is achieved rather quickly over 1-2 days even, sometimes newer vital information comes shortly after. But generally, one week is abundant. In cases, where it is borderline, adding 7 days won't help much. From a stress point of view, the longer, the more stressful. I recently self-nominated myself unsuccessfully for an RfA and do think there are other ways to reduce the stress, namely having more people run simultaneously. ~ 🦝 Shushugah (he/him • talk) 18:42, 4 September 2022 (UTC)

Shorten the time then?

Given the consensus here, should we shorten the RfA time then? 3 days should be good for me. CactiStaccingCrane (talk) 11:13, 14 September 2022 (UTC)

As stated by Hawkeye7, having a one-week period allows editors who only participate once a week to contribute to the RfA. isaacl (talk) 20:33, 14 September 2022 (UTC)

Pre-RfC: New antivandalism role: Responder

Please see Wikipedia:Requests for comment/Responder role. The RfC will start soon, but I wanted to ask VPI if you all had any feedback on the RfC before it goes up. Is there anything confusing? Should anything be clarified? Enterprisey (talk!) 04:18, 11 September 2022 (UTC)

@Enterprisey to me, one likely objection that has not yet been answered is as follows: this method will presumably be eventually used incorrectly. That's not the issue.

The issue is that there will then be a block in that user's log, with negative reputational impact, and getting blocks overturned when some level of misconduct has occurred is difficult.

In the path of offering solutions not problems, I would suggest that where one of these blocks was issued and the admin does not directly then block the IP/user, the block will be viewed as being incorrect and must receive a blank "1 second block" log note to specify that the block was incorrectly granted. I'm not advising that such an overturning is an automatic revocation of the responder userright (though the admin obviously should review), but making sure that the block doesn't stick when warnings would have been issued. Nosebagbear (talk) 14:52, 14 September 2022 (UTC)

Perhaps a tag on the given blocks? Say, if a block was handed out by the responder, the (Responderblock) tag could be made in the edit filter, and the block could be tagged with (Responderblock, Overturned) if an admin finds it as inappropriate. I have little technical knowledge, so I'm not sure if this is feasible. CollectiveSolidarity (talk) 19:29, 14 September 2022 (UTC)

@CollectiveSolidarity: From a technical standpoint, implementing this would not require much effort, as bots can easily make blocks with added tags, and also add tags to a block log entry afterwards. 0xDeadbeef 02:12, 15 September 2022 (UTC)

There may be some things wrong with this proposal (if I was prepared to spend more time on Wikipedia I would lobby for a more "easy-come, easy-go" form of granting and removing admin rights in general) but I don't think that reputational impact should be an issue. If anyone is judging people by entries in their block logs they should simply be told to stop doing so. Phil Bridger (talk) 20:06, 14 September 2022 (UTC)

New essay: Wikipedia will exist in the far future

Apologies if this is not the right place to notify, but I have created a new essay: Wikipedia:Wikipedia will exist in the far future. It started out as just a fun idea but I thought there was some value to writing it up, although you might disagree! Barnards.tar.gz (talk) 12:10, 9 September 2022 (UTC)

I love the optimistic view. CactiStaccingCrane (talk) 12:46, 9 September 2022 (UTC)

Agreed! Sm8900 (talk) 14:00, 9 September 2022 (UTC)

1 Jo-Jo Eumerus (talk) 15:24, 9 September 2022 (UTC)

Most interesting part:
The 21st century may be as relevant to the 301st century as the year 27,978 BC is to us.
Gave me a new point of view about history in general and more. — Klein Muçi (talk) 15:30, 9 September 2022 (UTC)

ok, that's one way to look at it. but I have a slightly different view on such things. here is my version of that; Winston Churchill, Franklin Roosevelt, Hitler, and Mussolini, as the major leaders of the World War II era, may be as relevant in 2,000 years as Alexander the Great, Julius Caesar, and Octavius are to us today. Sm8900 (talk) 15:49, 9 September 2022 (UTC)

Sm8900, they certainly may. The interesting part was in regard to how history gets compressed with time. Consider how we talk about the past elections or all the elections that have happened in this decade and how we treat whole periods spawning centuries in the Ancient Rome that get summarized in 2-3 sentences. This most likely holds true even for Wikipedia and inclusionist/deletionist parties will have their fun forever and ever over this aspect. The good thing though is that the information is still there if someone wants to decompress it. Be that for Ancient Rome or for Wikipedia archives. — Klein Muçi (talk) 16:14, 9 September 2022 (UTC)

But those people are much closer to the present time than they are to 30,000 years ago. Phil Bridger (talk) 17:33, 9 September 2022 (UTC)

true, but look at the different nature of the replies, on an intrinsic level. since our history currently does encompass several time periods of 2,000 years in length, it is possible to extrapolate, on that basis, what might be the relevant significance of events and figures from our own time, once another 2,000 years have passed.

however, for a period of 30,000 years, we have no comparable time period within recorded history to compare that to; therefore, any analogies, assumptions, extrapolations, might be better off if applied to periods for which we already have some example.

of course, i think that in 30,000 years, the Internet will be built into every streetlight and bus stop, and wikipedia will be sentient being that is friendly and helpful, but that's just me. lol Sm8900 (talk) 13:56, 14 September 2022 (UTC)

I really enjoyed this. Two things came to mind while reading it. First, there have already been efforts to preserve Wikipedia. There are many physical copies floating around, including one on the Moon. Second, regarding language change, it presents a good reason why we should use standardized language, but it's also a good reason of why we should record and describe non-standardized language. Thebiguglyalien (talk) 18:31, 9 September 2022 (UTC)

Any chance you'd be alright having this republished in the Essay section of The Signpost? Adam Cuerden ^(talk)_{Has about 8.1% of all FPs} 18:22, 11 September 2022 (UTC)

Thank-you so much for sharing this. It reminded so me of the optimism I felt reading Isaac Asimov 's /Heinlein's/Douglas Adams'/Encyclopedia Galactica (and maybe a little of Multivac in the The Last Question ). Your essay led me to a List of fictional computers and Time formatting and storage bugs, and List of IP version numbers(of which I understand little. Wakelamp d[@-@]b (talk) 04:47, 17 September 2022 (UTC)

Biology articles and human focus

There's currently no standard for how to divide humans, other animals, and other living organisms in biology articles (especially in anatomy and medicine articles). A few examples:

• Articles such as Heart are nominally about the entire topic, but the vast majority of the content is about how they occur in humans and an "Other animals" or "Other organisms" section covers all other examples.
• Articles such as Brain and Human brain have both a general article about the topic and a specific article about the human version.
• Articles such as Digestion and Human digestive system similarly split the article, but they follow WP:SUMMARIZE and have a section in the parent article summarizing the child article.
• Articles such as Sleep and Sleep in animals treat the human aspect as the default with a separate article for all other occurrences.
• Articles such as Anatomy#Human anatomy and Human body#Anatomy intersect and talk about the same subject relative to two different ideas.
• Articles such as Tissue (biology) discuss all forms without making any distinction between different types of animals.
• Articles such as Health lack information on the subject in the context of any living thing— animal, plant, or otherwise— except for humans.

Each of these formats seems to have different drawbacks. How can we come up with a method to determine which formats to use and when they should be applied? Should there be a codified standard that's used? Thebiguglyalien (talk) 06:00, 8 September 2022 (UTC)

Thebiguglyalien, a couple of months ago I was working in translating the articles about the human body and the heart and noticed more or less the same thing as you do. It left me in a confused state: Is this anthropocentrism? If it is, is it coming just because humans tend to write articles about themselves more than animals do about themselves and humans will join them later and help or because it is a current global practice in anatomy that it should deal only with humans unless otherwise specified? If it is not, maybe tissues and organs don't change THAT much between all species as to warrant different articles beside some exceptions? Between those 3 options, judging by the rather-chaotic (at least for my eyes) appearances some of the anatomical articles tend to have, I believe it is most likely that we've fallen in anthropocentrism just because there has, naturally, been more interest to write for ourselves than for animals. If that indeed is the case, I'm glad for this discussion and I believe we should act on addressing this problem. — Klein Muçi (talk) 09:09, 9 September 2022 (UTC)

Inconstistency is a pretty inevitable feature of Wikipedia, given its model. This is probably best discussed at Wikiproject level, maybe Wikiproject Biology. I would say here that Wikipedia's job is to reflect what independent reliable secondary sources say, so if much of the research into subjects is about humans then we reflect that. Phil Bridger (talk) 09:20, 9 September 2022 (UTC)

The article Baylisascaris shroederi is prime example of inappropriate focus on humans. It's a parasite of giant pandas and there are NO known cases of humans being infected by it. But it's written primarily as a medical article with sources about a related species that does infect humans. Plantdrew (talk) 20:09, 9 September 2022 (UTC)

I would not say that it's primarily written about humans, although it contains some information about us. There's more content in the article about giant pandas than humans. I have no idea whether the article conforms to WP:DUE, which expects that articles will be written in a way that reflects the amount of content that exists in independent reliable secondary sources. As I said above, this is better discussed at a Wikiproject, such as Wikiproject Biology, where you should find subject-matter experts. Phil Bridger (talk) 20:28, 9 September 2022 (UTC)

As a possible solution the article could be divided into two sections, one of which would be about human and another would be about pandas. But if the sources describe that parasite separately in human and pandas so we should also do so. The current article is mixed up, especially with Prevention section about humans and epidemiology about pandas. The article can be split into two different articles, one of which would be related to medicine and another would be related to biology or veterinary. D6194c-1cc (talk) 15:30, 10 September 2022 (UTC)

And also more appropriate name for the disease article would be Baylisascaris schroederi infection. And the article about parasite in context of biology could be called just Baylisascaris schroederi. D6194c-1cc (talk) 15:45, 11 September 2022 (UTC)

Oh dear. I hadn't really read the B. shroederi article when I brought it up (it's just something I'd noticed awhile back that had a bizarre focus on humans for something that has never been reported to infect humans). The B. shroederi article is mostly a clumsy copy of Baylisascaris procyonis, with some words replaced (e.g. racoon->giant panda). And the B. procyonis article has more content on humans than I think is appropriate (but at least it does infect humans), which could be split to Baylisascariasis. Going back to B. shroederi, I think the article may represent a hoax or breaching experiment (I have difficulty imagining that incompetence alone could account for the medical misinformation in that article, and if it was a breaching experiment it succeeded; Doc James edited it without noticing any red flags (and neither did I when I first edited it)). Plantdrew (talk) 01:28, 18 September 2022 (UTC)

I think that part of the inconsistency is due to our system for naming articles. For the first article, someone starts an article about ____, and says "Okay, this page is about _____, so I'll name it that". The contents are a few human-centric sentences. Then the next person comes along, and has to choose between:

• expanding the existing article (e.g., adding non-human information);
• renaming the existing article to "Human ____", so that another human-centric page can be created about ____ in general; or
• starting another page about ____ in general, and naming it something like "____ in animals".

(Some of this is outlined in the Wikipedia:Article titles and scopes.) Different editors will choose different answers, based on both their personal preferences and the specific facts and circumstances (e.g., whether that infection or bit of anatomy is more of a human disease or more of a livestock problem). Sometimes the differences in the names just hint at which article was written first.

Also, as a side note, it is not unreasonable for an encyclopedia that is written by humans, for the purpose of being read by humans, to focus on humans. There might be something like 500,000,000,000,000,000,000 individual bacteria for every human on Earth, but that doesn't make them the focus of every article. WhatamIdoing (talk) 19:11, 12 September 2022 (UTC)

I have given some thought to the issue myself, and brought the topic up at WikiProject Palaeontology, and I am of the opinion that having two separate articles on each anatomical structure is the most appropriate, with one article on the structure in human anatomy and another article giving an overview of the structure across the animal kingdom. The needs of an article giving a general overview of the structure across all animals and the needs of a more medically-focused article on the structure in humans are just too distinct to reconcile them into a single well-formatted article. I am undecided on whether it would be more appropriate to treat human anatomy as the primary topic (e.g. "Femur" and "Femur (vertebrate anatomy)") or not (e.g. "Femur in human anatomy" and "Femur"). My instinct, as a paleontologist, is to favor the general topic as the primary topic over human anatomy, but I suspect that many readers may expect otherwise. Ornithopsis (talk) 23:54, 12 September 2022 (UTC)

I largely agree with Phil Bridger's opinion - I tend to find that most of the academic articles I find on a topic are about humans rather than animals, and this is reflected in the content that I can possibly write in articles. This is mainly in relation to anatomy, but probably applies to other fields. Bibeyjj (talk) 09:33, 13 September 2022 (UTC)

It depends on what sources you read. As a paleontology-focused editor, I am familiar with many sources on various bones in non-human vertebrates, for instance. In any case, I would argue that the fact that there are so many sources specifically on anatomical structures in humans further emphasizes my point that human anatomical structures are a specific case of general anatomical structures that are notable enough to merit separate articles. I just don't think there's a good way to create a single article that adequately covers the needs of both. Ornithopsis (talk) 00:31, 14 September 2022 (UTC)

I agree with Ornithopsis here that the two concepts are not reconcilable into a single coherent article, and need to be separated. Hemiauchenia (talk) 00:39, 14 September 2022 (UTC)

I have to say I agree with the observations above. A lot more content is uploaded about humans and present here than other animals. In my experience to date there's a lot of discussion about why, but discussions don't usually end up resulting in more non human content that actually gets added. Most articles seem to follow the general WP principles where if there's lots of content it gets split; otherwise articles get merged. It's a pity because some big areas (including culinary, history and symbology) are often completely missed. Tom (LT) (talk) 11:09, 14 September 2022 (UTC)

This is my concern. There doesn't seem to be much will to address this issue, and as a result we get gaps in coverage if not totally misleading articles. For example, the article on Heart says in the lead that "The heart beats at a resting rate close to 72 beats per minute", claiming this as a fact about hearts in general. I'm hoping that raising this issue here can bring about some sort of improvement, but I'm not optimistic. Thebiguglyalien (talk) 21:55, 14 September 2022 (UTC)

I think that one of the reasons these discussions don't lead anywhere is that nobody knows how to write a good article that effectively balances both medical/human-focused and zoological/paleontological perspectives. This is exactly why I think we need to just go ahead and make the split. Otherwise, the lack of clarity on how to structure these articles is going to keep holding them back. I suggest we treat the zoological article as the primary topic and title the human article either "[X] in human anatomy" or "Human [x]". Ornithopsis (talk) 00:11, 16 September 2022 (UTC)

That's the conclusion I came to as well, but it seems to be controversial. The discussion on Talk:Injury proposing such a split is what prompted me to make this post here, as I realized it was a site-wide problem rather than an issue with one specific article. Biology, anatomy, and medicine articles are only going to get more muddled until there's an active push to fix them. Thebiguglyalien (talk) 00:48, 16 September 2022 (UTC)

Good idea to try - I suggest pick one or two articles to see how it goes. If you're adding content and the changes are an improvement, mainspace is a good idea. Please do add content rather than reorder things, as otherwise you're not really helping fix the underlying problem of editors not uploading non human content. Once you have a few articles of a high standard in the way you want it'll provide a template for future editors to model their own articles on. Preferably don't start with FA or GA class articles as they're at least somewhat polished. You can also trial it out in draft space or sandbox. You can get feedback from the related Wikiprojects - WP:ANAN and WP:ANAT if it's anatomy, or others depending on what you're editing. Tom (LT) (talk) 00:08, 18 September 2022 (UTC)

Reduce Non Notable/Junk Article creation

Problems are new editors rage quit after we delete their articles, we need to encourage more people to become editors (because it's nice), reduce excessive workload on NPP, create start class not stubs, reduce AfD conflicts, and do something about the diversity complaints that WMF is trying to address in the Endowment . Also World peace.

That the Create Article Process (which includes the Wizard) has problems is a truth universally acknowledged; the navigation is confusing, it teaches rather than trains, and it provides too much information (links to 300 sections and more than 10k words) repetitively and at the wrong time. It is actually not really a Wizard as nothing is added to the page - more like a disclaimer, or a health warning.

Suggestion is

1. We give the new editors free choice - to only choose the New Article Wizard
2. Simplify navigation They are currently on various notices, on the search screen, and in the Wizard
3. Provide only enough information to avoid 80% of the AfD reasons, so editors do that. If they screw up a citation format, then it's minor.
4. Warn Editors that 95% of new articles are speedily, or AfD deleted, and that a new article needs 2 to 3 hours of their time, "normally" 3 quality references, and a few paragraphs of referenced non plaigarisied texts.
5. Ask about the most likely problem for deletion first, on one screen. When they go to publish ask them the same questions again, and have it one document that is linked in the article they create, or maybe as a checklist in the article, Have one link from that document to further information.
6. Have them enter their 3 references (for new editors better more than less) in the Article Wizard (yes @Kudpung you knew it was coming) and have the NPP notability reference check them. BUT If it is a bio, then ask them the diversity questions about the Bio, so we can funnel/invite them to Women in Red. Otherwise advise them that help is needed on FA, Core etc.
7. General Area.The submit to AfC from draft asks them to enter a general area, and project tags. I think AfD does this automatically. Standardize the process and make it part of the Article wizard, and use it to create an infobox. Depending on the Infobox, then bring up the relevant warnings - why warn someone creating a Pokemon defamation?
8. Create a desire path called Vote for an article. Say we will contact them , if they stay logged in, when the article is created, or if we decide that it doesn't meet notability. Suggest they help somewhere
9. Change "More" on the Taskbar to Editor, and make it greyed out unless you are logged in. Add Redirect under this, and insert common templates. Add something snazzy such as a deletable user path network. Suggest they work on FA.

Wakelamp d[@-@]b (talk) 08:11, 19 September 2022 (UTC)

Ability for users to change their edit summaries

It would be very helpful if editors could have the ability to edit their own edit summaries they made. I personally know of many times I've made errors when describing an edit summary I made, or wish I could go back and make changes or grammatical corrections to particular edit summaries. Helper201 (talk) 22:25, 19 September 2022 (UTC)

You can make a WP:dummy edit to accomplish that. Sungodtemple (talk) 13:15, 21 September 2022 (UTC)

Adding a Clock to the front page and or users' homepages

I don't know if this is implemented somewhere else on Wikipedia but I was thinking that having a clock on the main page and or each users' homepage might be a usual tool for editors. It might be more useful on users' homepages as it would tell time but also show new users the time zone Wikipedia uses. Thank you for time, have a good day! DiscoA340 (talk) 01:08, 21 September 2022 (UTC)

There is a gadget that puts a clock in the upper-right of all pages. Any logged in user can enable it for themselves via Special:Preferences#mw-prefsection-gadgets, in Appearance... "Add a clock". Does that do what you are envisioning (so your idea is just to make it default-active for everyone)? For the record, it's broken under the new Vector2022 skin. DMacks (talk) 01:14, 21 September 2022 (UTC)

@DMacks I was thinking of it only being added to those two pages and it would look like an analog clock (like the clock app on Apple devices) but that works too, Thanks for showing me that. DiscoA340 (talk) 01:29, 21 September 2022 (UTC)

You may also be interested in some of the templates here: Category:Time, date, and time zone user templates. For the most part they are static (they only update when the page is purged/reloaded). We are very unlikely to put "dynamic content" like a moving clock on the main page or default userpages, as could spoil caching (here or downstream). — xaosflux ^Talk 13:54, 21 September 2022 (UTC)

There probably isn't an operating system anywhere that doesn't include a clock. I've got two on my desktop (local and UTC). What functionality would having a clock on the main page serve that isn't already handled by the OS or by some browser plugin? -- RoySmith (talk) 14:13, 21 September 2022 (UTC)

Lighter page colours in dark theme

Moved from Wikipedia:Village pump (proposals)

I've been using the new-ish dark theme for a while now, and it has a slight problem where the page colour and background colour are the same pitch black. Having pitch black sidebar and page contents is not good for contrast, and in my opinion makes it harder to distinguish foreground elements from background ones. I would suggest taking inspiration from Apple's UI to improve the look of the dark theme. -A Fluffy Kitteh | FluffyKittehz User Profile Page 13:40, 21 September 2022 (UTC)

@FluffyKittehz: I moved this to Idea Lab for now, this would need to be fleshed out a bit before a community proposal would be needed (and it might not even be needed). There are several "dark" modes/themes/skins - can you be specific about which one you are referring to? (e.g. is this the IOS App Dark Mode, Android Darkmode, the enwiki Gadget for desktop, etc) ? — xaosflux ^Talk 13:52, 21 September 2022 (UTC)

We don't have a dark "theme". If you're talking about the dark mode gadget, calling it an inverted filter would be more accurate. As discussed here, if someone comes up with neat CSS to make it less dark, we can make it an option. Nardog (talk) 22:01, 21 September 2022 (UTC)

Repeatedly Deleted Drafts

I often get notices that drafts that I have created have been deleted as G13, expired drafts. I didn't create the drafts. I moved them from user sandboxes to draft space, and Twinkle notifies me rather than the real originator. That isn't the primary issue I am identifying here, although I would like to see it notify the real originators of the drafts. However, sometimes, I see that the draft topic has been repeatedly deleted as G13. I think that we have a minor persistent problem, which is proponents who repeatedly create a draft, either in a sandbox or in draft space, and then don't do anything with it. For example, see Draft:American Cochlear Implant Alliance, which has been created four times, and deleted once as spam and three times as expired. In its case, the drafts are probably being created by the alliance, and so are straightforward COI, which is probably the most common reason for repeatedly created and abandoned drafts. Does anyone have any ideas as to anything to do about repeatedly created and abandoned drafts (either than delete them repeatedly)? Robert McClenon (talk) 18:54, 10 September 2022 (UTC)

WP:Salt? ϢereSpielChequers 17:50, 22 September 2022 (UTC)

Faciliate access to article history

This is a technical idea, but it's preliminary enough that it belongs here and not in the Technical village pump.

Major issues that this idea aims to address:

• Article decay (experts get article to Featured status; the quality starts decaying after they leave from cumulative non-expert edits). Addressing this will have positive knock-on effects on retention of domain-expert Wikipedia contributors.
• Article decay may also result in expert contributors being turned off from editing Wikipedia, since edits they invest a lot of time and effort into may be reverted one day, and those reversions may not be adequately reviewed by enough editors. Removing disincentives to domain-experts contributing to Wikipedia would go a long way to improve the encyclopedia's quality, expand the number of contributors, and reduce knowledge inequity.
• Low scrutiny for many edits, combined with Warnock's dilemma (WP:WEAKSILENCE). Many editors may assume that "others have reviewed and agreed with various changes and haven't found a need to comment" in instances where no one has reviewed anything. This lack of scrutiny is compounded for articles that are edited frequently (dozens of times a day), or very infrequently (a few times a year).
• Using WikiBlame to find the relevant edit summary for an addition is exceedingly time-consuming, since any sentence has likely been repeatedly reworded; editors need to repeat the WikiBlame process multiple times to distinguish the "insertion" of a rewording with the insertion they are looking for, that of the original addition of the sentence, if they want to read the corresponding edit summary to gain context.
• New contributors may be intimidated by the idea of editing an article they've never worked on before, because they missed out on old discussions in talk pages and edit summaries (that may not have been read or acknowledged by any other editor, see "scrutiny" above). As a result "newbies" prioritize copyediting over content edits due to lack of confidence, which results in fewer contributions to Wikipedia than if newbies could feel that they can easily access the proper context.

Description of the idea:

• It would be nice to add a mouse hover action while in the Visual Editor, that shows the edit summary corresponding to the latest time the highlighted word or sentence was changed; as well as being able to easily enter a "paragraph history mode" that would make it easier to see when a paragraph was first inserted started, and how it evolved. I would assume there are ways to optimize this, in order to minimize server load (for example, for every word in an article, pre-computing the date of the last revision in which that word was edited; this would only need to be done once per article and could be kept up to date with minimal resource usage upon each subsequent edit).
• It would also be nice to have a feature in the revision history that adds icons to the left of "significant" revisions, for example highlighting a revision that was peer-reviewed by 2 editors (see Wikipedia:Peer review), or a revision that passed a Good Article review or Featured Article review. Revisions that implement an article review's suggestions shouldn't look indistinguishable from other revisions. Being able to easily see side-by-side Visual Editor/wikitext comparisons between the current revision, and the last reviewed revision would make it far easier to focus copyediting and review efforts on newly added content, and would be a huge step in addressing article decay. Article decay is a major concern, since for example, there are more Vital articles that were delisted from Featured article status (89), than there there are Featured Vital articles in total (70); preventing decay would lower the maintenance burden, improve the encyclopedia's quality massively, and may help reverse both the decline in overall editors, and the skepticism of some domain-expert contributors who currently see contributing to Wikipedia as a waste of time (see issue #2 above). I've personally witnessed genuine experts retiring from Wikipedia due to being overwhelmed by article quality decay issues, so I feel this is a very important issue to address.

Examples of use cases:

• If I come across the "citation needed" template, it would be nice to see the original edit summary for the sentence's addition, which could contain more context or detail, or would allow me to contact the editor that added the claim. I am sure that many domain-experts have had the experience of adding an unsourced but worthy-of-inclusion claim to scientific articles, and having that claim removed after a few years because non-domain-experts reasonably but inaccurately thought the claim is unsupported by evidence (maybe the claim can be reliably sourced from a book or non-open-access study that is hard to find using Google). This would also make newbie contributors far more comfortable when editing major topics.
• If a sentence has been repeatedly reworded as a result of an edit dispute, or if it's been reworded following a talk page consensus, my proposal would make that far easier to see within the Edit view. If a sentence has been added 3 days ago and never copyedited, it deserves more strutiny than a sentence that was added 8 years ago after a talk page consensus and has been copyedited and improved a dozen times since. My proposal would allow for editing and content review efforts to be more concentrated on passages that need it most; largely addressing the "low scrutiny" issue.

DFlhb (talk) 18:12, 22 September 2022 (UTC)

Some of this is already available via mw:Who Wrote That? -- RoySmith (talk) 18:36, 22 September 2022 (UTC)

Brilliant; I just added a link to this discussion on that tool's talk page. I hope it gets rolled out more widely. It unfortunately doesn't include a "paragraph history viewer" that would be a true godsend. DFlhb (talk) 19:58, 22 September 2022 (UTC)

Unedit/Unsend And Incognito mode - Anonymous experienced editors, and UnEdit

If there was a way for an editor to choose to post, add watch, or edit incognito using their exisitng account would you use it? (Checkuser could see the link, but not Admins)?

Sort of related. Sometimes I see people regretting their edit or email. If you could unedit/unsend (so it was deleted off your history) would you use it? (Subject to no one editing after you, and the User not opening the message? Wakelamp d[@-@]b (talk) 02:04, 25 August 2022 (UTC)

It's perfectly fine to use an alternate account to go "incognito", as long as you don't do so in a deceptive or disruptive way. – Joe (talk) 09:07, 25 August 2022 (UTC)

True. But isn't that inconvenient to log in and out? Wakelamp d[@-@]b (talk) Wakelamp d[@-@]b (talk) 13:20, 25 August 2022 (UTC)

As regards your second question, I'm sure that anyone who has regularly edited Wikipedia is embarrassed by something they have done, but such a feature would just make us less careful in the first place. Phil Bridger (talk) 09:34, 25 August 2022 (UTC)

I was mainly thinking about editors being angry rather than careless Wakelamp d[@-@]b (talk) 13:09, 25 August 2022 (UTC)

Simple rule… don’t post when angry. Problem solved. Blueboar (talk) 13:27, 25 August 2022 (UTC)

Possibly not quite so simple. Unfortunately, there are things that can make one angry or frustrated in any collaborative workplace, whether online such as on Wikipedia, or in the office or in the factory, voluntary or salaried and feelings get blurted out. Wikipedia is however the least tolerant of places and where the slightest lapse can have devastating consequences; mean spirited, governance obsessive people will gang up and get productive editors sanctioned, blocked, or even banned. It's easy when everyone is just a user name. It's harder for those however who were known personally for their highly motivated work for Wikipedia in real life such as organisers and leaders of editathons, for example. Understandably, after years of service they just give up and don't come back. Kudpung กุดผึ้ง (talk) 13:36, 25 August 2022 (UTC)

@Kudpung That was exactly the situation was I was concerned about. There was some research I can't find that disucsses that experienced productive editors often leave after an angry email or comment on their talk page. I think @Blueboar sadly may be unique in never having regret.

I just through of another use - Both people in a conflict could agree to withdraw (rather than delete) their comments. It was normal in duelling :-), but I think it is a good process,

With the people highly regarded for editathons etc, maybe a modification of the anonymous idea would help. The Editor name would stay the same, but there would be an option to link a second private Editor name. Wakelamp d[@-@]b (talk) Wakelamp d[@-@]b (talk) 01:02, 28 August 2022 (UTC)

In RL Blueboar has an admirable career and pursuits which call for people management and a calm disposition, and on Wikipedia he appears to eschew the contentious areas which we often refer to as 'the coal face' or 'working in the trenches'. He may therefore not be familiar with the challenges of maintenance work or the Wiki kind of dispute resolution. Those who choose to be active in the contentious areas here have a Sword of Damocles hanging over them and their position is often precarious - even acting in the very best of good faith in the interest of Wikipedia can get them scolded or sanctioned to the extent they will just walk away and end their years of excellent contributions. Kudpung กุดผึ้ง (talk) 01:39, 28 August 2022 (UTC)

Thanks for the kind words… FYI, I’ve done my time “in the Wikipedia trenches”, and have faced all the aggravating situations you are talking about. I have learned from experience that, when angry or frustrated, the best thing to do is… not type. I take a break and address the issue later, once I have calmed down. Blueboar (talk) 02:19, 28 August 2022 (UTC)

People seem to also use IP accounts to go ingonito< I owonder what Portugal did after they turned off [|IP Editing.

I have looked through Wiki and their seems to be 4 case with a few cases why IP is used

• Can't be bothered, don't think it is needed,
• Privacy (even though they are private with an account)
• Permanent IP editors
• Editors going anonymous to avoid controversy. I imagined an incognoto made mode that still added to history, and both Ip and normal (or even alternate id) would still come to the main account Wakelamp d[@-@]b (talk) 04:49, 1 September 2022 (UTC)

There are many reasons to edit without logging in. To your list, off the top of my head, I'd add "don't want to type my password on this shared computer", and I'd split your last into two. There's the "don't want controversy (I really shouldn't be editing this, but I know enough to know that nobody will object to this edit as long as they don't know who's making the edit)" case and the separate "don't want this controversial content associated with my real account".

More broadly, almost everything listed under Wikipedia:Sockpuppetry#Legitimate uses is also an expected use case for people editing while logged out.

I don't know what work is being done on the Portuguese Wikipedia, but I have of two other Wikipedias that considered requiring registration: The Persian Wikipedia was following in ptwiki's footsteps until they figured out that what they really wanted was an anti-vandal bot, so they voted to keep IP editors. The French Wikipedia recently had a big RFC that ended with a decision to keep IP editors (I don't happen to know the main themes of the discussion, just the outcomes).

I also ran across an interesting, if old, bit of research that determined the most productive editors were the people who first made an edit as an IP, and then (within an hour) decided to create an account. Whatamidoing (WMF) (talk) 22:03, 7 September 2022 (UTC)

"I also ran across an interesting, if old, bit of research that determined the most productive editors were the people who first made an edit as an IP, and then (within an hour) decided to create an account. " That makes sense - It is easier to not create an account To explain

When you choose to edit an articles

1. Popup saying welcome to WP (only happens once)
2. "This is a talk page. Please respect the talk page guidelines, and remember to sign your posts by typing four tildes: ~~~~. You are not logged in. Your IP address will be publicly visible if you make any edits. If you log in or create an account, your edits will be attributed to a username, among other benefits."

This goes against the standard UI experience for login (Popup giving options to Login, or create account, (or in our case do an Ip edit with warnings)). So the question is why hasn't the WMF done this? Wakelamp d[@-@]b (talk) 01:58, 10 September 2022 (UTC)

Would you like the editing workflow to be a bit more aggressive about getting people to login? I could be wrong (I am not an engineer), but a box that says "Please log in or create an account", even if it has a small "skip this and display my IP address" option for those who don't want to be bothered creating an account, doesn't sound like a ton of work to me. (It might require a ton of double-checking that it won't break everything, and they might want to run A/B tests, but we're still talking about months, not years.)

Are people interested in this? If you are, I'm willing to present it to a team and see if they'll do it. Whatamidoing (WMF) (talk) 19:20, 12 September 2022 (UTC)

I think it would be a good idea. It's also got the advantage that it isnt a exisiting editor Use case don't use this part. Wakelamp d[@-@]b (talk) 07:57, 19 September 2022 (UTC)

@WhatamIdoing Is it worth a go as proposal. or should I let it disappear into archive land for the moment? Wakelamp d[@-@]b (talk) 07:40, 26 September 2022 (UTC)

I've started asking around internally. It's much too soon to see whether anything might come of it, but there's no need to worry about whether this discussion gets archived. Whatamidoing (WMF) (talk) 21:18, 28 September 2022 (UTC)

Just to make this a little easier to track, this is phab:T319084 and I've linked there a similar task, which might happen fairly soon. Whatamidoing (WMF) (talk) 23:25, 30 September 2022 (UTC)

Editing of a template before posting

Sometimes , I feel some templates are harsh. I remember in Corrosion Engineering an editor put "Need attention from an expert". I know the person and did my Masters with him at a prestigious university. He is Managing partner in a corrosion engineering company. He is an expert! --- in corrosion engineering. Perhaps not in all the WP policies about formatting, reference format and citations etc. I recently had an editor put a very harsh message about an edit I made in all innocence. Here it is:

Extended content

Copyright problem: Corrosion engineering Hello GRALISTAIR! We welcome and appreciate your contributions, such as Corrosion engineering, but we regretfully cannot accept copyrighted material from other websites or printed works. This article appears to contain work copied from https://dokumen.pub/practical-power-plant-engineering-a-guide-for-early-career-engineers-9781119534983-1119534984-9781119534990-1119534992-9781119535003-111953500x-9781119534945.html, and therefore to constitute a violation of Wikipedia's copyright policies. The copyrighted text has been or will soon be deleted. While we appreciate your contributions, copying content from other websites is unlawful and against Wikipedia's copyright policy. Wikipedia takes copyright violations very seriously, and persistent violators are likely to lose their editing privileges. If you believe that the article is not a copyright violation, or if you have permission from the copyright holder to release the content freely under license allowed by Wikipedia, then you should do one of the following: Have the author release the text under the Creative Commons Attribution-ShareAlike 3.0 License (CC BY-SA 3.0) by leaving a message explaining the details at Talk:Corrosion engineering and send an email with confirmation of permission to "permissions-enwikimedia.org". Make sure they quote the exact page name, Corrosion engineering, in their email. See Wikipedia:Requesting copyright permission for instructions. If you hold the copyright to the work: send an e-mail from an address associated with the original publication to permissions-enwikimedia.org or a postal message to the Wikimedia Foundation permitting re-use under the Creative Commons Attribution-ShareAlike 3.0 License and GNU Free Documentation License, and note that you have done so on Talk:Corrosion engineering. See Wikipedia:Donating copyrighted materials for instructions. If a note on the original website states that re-use is permitted "under the Creative Commons Attribution-ShareAlike License (CC BY-SA), version 3.0", or that the work is released into the public domain, or if you have strong reason to believe it is, leave a note at Talk:Corrosion engineering with a link to where we can find that note or your explanation of why you believe the content is free for reuse. It may also be necessary for the text to be modified to have an encyclopedic tone and to follow Wikipedia article layout. For more information on Wikipedia's policies, see Wikipedia's policies and guidelines. See Wikipedia:Declaration of consent for all enquiries for a template of the permissions letter the copyright holder is expected to send. Otherwise, you may rewrite this article from scratch. If you would like to begin working on a new version of the article you may do so at this temporary page. Leave a note at Talk:Corrosion engineering saying you have done so and an administrator will move the new article into place once the issue is resolved. Thank you, and please feel welcome to continue contributing to Wikipedia. Happy editing! BalinKingOfMoria (talk) 00:06, 23 August 2022 (UTC)

So even a suggestion I could get banned from editing!.

Am I missing something? Is there already a way where templates such as this can be toned down and less harsh?

GRALISTAIR (talk) 14:58, 28 September 2022 (UTC)

This is hardly harsh. A WP:COPYVIO is very serious and requires this tone. Meanwhile, {{uw-test1}} has a light tone. Sungodtemple (talk) 15:32, 28 September 2022 (UTC)

Fair enough if I am over reacting GRALISTAIR (talk) 19:02, 28 September 2022 (UTC)

@GRALISTAIR: This may feel harsh and I understand why. Coming from someone whose first (and only) article was on the verge of speedy deletion due to it. See, as copyright violations can cause legal issues, it is seen as a very serious issue on Wikipedia, more so than minor editing disputes. Basically, you need to rephrase anything you write in your own words and you're fine. You can use copy-paste material only if certain specific copyright licenses have been used by the original creator/publisher, which might not have been the case here. —CX Zoom[he/him] ^{(let's talk • {C•X})} 19:43, 28 September 2022 (UTC)

While it's certainly intimidatingly long, at least it starts with We welcome and appreciate your contributions, such as Corrosion engineering, but we regretfully cannot accept copyrighted material from other websites or printed works. I'm sure efforts to shorten it while preserving meaning would be welcome. Enterprisey (talk!) 01:47, 29 September 2022 (UTC)

As I did this inadvertently, is there any easy way to check any article for copyright violation. GRALISTAIR (talk) 22:29, 29 September 2022 (UTC)

@GRALISTAIR you can use tools such as Earwig's Copyvio Detector to search for copied text. It is only a tool, and results must always be manually validated. Keep in mind, it often finds external sites that are violating our copyright! — xaosflux ^Talk 00:35, 1 October 2022 (UTC)

Capital letters in titles of entries

Moved from Wikipedia:Village pump (policy)

This is a proposal for all wikipedias. Would WMF consider: The titles of entries at Wikipedias, to be written exactly as they are written in their languages.

• No uppercase initial letter without grammatical reason.
  
• Case sensitive (in unison with the Wiktionaries of the same language).

Thank you. From el.wiktionary, Sarri.greek (talk) 18:31, 29 September 2022 (UTC)

@Sarri.greek: note, I moved this to VPI, as it is not a ready-to-go policy proposal specific to the English Wikipedia. — xaosflux ^Talk 18:40, 29 September 2022 (UTC)

There are technical reasons why titles start with capital letters; see WP:NCLOWERCASEFIRST. This change would also mess up a bunch of links with words at the start of sentences that are capitalized; it currently doesn't matter if the first letter is capitalized, but this change would break that. To me, it would cause too many issues to make the change worthwhile. RunningTiger123 (talk) 21:22, 29 September 2022 (UTC)

It would be a major upheaval but might still be worth discussing. It might not affect too many articles. I just clicked Random article twenty times and none were within scope (18 were or began with proper nouns; two were genera). There are advantages: apple could be the fruit and Apple the company. There are, of course, also disadvantages: most affected pages would need a new redirect from initial capital (to avoid millions of sentences needing piped links: [[pear|Pear]]s grow on trees.) Finally, having an article titled pear rather than Pear would be unconventional and take some getting used to. Certes (talk) 23:11, 29 September 2022 (UTC)

@Sarri.greek, what problem would this be trying to solve? What would be the point of this major change? Schazjmd (talk) 23:18, 29 September 2022 (UTC)

[answers by Sarri.greek]: I started a discussion here, a) because this is the largest and most infulential WPedia, with the most experienced editors. And b) I wouldn't know where in WMF I should have asked: also, the upper 'administration' would not be the first, but the last level to approach, after hearing from the front line.

Technical issues for case-sensitivity (@RunningTiger123:, thank you for the link), are outside my knowledge, but we know (from Wiktionaries) that it is possible. Anyway, even if we forget case sensitivity, the lemmata can be changed to lowercase, redirects updated, etc.

Yes, @Certes:, I am aware that it is a «major upheaval». So, why? («what is the point?» @Schazjmd:). Because the spelling of a word should not be a lie, should not be 'not true'. It misleads the readers.

• the first sentence of all lemmata entries (often, I see, called 'articles') desperately tries to 'fix' the problem by stating the correct spelling. Example apple (Thank you Certes for the handy example): «An apple is...»
• a lemma, an entry, is not a book, is not a treatise, is not a PhD, ... to have a title (with uppercase initials). It is just an entry, exactly as a lemma in a dictionary. It should not be different.
  I confess, that I have not done any research to existing encyclopaedias (printed) to see their style-rules. But even if they had uppercase for all their entries, I would still disagree.

Coming from a wiktionary, my training and 'ethics' are lexicographic and not from WPedia. But why these should be different? The word! the word! (words, words, words and not Words, Words, Words) are the beginning of everything! Thank you all. Sarri.greek (talk) 01:36, 30 September 2022 (UTC)

Implementing a change is a balancing act; do the benefits of the change outweigh its harms? Here, it's weighing the benefits of showing article titles in lowercase versus updating hundreds of thousands (if not millions) of article titles and millions of links within pages. I would argue those downsides are far, far greater. I have reasonable confidence that most readers recognize capitalized titles for words that are not normally capitalized as a stylistic choice, not as the actual use of the word(s), which renders the change generally unhelpful. RunningTiger123 (talk) 01:44, 30 September 2022 (UTC)

I would support this change to distinguish upper and lower case in first letter. Technical restrictions can be overcome as happened for Wiktionary. For some Chemistry articles an upper case letter for starting a sentence should be later in the title, eg 1,2-Dichloroethane or alpha-Ethyltryptamine. And if we use the Greek latter α it capitalises to a confusing Α. We can override this with DISPLAYTITLE though. For some astronomy articles it makes a difference if it is b or B eg B Centauri has to be a disambiguation page with alternative naming. for Brown is it the colour brown or the name Brown? For common nouns we can create a capital redirect. And these would be produced if the change was introduced, and articles renamed. Graeme Bartlett (talk) 07:33, 1 October 2022 (UTC)

Ask Top sources for Share Cite Option

Problem : Wikipedia:Bare URLs

But : They should use {{cite web}} or one of the Browser addins such as Except they don't

Proposal : Approach the content providers and ask for Cite to be added to their share option. Stop Bare Url at software, by asking developers of popular Github projects eg Social Share Privacy

In Australia, our National Archive has this option already as an alternate to Template:Cite Trove newspaper Wakelamp d[@-@]b (talk) 07:15, 23 September 2022 (UTC)

Nup - drop it. I can't work out how they could create a bare Url with vector 2022 Wakelamp d[@-@]b (talk) 13:54, 2 October 2022 (UTC)

Signpost for all (second time)

Could the Newsletter Extension be used to do this? Wakelamp d[@-@]b (talk) 03:07, 24 September 2022 (UTC)

@Wakelamp The newsletter extension is unsupported and will not be installed on any new wmf sites. See meta:Limits to configuration changes#forbidden-extensions. 192.76.8.81 (talk) 10:35, 25 September 2022 (UTC)

And, we would never forcibly subscribe 44million users to a notification. — xaosflux ^Talk 11:23, 25 September 2022 (UTC)

Many people deeply resent any bulk message that can be perceived as spam. Cullen328 (talk) 01:56, 29 September 2022 (UTC)

And many others deeply resent not being personally notified about things.

If you were going to spam everyone, you'd probably want to limit the notifications to highly active editors (e.g., ≥1,000 edits) and only people who have been active recently. I'd guess that would mean contacting something on the order of twenty thousand editors. You could reduce this by limiting it to people who have made ≥100 edits this year.

It's generally considered polite to do this only once, and to require people to actively opt in after a single notice. Whatamidoing (WMF) (talk) 23:08, 30 September 2022 (UTC)

@Cullen328, @Whatamidoing (WMF) What about as has already been sort of suggested if the watchlists top dismissable message appeared addiionally in the same position as '"you have talk? Would that be too spammy?And I agree about the limits

I agree with the spam concerns, but the other side is how do we increase editors and I think that means community. @Whatamidoing (WMF) discussed on proposal (i have forgotten which, so apologies) that people have to be conscious of the ratio of WP to Main.

Myabe Main is attractive, becuase editing we do is solitary except for 3 or 4 projects, dev, and various admin things. Solitary is great, but it's nice to have some community.

The signpost is global, but also a way is needed to have people feel they have purpose and support as they edit. I was looking at Phab, and I was thinking that it was like a talk page, and wondered if there was someway of creating a newfeed with a nice UI for an area of interest, where experienced editors could edit. respond and see what each is doing.

(And I would also like to thank everyone in general for discussing these ideas.) Wakelamp d[@-@]b (talk) 11:52, 2 October 2022 (UTC)

Related discussion on Meta:Babel about mass messages.

Wakelamp d[@-@]b (talk) 14:13, 7 October 2022 (UTC)

Images of (Historical) Religious figures

Recently during this discussion about images of Mary, I noticed that almost everyone had differing ideas about what purpose images of religious figures should serve on Wikipedia - and there didn't seem to be any policy, guideline, or MOS, that provided clarity. Should they try to be accurate to when the person lived? Or to common modern portrayals? or to famous depictions? or simply what looks best?

While I understand that for modern religious figures such as Pope Francis this isn't going to be as much of an issue as we can simply use photographs, for figures such as Jesus and Gautama Buddha this is likely to recur.

So my idea is: create a policy or guideline which establishes what criteria an image of a (historical) religious figure should have to meet for inclusion, and which can be used to help editors decide when to use one image over another.

Your thoughts and ideas would be greatly appreciated. Tomorrow and tomorrow (talk) 22:55, 7 October 2022 (UTC)

​​​​​​​​​ Just hidden-pinging everyone who contributed to the discussion on Mary as they may be interested. Tomorrow and tomorrow (talk) 00:37, 8 October 2022 (UTC)

@Tomorrow and tomorrow - Thanks for the ping :) — That Coptic Guy (talk) 01:34, 8 October 2022 (UTC)

I think it's usually going to be best to balance the 4 criteria you mention above. For example, the earliest (beardless) depictions of Jesus might satisfy the first criteria, but not the other three.

As a side note, perhaps it might be better to have a pool of suitable lead images and switch between them every now and then. 𝕱𝖎𝖈𝖆𝖎𝖆 (talk) 00:56, 8 October 2022 (UTC)

Thanks @Ficaia. I'm not sure I agree with you about the rotation of images but that's a bit of a side note.

Taking what you said about balancing, which I broadly agree with - would you be supportive of a criteria model based on a reasonable balancing (where possible) of:

• Historical accuracy
• Similarity to common modern and historical depictions
• Clarity and completeness of image

I've rolled modern and historical depictions together, and I think we could maybe say in the case of a significant divergence use a modern picture alongside a historical one. And I've reworked the "looks best" one, in an attempt to make it less subjective. Tomorrow and tomorrow (talk) 01:08, 8 October 2022 (UTC)

Seems fine to me. 𝕱𝖎𝖈𝖆𝖎𝖆 (talk) 01:16, 8 October 2022 (UTC)

• I think this is pretty unlikely to be accepted. For the foundational generation(s) of the major religions, you can forget about "Historical accuracy" completely, but this is very important for later figures, when there is a chance they derive in some way from an actual likeness. Otherwise early images of high quality (both the work and the reproduction file) are usually best. I'm not against rotating images at long intervals - say annually. Johnbod (talk) 02:14, 8 October 2022 (UTC)
• So, in general, in its current state, I think this would be pretty difficult to implement, simply because we don't know much about what Christ or St. Mary looked like officially, for instance. Even the most accurate image of Jesus we have so far, the Shroud of Turin, hasn't been officially accepted as the depiction of Christ. Speaking as to Jesus and His Mother, the problem is that there are so many different depictions of them that differ by the time period in which the painting was drawn, culture, and other interpretations. There's just so many out there... That said, I'm thinking more along the lines of what 𝕱𝖎𝖈𝖆𝖎𝖆 said, in that we should keep the images fresh, instead of having the same one for years on end. I like the rotation idea, as well as satisfying the 3rd criteria you set forth (Clarity and completeness of image). — That Coptic Guy (talk) 02:24, 8 October 2022 (UTC)

Is this about WP:LEADIMAGE specificly? Fwiw, here's a similar discussion:Talk:Moses/Archive_6#Not_a_fan_of_this_image_of_Moshe. I know there's been several at Jesus. Gråbergs Gråa Sång (talk) 08:58, 8 October 2022 (UTC)

• The lead image for a religious figure should be something close to a "typical" depiction of that person, and if you ask me the most useful purpose it can serve is to help readers identify other images of the same figure. Obviously the more types of images of a figure there are, the harder it will be to find a single typical image, which will be the case for Jesus and Mary. But in the case of a Christian saint, say, the image should have the usual attributes of that saint in artistic depictions – and these should be fairly prominent in the image. Some bad examples are the current lede pics at Mark the Evangelist, Luke the Evangelist and John the Evangelist, which are clearly intended to form a "set" of high-quality images, but you can barely make out any distinguishing features. And the set isn't even complete, because Matthew the Evangelist is also Matthew the Apostle and so is part of another set with the Apostles Peter, Andrew, John, James the Greater and the Lesser, Philip, Bartholomew and Simon the Zealot, which all use paintings from a set of Twelve Apostles by Rubens (unlike Thomas and Jude, which don't). Ham II (talk) 15:23, 8 October 2022 (UTC)

There is indeed a guideline that regulates this sort of thing and I find it perfectly reasonable. MOS:LEADIMAGE:

Lead images should be natural and appropriate representations of the topic; they should not only illustrate the topic specifically, but also be the type of image used for similar purposes in high-quality reference works, and therefore what our readers will expect to see. Lead images are not required, and not having a lead image may be the best solution if there is no easy representation of the topic.

Surtsicna (talk) 15:32, 8 October 2022 (UTC)

I am not sure that we need any more advice than we already have in Wikipedia:Manual of Style/Images#Choosing images:

"Strive for variety. For example, in an article with numerous images of persons (e.g. Running), seek to depict a variety of ages, genders, and ethnicities. If an article on a military officer already shows its subject in uniform, then two more formal in-uniform portraits would add little interest or information, but a map of an important battle and an image of its aftermath would be more informative."

So put up several images, and whenever you can source it, add a caption that says something like "this depiction is thought to be realistic" or "this depiction was popular during the ____ century" or "this depiction is typical of <region>". WhatamIdoing (talk) 17:34, 10 October 2022 (UTC)

Adding more debugging tools to modules of Wikipedia

Hi, nowadays in an IDE like Eclipse, debugging tools like setting a breakpoint, and line to line execution, etc. are very common and helpful tools. For example in Eclipse, after setting a breakpoint, and executing a debug action, the value of all variables can be viewed in a sidebar window at that breakpoint. This and other similar debugging tools can be implemented at server-side for Lua modules of Wikipedia, by partial submitting, resulting in some HTML windows containing the values and types of variables.

For example after setting a breakpoint at line 2545 of Module:Wd and doing a partial submit, the programmer can see the values and types of all variables at that point of code in some returning hide/show HTML windows. I should note that this capability is very common in contemporary IDEs, but we can implement similar idea for WP modules, by partial submitting. Please discuss about my idea. Thanks, Hooman Mallahzadeh (talk) 15:24, 11 October 2022 (UTC)

Idea for sports articles

Wikipedia has a history of sports -- all kinds of sports-related articles. Enthusiastic editors for sports articles are... enthusiastic. Editors that aren't enthusiastic about sports articles are ... also enthusiastic about deletion.

Here's my idea: would/could/should the Wikimedia Foundation create a free online "sports almanac" specifically for extensive sports-related articles? That would create an outlet for those enthusiastic about sports and also create a limitation in Wikipedia. I'm not sure if it's a "good" idea, but it is an "idea" -- thoughts?--Paul McDonald (talk) 14:43, 4 October 2022 (UTC)

It is true that we have bias. I think before you resort to creating a new Wikimedia project, maybe try to change the WP:NSPORTS-related deletion guidelines. I think the basic idea is that notability is when you have enough sources such that you know the information in the article is accurate (v) Sungodtemple (talk) 21:17, 4 October 2022 (UTC)

By the way, I'm in the 'aren't enthusiastic about sports articles' category. Sungodtemple (talk) 21:19, 4 October 2022 (UTC)

I'd be all for it. Along with changes to Enwiki notability guidelines. For example the sports world cranks out medals and awards, naturally; our guidelines follow this trail and we end up with excessive numbers of articles. -- GreenC 23:30, 12 October 2022 (UTC)

Proposed mandatory admin recall procedures

Any editor may open a petition to recall an admin, unless the admin in question has passed an RfA or recall RfA within the past year.

If the petition receives more than twenty signatures in thirty days, the admin in question would be required to open a recall RfA within thirty days of the twentieth signature being applied. This recall RfA would function in the same manner as a standard RfA but with reduced requirements; the discretionary range would be between 50 and 55 percent support.

Admin's who don't open a recall RfA within the thirty days period are automatically desysopped, but this period may be extended at the discretion of the bureaucrats; admins who are granted an extension may not use their admin powers during the extension period.

This was prompted by the recent restoration of Staxringold's permissions, but I hope it would also lower the stakes at RfA. I also believe it will, through the use of reduced requirements to pass the recall RfA, and through limiting recall RfA's to one per year, address past concerns about mandatory recall procedures. BilledMammal (talk) 02:15, 28 September 2022 (UTC)

A spicy topic, to be sure. No comment on your proposal because I haven't thought about it enough, but onlookers not familiar with the history may want to check out some of these RfCs, especially the latest one in 2019. Enterprisey (talk!) 06:41, 28 September 2022 (UTC)

I would suggest the wording to be changed to "recieves more than twenty signatures from editors in good standing in thirty days". I would support this proposal, but I think if the proposal applies to current admins, it would probably result in opposes from some admins. However, only having this for future admins seems a bit unfair and will take a long time before actual solving the problem suggested by the recent event. 0xDeadbeef 09:25, 28 September 2022 (UTC)

• This will never fly. Smacks to much of mob justice, and we already have enough of that: peanut gallery at the gutter level and from a coven of governance obsessives at the highest coveted rank. Even a solution launched by me and Worm That Turned was turned down.Kudpung กุดผึ้ง (talk) 11:11, 28 September 2022 (UTC)

  I'm not sure I'd call myself a governance obsessive, despite being entrenched in said coven.

  Regarding the proposal though, @BilledMammal, the community has recently stated clearly that The atmosphere at RfA is deeply unpleasant at the most recent RfC last year. Every individual who would go through your proposed process would be divisive, simply by getting to the point that they are in the recall RfA, meaning that they would instantly be put through a massively unpleasant process. I cannot get behind a process which is designed to humiliate and I struggle to see that using RfA in these circumstance is anything else Worm^TT(talk) 14:07, 28 September 2022 (UTC)

• Mandatory recall is not the way to go about admin accountability. If someone is able to get 20 signatories for demotion, there is enough for an ArbCom case. I'm not saying that requiring ArbCom is a good process, but this proposal seems redundant. Anarchyte (talk) 14:05, 28 September 2022 (UTC)

This isn't a solution it just means that "unpopular" admins have to jump through hoops every year until they don't want to be an admin anymore. Terasail^[✉️] 14:10, 28 September 2022 (UTC)

If someone breaks policy without a good reason then make it easier to remove the admin, this could impact people who are not breaking policy but a group of users just doesn't like them, which is entirely unproductive. Terasail^[✉️] 14:13, 28 September 2022 (UTC)

• I am leery of this. A lot of admins already clearly avoid closing contentious discussions because they know that no matter how the discussion is closed, it will be challenged with recriminations against the closer. I can imagine how much more stressful those closes would be if the closer faced the prospect of a desyssop effort. BD2412 T 23:35, 28 September 2022 (UTC)
  • Welcome to Wikipedia, where the reward for someone with decades of experience doing disagreeable work for free is a week-long bout of criticism and dug-up grudges, but it'll be ok because it can "only" happen once per year. —Cryptic 23:57, 28 September 2022 (UTC)

  A similar problem on the horizon is that the admin may be tagged as harassing Wakelamp d[@-@]b (talk) 00:17, 4 October 2022 (UTC)

  Ooh, I remember that "tool". —Cryptic 00:40, 4 October 2022 (UTC)

• It would be useful to have a standard recall procedure. As it is, it is up to each admin to create their own. Hawkeye7 (discuss) 00:07, 29 September 2022 (UTC)
• It seems that no one linked the actual last one, Wikipedia:Requests for comment/Desysop Policy (2021). 0xDeadbeef 07:17, 2 October 2022 (UTC)

  In the last RfC, there was no WP:AARV, and therefore a requirement was for an AN(I) thread to be closed with the consensus being the admin behaving inappropriately for the process to kick off. I think it would improve if the requirement was instead changed to "at least two overturned admin actions in one year, OR an AN(I) thread closed within six month as the admin behaving inappropriately". 0xDeadbeef 07:42, 2 October 2022 (UTC)

  Need to make sure that "overturned" is based on a consensus of a discussion that there was a substantive mistake of judgement. Good admins shouldn't feel threatened to make a hard and urgent call early at risk of later discussion with more ideas and new details. Judgement might be fine but an appeal and second thought had. And many actions might be unquestionably correct at the time, but later overturned when facts of the situation have changed. As the simplest example, I close an AFD 'delete--unanimous--nonnotable/TOOSOON', and years later someone successfully appeals at DR that it has eventually become notable. That's an "overturned deletion" but was absolutely the correct action to have taken. DMacks (talk) 03:07, 18 October 2022 (UTC)

• oppose it's hard enough to be an admin and try to do good stuff without an angry mob able to just drag you in front of a new RFA whenever they get enough support. Andre🚐 00:20, 4 October 2022 (UTC)

Wikipedia in More "Simple" Languages

This would be just like the Simple English Wikipedia. It would benefit people needing simpler and easier to understand descriptions in their native language AND benefit people who are learning a language or are new to a language. Obviously, this idea would likely be used for major languages (Spanish, German, French, etc.).

I think something like this would really benefit wikipedia and its users. KeshTM (talk) 13:39, 11 October 2022 (UTC)

The English Wikipedia doesn't have control over the creation of Wikipedias in other languages. If you're interested in working on this, consider checking out m:Help:How to start a new Wikipedia. RunningTiger123 (talk) 13:56, 11 October 2022 (UTC)

KeshTM consider bringing this up at Meta-Wiki. 104.37.211.87 (talk) 18:26, 12 October 2022 (UTC)

We don't even need the Simple English Wikipedia. --User:Khajidha (talk) (contributions) 16:56, 17 October 2022 (UTC)

@KeshTM: I agree with your idea, but the question is who would write that? Most of the non-English projects suffer from severe lack of editors already, edit sourceand by splitting it further, we will get two articles on Earth that aren't all very different but that will take away so much editor time and effort that we may never get an article on Triton in either of those. Also, the English Wikipedia doesn't have any control over what new projects are created, Meta has. —CX Zoom[he/him] ^{(let's talk • {C•X})} 17:16, 17 October 2022 (UTC)

Overlaps with the user friendly topic above Wakelamp d[@-@]b (talk) 03:45, 18 October 2022 (UTC)

I support the sentiment of this proposal completely, I agree that it would be useful for the users.

I don’t agree that it would lead to a lack of focus for the editors, spreading their limited time too thin. For one, it’s conceptually understood on a general level that the “Simple” language version is in a way secondary to the “standard” language version. The simple language version is clearly more of a pedagogical aid than a ubiquitous information repository. Therefore, there would be no ambiguity for an editor in which article they should work on. The standard version would always appear as the primary focus. Someone who specifically wanted to adapt a Wikipedia article into simpler language could work on the Simple language version - only if they had that much more niche, specific goal.

Another important point is that it isn’t as hard to edit Simple Wikipedia because it amounts to rephrasing the original corresponding Wikipedia article. The original is much more about research and citation. The simple version can be seen as completely derivative of the standard version: you already have all the reference information you need in one place. Wikipedia already has a “translation” mode to facilitate easily converting an already robust Wikipedia article into a different language, saving you the toilsome research process. Simple Language Wikipedias amount merely to translation. It’s a downhill ride because someone who’s already versed in that topic can easily rephrase themselves in simpler terms. It’s a walk in the park. It’s also inherently rewarding if you care about a topic, to feel like you’re making it more accessible to a wider world, rather than burdensome. The idea that “more is bad” could imply that we shouldn’t have Wikimedia or something, since it detracts from time and effort put into Wikipedia. It’s not true. If people find something valuable to do, a way they want to contribute to Wikipedia, they will. If you build it they will come. More is better.

I suggested we make difficulty levels a standard feature of all Wikipedias here, then observed this post expressing a similar sentiment.

https://en.m.wikipedia.org/wiki/Wikipedia:Village_pump_(idea_lab)#Make_graded_article_levels Juliushamilton100 (talk) 14:39, 18 October 2022 (UTC)

Articles for Simple Wikipedia are not, in the general case, just translations of the English Wikipedia article. Writing for a lower language level can mean that a different article structure is preferable, and a topic might be broken down or grouped together into different articles. As an analogy, an elementary school textbook is not just a translation of a high school textbook. By all means, anyone interested in encouraging more article writing at different reading levels should go ahead and work on whatever initiatives interest them. It will involve more time and effort, and thus either more editors or more time spent by existing editors. isaacl (talk) 21:50, 18 October 2022 (UTC)

Make graded article levels

Wikipedia Simple English appears to have been very successful, well received, wide range of contributors, people seem to see the value and purpose of it.

However on a conceptual level it actually doesn’t make as much sense to me that this would be considered a separate language. As a native speaker I sometimes wish to read the Simple English version of a page just because the topic is so dense, usually something theoretical in computer science, mathematics or physics, for example. It’s like the concept of “Explain like I’m 5.”

Also, it doesn’t seem there’s any other “Simple” language site for any other language than English, which doesn’t feel right - whatever reasons Simple English is useful must surely apply to others languages as well. Of course, English is the largest language on Wikipedia, but there are other languages large enough to warrant this (including French, German and Russian).

To me, “level/difficulty” should be abstracted away from “language” - it should be a separate parameter. I think it would be really easy to implement. Just create a little icon at the top of the page for a difficulty level of 1, 2 or 3, or “Beginner”, “Standard” and “Expert”. It can be optional, as in it doesn’t appear for pages which don’t have difficulty versions.

I think this makes enormous sense because Wikipedia has become a universal information resource but it doesn’t acknowledge in its design that reference materials do and should come in levels, depending on who’s reading them - a five year old kid or a Professor?

Having this feature could actually influence people to take advantage of it - once people realize you can make different levels, I believe Wikipedians would feel intrinsically motivated to do so for the articles they care about. They would find it fun.

It also helps keep articles more organized. Authors would think more about what kind of information should go in which levels. It could help decrease the sense of “jumbledness” that some articles have.

Also, it’s significant to note that the current digital Encyclopedia Britannica actually already does this.

I read that this forum is for airing ideas before formally proposing them.

So, any experienced Wikipedians out there:

What steps do you think I should take in preparing a formal proposal?

What obstacles might I face, or what opposing opinions, about why it’s not a good idea?

What would I need to consider about implementation? I believe pages have “Index pages”, don’t they? So the page versions could be bound together on one index page.

A preliminary set of steps could be:

-> prepare a draft of source code for implementing a “level” icon on all Wikipedia pages - like the “languages” icon, it triggers a drop down menu with the available difficulty levels. If there are none, it doesn’t show. Also, every page has a “standard”, default version/level.

-> Consider how to link the pages together. I need to learn more about the software architecture of Wikipedia pages but maybe it could be as simple as a special tag at the top of the Wikitext with a link to the other versions. The page could be stored under a url structure which introduces another parameter, like:

https://en.m.wikipedia.org/wiki/Offham_Hill

https://en.m.wikipedia.org/wiki/Offham_Hill/l1

or

https://en.m.wikipedia.org/wiki/Offham_Hill/beginner

https://en.m.wikipedia.org/wiki/Offham_Hill/expert

So there’s my proposal.

I’d love for experienced Wikipedians to air their opinions on why they don’t think it’s a good idea, or if they think it is, any advice and assistance on the concrete steps for trying to make it happen.

Thanks very much. — Preceding unsigned comment added by Juliushamilton100 (talk • contribs) 14:18, 18 October 2022 (UTC)

I think the idea of simple version of other languages is wonderful. The only question is whether there's enough interest to make it happen. You should read meta:Proposals for new projects and/or meta:Requests for new languages to learn about the process of proposing a new wiki. From what I can see from the list of past proposals, simple language versions of existing languages have been suggested a number of times and most of them of been rejected. meta:Requests for new languages/Wikipedia French Simple 3 looks like the only one that even got out of the starting gate. A good place to start would probably be to read that proposal and some of the rejected ones to learn what it might take to be accepted.

You should also look at Wikispore, which is relatively new. I don't know much about it beyond that it's supposed to be a project incubator space. That might be a place to get your project off the ground. Maybe pick 100 topics from the French, German, or Russian wikipedia and convert them to simple language versions in wikispore. Then, at least you'd have some kind of concrete progress to point to when trying to convince more people to help out. "I want to do X" is never as powerful an argument as "Look at the start I've made on X, who wants to help me work on it more?" -- RoySmith (talk) 22:53, 18 October 2022 (UTC)

Incentivize WP:Vital article improvement

It's been a few week since a series of threads popped up about this issue, and it's pretty clear in these discussions that improving Vital articles is important. I tried to apply that to practice with the WikiProject Vital Articles, but it seems to me that we can do more than that. Should a contest like WikiCup be set up? (like WP:The Core Contest but year-long) CactiStaccingCrane (talk) 14:33, 4 September 2022 (UTC)

I think the problem goes far beyond incentives. Wikipedia has milions of articles and the attention and energy of most editors gets wasted on non-vital, rarely-viewed articles. The Wikimedia presentation is extremely clear on that. The Wikipedia:WikiProject_Vital_Articles isn't being promoted enough; only a minuscule minority of editors will ever visit the Village pump, WikiProjects, or any of the sort. I think the Vital Articles WikiProject should be made very prominent in all the places where people learn how to edit Wikipedia, like the the "How to contribute" guides, Help:Introduction, the Edit an Article section on Help:Contents. Ideally the Wikiproject would be permanently linked in the sidebar, and would be promoted in a sitewide banner like the ones we regularly see. There needs to be some kind of large, central effort; right now barely more than a dozen have joined the Vital articles project. A lot of vital articles are languishing; most of them haven't seen a truly significant revamp in more than a decade. It's perfectly fine that editors create articles for high schools or TV show episodes, but I wonder how many of them would continue if they knew about the poor quality of many vital articles. DFlhb (talk) 19:39, 13 September 2022 (UTC)

Adding to this: I think Help:Introduction and Wikipedia:Task_Center (linked to in Help:Introduction) in particular need to make the WikiProject Vital Articles extremely prominent. The task center should be simplified to the bare minimum: when you want help on where to start, and are met with an enormous list, that can be extremely discouraging. Move most of the Task Center contents to Wikipedia:Maintenance (which is linked at the top; and should also be simplified, BTW), and only list the most important things: contributing content, fact-checking, and copy-editing.

I generally think the entire non-content part of the encyclopedia (everything [[Wikipedia:]] ) should be dramatically simplified and cleaned up, since it overwhelms newcomers and makes them feel lost. There's further things the Wikimedia people can do to encourage quality contributions, like gamification, but I just don't think they have the right focus. Poach Duolingo's staff, hire a hundred academics fulltime to contribute to articles, and watch Wikipedia get revitalized in just a few years. DFlhb (talk) 20:10, 13 September 2022 (UTC)

DFlhb, I don't feel comfortable promoting the Vital WikiProject, but given that the first GA drive has lots of trouble gathering members even after MassMessages, I guess that there is no other choice... In my opinion, some of the proposals would be easy as pie to impliment (Intro, Task center, etc.) while others would be neigh impossible (add the project to the sidebar for example). There should be a fair amount of discussion about this given the scale of the proposal. I'm gonna ping User:Sdkb as they has done work at rewamping these sort of pages. CactiStaccingCrane (talk) 11:12, 14 September 2022 (UTC)

Good points, and good idea on the ping. Editors' efforts being spread too thin and concentrated in the wrong areas seems like a major Wikimedia Foundation viewpoint, so I hope this can be addressed both comprehensively and (hopefully) promptly by the community here. DFlhb (talk) 11:37, 14 September 2022 (UTC)

To address the deeper point:

I don't want to be pessimistic but as a newcomer (who probably won't have much time to keep contributing so much) I get the strong impression that Wikipedia is collecting dust. The Technology page was absolutely dreadful, below the quality of a high school essay in some sections before I revamped it (still not done).

I checked, and the page has basically only received minor improvements since 2008-2010. It seems that when it comes to Vital articles, most people are more interested in minor changes, grammar, etc than in actually improving content quality (likely due to lack of time). Getting Vital articles to encyclopedia quality is a lonely effort. WikiMedia has remarked on the decline of active editors, but I'd also guess that there's also been a shift from major content-based contributions to more minor spelling, grammar, or maintenance edits. I'm becoming skeptical that a volunteer effort can properly prioritize efforts and achieve high quality. Only 70 of the 1000 vital articles are Featured articles. There are more Vital articles that were delisted from Featured article status (89 total), than there there are Featured Vital articles in total (70!). Looking at the Featured Article log, the number of monthly promotions is way down since 2008. And of the 205 articles that became Featured Articles this year, only 6 were Vital articles. Last year, 9 out of 325 newly Featured Articles were vital. That's way down from 2008, when 33 articles were promoted to Featured that are considered Vital today. I used Petscan to compile all these numbers.

An incoherence I see in the Wikipedia model is that collaboration and "free knowledge" doesn't necessarily mean zero central coordination; yet Wikipedia's managers seem to confuse the two ideas. I don't think you can have a ship without a captain. If Linux wasn't heavily coordinated, with domain experts in charge of certain parts of the code, it might flouder similarly.

At this point, I think the changes I proposed above, while crucial, won't be enough to fix the problem. The main problem I see is people's lack of desire to take hours and hours to improve contents of articles they might not be personally interested in; resulting in a focus on irrelevant articles or minor maintenance changes. The only true solutions I see are:

1. Wikipedia hiring and paying a few hundred academics to revamp all vital articles

2. Developing GPT-style AI that's advanced enough to read the major literature, summarize it, cite it and format it well, follow the MOS, and let it become the predominant contributor to Wikipedia. Doesn't exist today, but companies like Meta Platforms are working on this. Should exist within the next 5 years; and Wikimedia should make developing this its number 1 priority.

Option 1, Wikimedia can do today; they have the funds. Option 2 is a long-term investment the Foundation should make that will comprehensively fix Wikipedia's coverage and quality issues, far better than any of their proposals that I've seen. I have very few hopes they'll listen. DFlhb (talk) 12:38, 14 September 2022 (UTC)

Thanks for the ping, @CactiStaccingCrane! I don't think emphasizing vital articles in onboarding pages can be the solution — space on those pages is extremely tight, given all the information we need to teach there, so there's not really room, and many people don't read those pages anyways.

I think the solution needs to come in the form of tools that make it easier to see one's impact, which will then lead to a culture that values that impact. Right now, our best metrics are things like edit count and GA/FA count, which are extremely crude. When at RfA, I'd much rather be able to know how metrics like an editor's authored words  ×  pageviews on those words. The Growth Team's impact module is a rudimentary form of those impact metrics, but I've been encouraging @Trizek (WMF)/@MMiller (WMF)/@KHarlan (WMF)/KStoller-WMF to improve it and make it more useful for experienced editors as well. Making it work well will take some technical effort, but I think it could have a transformative impact on Wikipedia's culture, as it'd incentivize editors to make meaningful contributions to important articles rather than just gnoming. Fundamentally, right now it just doesn't feel any different to edit an article with 50 pageviews a month vs. one with 50,000, and until it does feel different, no amount of pleading to go edit important articles instead of niche ones will have enough impact to truly move the needle. Cheers, {{u|Sdkb}} ^talk 23:20, 14 September 2022 (UTC)

Although I think it is harder to make substantial contributions to articles for major topics, I think it can in some cases be easier to find editors interested in them than it is to find editors interested in making tidy-up edits, which go largely unheralded. As long as the Wikipedia community is welcoming to anyone to edit, it needs gnomes to shape the raw content added by editors unfamiliar or uninterested in Wikipedia standard writing conventions. I agree that ideally major topics would get more attention, but I disagree that we should be trying to get editors to edit them by dropping their tidy-up efforts. isaacl (talk) 23:57, 14 September 2022 (UTC)

Fair point. But I disagree gnoming is currently unheralded — it puts you on a million watchlists, making your username more recognizable, and it rapidly boosts your edit count (which despite all protestations to the contrary does mean something to most editors). We need both gnomes and writers; we have enough of one, but definitely not the other. {{u|Sdkb}} ^talk 00:23, 15 September 2022 (UTC)

I wonder if it would be feasible and useful for a tool to show "viewed edit count" or similar. Each edit would be multiplied by its pageviews before summing, rather than edits to Charles III and Lake Charles (Nova Scotia) each counting once as in the standard edit count. Of course, not everyone skimming a huge article will reach the typo I fixed on line 1234. The score probably needs to be something like sum of (pageviews × Levenshtein distance) of unreverted edits, which is getting complicated and computationally expensive – but this is an idea lab! Certes (talk) 11:18, 15 September 2022 (UTC)

I actually like this idea... It needs to be weighted though, so for example, adding/removing these [[]] should be weighted less than fixing typo for instance. There's a small dataset that you can toy your algorithm with at WP:30 kB drive; you just need to compare the old and new version, churn out a score and sort the articles. If the metric makes sense, so be it. CactiStaccingCrane (talk) 14:03, 15 September 2022 (UTC)

P.S., I think we can make do with a very crude metric for the Levenshtein distance using diffs, which it equals to the number of chars highlighted. CactiStaccingCrane (talk) 14:06, 15 September 2022 (UTC)

Quarry:query/67384 attempts to measure contributions to level 1 and 2 vital articles. For the simplest types of reversion, I've excluded both edits, but some of the contributions may be multi-edit vandalism or reversion of multi-edit blanking. I may also be ignoring good contributions which were reverted either in error or maliciously. I've arbitrarily assigned level 1 articles twice the weight of level 2. We could probably add level 3, but checking all 50,000 vital articles might set the servers alight. Certes (talk) 00:44, 16 September 2022 (UTC)

Weighting edits by their impact is certainly a good idea. I think there are 4 things Wikipedia should weight:

• edits to popular articles as opposed to unpopular articles, pure pageview metric (their Impact module seems to fix this)
• edits to vital articles, as a separate and additional weight factor to the one above. Popular but non-vital pages should get weighted less, popular and vital, more. Special attention should be dedicated to this, as we certainly don't want to disincentivize people editing currently-trending articles like the Elizabeth II page; but we also want to strongly incentivize people to work on lower-views, but vital articles; and to incentivize vital articles over equally-popular non-vital ones.
• edits that don't get reverted (weighted by how long they've lasted) versus edits that got reverted; Levenshtein addresses this
• edits that contribute content, as opposed to style/grammar/MoS edits (as you're proposing, and I agree). "% of mainspace, non-talk page article written by [USER]" weighted by that article's pageviews. XTools already has authorship percentages, so it's clearly possible through the Wikipedia API. That way, not only would unreverted edits be weighted more, people would also be incentivized to contribute lots of content rather than to make minor fixes. This should be combined with the Levenshtein metric since we don't want to incentivize fluff or non-encyclopedic essays; so the metrics shouldn't just be added. For example Page-popularity-metric vital articles metric Levenshtein Levenshtein * (content contribution * pageviews).

It would be best if this were combined into a single metric; and maybe if they were also each placed in a single place on a dashboard to "break down" the single metric (highlighting vital articles contributed to, etc.) The community could discuss the different weights to give to these elements.

When it comes to changing culture, I'm hopeful that this will have an impact, but there's obviously always inertia when trying to change incentives. I'm not sure how to ensure that people don't keep valuing raw edit count as the "main" value to increment, as opposed to a weighted alternative count. A good start would probably be to remove "edit count" from Preferences and replace it with the new metric; and to shift the autoconfirmed, extended-confirmed, etc thresholds to a corresponding value in the weighted metric. A big thing to do would also be to create something like WP:WBE, for the "highest achievers/most competitive", so they can compete on that more complex/nuanced metric in a straightforward, easy way. Creating a central acknowledgement would increase the incentive.

Also I'm not aware of any userpage templates that update dynamically (for example, and edit count userbox that updates itself); would be nice to create one for the weighted metric that autoupdates, that people can add to their page. DFlhb (talk) 16:22, 15 September 2022 (UTC)

Do you want to target popular articles or vital articles? User:HostBot/Top 1000 report lists this week's 2204 most-viewed articles (those in the top 1000 on at least one day). That includes 1/10 level 1 vital articles (Earth), 4/100 level 2 and 69/1000 level 3 articles. So there's not much overlap between what readers choose to read and the topics marked as vital. Certes (talk) 10:20, 16 September 2022 (UTC)

Incentivize content edits to vital articles; incentivize correctness, proper sourcing and neutrality for popular articles. I mention 4 factors; but you can play around with weights on each factor, or have some factors be absolute numbers (can't drag the overall metric down, just bring it up). Those are implementation details. DFlhb (talk) 10:50, 16 September 2022 (UTC)

i am amazed by GPT-3. I am currently working on voter database related pages. Here is free text to table (via table convertor)

"In the 2004 presidential election in the United States, the Republican Party used the Voter Vault platform and the Democratic Party used DataMart. Currently, the Republicans use rVotes Data Center and the Democrats use Votebuilder from the Voter Activation Network (VAN). There are non-partisan firms that offer registered voter data in the United States, too: NationBuilder, Aristotle, eMerges and Labels and Lists. In 2015, a database of 191 million U.S. voters was exposed on the internet and included names, addresses, birth dates, party affiliations, phone numbers and emails of voters in all 50 U.S. states and Washington.

Date	Platform	Election	Political Party
------	------------	---------	-----------------
2004	Voter Vault	U.S. Presidential Election	Republican Party
2004	DataMart	U.S. Presidential Election	Democratic Party
2015	rVotes Data Center	U.S. Presidential Election	Republican Party
2015	Votebuilder	U.S. Presidential Election	Democratic Party

Wakelamp d[@-@]b (talk) 07:36, 2 October 2022 (UTC)

I enjoy the sarcasm (it's funny) but do note that GPT-3 was simply never made for that purpose, only for "human-like language". AI researchers have been working on verifier/truth modules for years. DFlhb (talk) 08:27, 2 October 2022 (UTC)

For once, i actually wasn't being ironic, It actually works surprisingly well. Wakelamp d[@-@]b (talk) 10:37, 2 October 2022 (UTC)

Huh; I wouldn't have thought "rVotes Data Center" would be a real name. The 2015 bit though was genuinely impressive indeed. DFlhb (talk) 10:58, 2 October 2022 (UTC)

The 2015 bit stunned me as well. With the names of the voting database companies, I agree the names are absolute rubbish or downright misleading, and the ownership structures are totally crazy, (which I am enjoying finding references for :-)), but I unfortunately now think that Cambridge Analytica may be really nice in comparison. :-(. Wakelamp d[@-@]b (talk) 13:01, 3 October 2022 (UTC)

@DFlhb: FYI, your link to the presentation is broken. This works: File:Wikipedia’s poor treatment of its most important articles.pdf. Qzekrom (she/her • talk) 05:05, 15 September 2022 (UTC)

Thanks DFlhb (talk) 14:50, 15 September 2022 (UTC)

I think editing existing articles can be intimidating, especially if they already seem pretty comprehensive and well put together. It takes a lot to make substantial edits to an article: you need to go dig into reliable sources, either ones that are already used in the article or new ones. This requires having access to the sources and the time and skill to pore through them. Then, you need to add missing information to the article. This requires finding a place to insert the information, paraphrasing the source without plagiarizing it, and making the inserted content flow with the rest of the article. There's a lot to get right, and if you get it wrong, you run the risk of your edits getting reverted (although it's probably more likely that other editors will either improve upon your work or do nothing for fear of "messing it up"). I think we need to make users more comfortable making substantial edits so they don't all become WikiGnomes like me.

I think it could be useful for WikiProjects to conduct scrum-style "sprint planning" meetings, in order to prioritize tasks and allocate them among project participants. For example, one person could specialize in adding information from sources, another could focus on organizing content, and another could focus on polishing articles. Specialization requires accountability (or redundancy), but it would allow us to work together more efficiently. If we're not comfortable with paying people to edit, WMF could hire scrum leaders to help WikiProjects coordinate. Qzekrom (she/her • talk) 04:33, 16 September 2022 (UTC)

Practically speaking, the WP:PVITAL WikiProject is already doing that. But we need more people. It is mentally tolling for me to see almost no activity on the WikiProject. CactiStaccingCrane (talk) 06:58, 16 September 2022 (UTC)

The absolute last thing Wikipedia needs is to hire more non-contributors, especially bureaucratic nightmares like Agile leaders being paid to waste unpaid volunteers' time in Zoom meetings. Did you know Wikimedia Foundation spent $70 million on its own salaries in 2020, $10m in grants, and $2m on internet hosting? That these salaries constitute 60% of the Foundation's expenses, and that by policy none of these people are allowed to contribute to Wikipedia anyway? (struck through 20 September 2022; I was imprecise, see below) Where's the evidence of impact for anything these people do? It's mindblowing that a "charity" is funneling money to "consultancies" for "strategic planning" (see: williamsworks) or "brand strategy" (Snohetta Design). Keep the devs and server farms, hire 100 academics to contribute content while staying away from controversial BLPs to avoid Section 230, and fire everyone else. The WMF people are making the same mistake Evernote made, wasting money while the core product (which is Wikipedia's written content) languishes. I'd like to remind them that once someone invents a more advanced GPT-3-type AI to write an encyclopedia, there is strictly nothing obliging them from giving that output to Wikipedia, or producing it under an open license. I wonder what an endowment would be useful for in that scenario (again, remember Evernote, the "100-year company" that may well die in the next recession?). DFlhb (talk) 08:07, 16 September 2022 (UTC)

But, you kinda need to shot on the WMF arms in order for them to do work though... I personally see that nothing will change unless there's a significant push to do so, i.e. competing encyclopedias or AI stuff. CactiStaccingCrane (talk) 09:53, 16 September 2022 (UTC)

DFlhb, what makes you think that by policy none of these people are allowed to contribute to Wikipedia? Whatamidoing (WMF) (talk) 16:41, 19 September 2022 (UTC)

I got that from the EFF [1]: In fact, the Wikimedia Foundation does not write or edit any of the content found on the Projects

From this Quora comment from Mark Hetherington: [2] The WMF has about 290 employees, and will probably grow to about 300 in 2017–18. These employees support Wikipedia indirectly. They don't create or edit content, because not doing so helps the WMF maintain its legal immunity as a “service provider” under Section 230 of the Communications Decency Act

And from Wikipedia:Wikimedia_Foundation that says: The WMF does not edit Wikipedia content (except for occasional office actions). "The community" (largely volunteer editors) handle content, because if the WMF did take responsibility for content, it would introduce liability issues per section 230 of the Communications Decency Act..

I'm assuming you're saying this is false? DFlhb (talk) 20:32, 19 September 2022 (UTC)

I'd say that it's incomplete, rather than completely wrong. As part of my job, I don't write articles, or otherwise create copyrightable content that isn't related to my job. This means that they don't want me writing whole articles for work but, e.g., that I can edit Wikipedia articles if I'm cleaning up a wikitext problem caused by a software bug. I personally tend to do that from my volunteer account, so that nobody's confused about whether it's okay to revert me, but the WMF has no concerns about me doing that from my work account. As an example, these edits comply with WMF policy.

But WMF staff are also welcome as volunteers, and volunteer-me has made more than a hundred thousand edits at this wiki. Dozens of WMF staff have made thousands of edits as volunteers, and many are admins at this and other WMF-hosted wikis. It's good for WMF staff to edit from a volunteer account on their own time (or during a designated training program; everyone wants new staff to learn the basics if they weren't hired out of one of the communities). Whatamidoing (WMF) (talk) 23:02, 19 September 2022 (UTC)

I see, I should have recognized that those statements were a little lawyer-y and that "WHM doesn't edit" meant something quite specific. It's good for WMF staff to edit from a volunteer account on their own time Absolutely agree, and it builds ties with the community. I'll strike through my comment above so no one gets misled.

I also hope you can relay my suggestion to the rest of WMF to hire some (ideally domain-expert) editors to contribute to making more Wikipedia:WikiProject_Vital_Articles Featured-quality, and I hope they consider the idea. I'm sure some would complain that it would unfairly advantage the English wikipedia, but a) there's no reason national affiliates can't do so too and b) that won't be a problem very shortly[3] DFlhb (talk) 02:14, 20 September 2022 (UTC)

AIUI the WMF never hires people to create article content, and the rules about getting grant money for content creation seem to have some limitations. I'm not involved in the grants process, so I am not a reliable source ;-) but my impression is that they are much more likely to fund an event during which volunteers create content than to pay directly for content creation. You could request a grant, if your group wanted to do something along those lines. Maybe invite an expert or two and a reference librarian to meet a handful of Wikipedia editors in a university library to work on a pre-determined set of vital articles? Whatamidoing (WMF) (talk) 23:12, 21 September 2022 (UTC)

There are many flaws to the idea of throwing together an event for domain experts to volunteer. AIUI, the grant money can't be used to actually pay contributors, just to pay for the costs of managine and hosting events. Even if it could, this effort would be far more effective if centralized.

A major issue with Wikipedia is domain experts feeling overwhelmed at the poor quality of major articles, and just quitting. These are exactly the kinds of contributors that Wikipedia needs, and they may be less incentivized to contribute than anyone else. I feel like such efforts may be asking far too much of overwhelmed volunteers who are for the most part already overworked at their actual jobs. Hiring a few of them full-time and paying them is the only good solution I can think of; I don't believe it would detract from Wikipedia's volunteer ethos; if anything, it has a good chance of increasing academics' faith in Wikipedia and reigniting their interest in volunterring their edits. I'd also expect such an initiative to receive positive coverage in mainstream press.

Asking you to relay my suggestion to WMF might have been a reach (it would be perfectly appropriate for you to only do so if you actually agree with the idea, frankly) so I'd like to ask instead if this is the right place to discuss WMF-related suggestions like that (which strays quite a bit from the overall Vital Articles discussion); is there another place I can ask this that would get the WMF's attention? Best, DFlhb (talk) 18:41, 22 September 2022 (UTC)

Are you familiar with the m:Wikimedian in residence idea? Other organizations hire Wikipedia editors. It appears to mostly be museums or cultural organizations, though volunteer-me is more likely to see John P. Sadowski (NIOSH) and others who work for public health agencies, since I'm more active in the health-related articles than art-related ones.

(I am happy to pass along ideas that have a chance of being useful or interesting, even if I don't agree with them.) Whatamidoing (WMF) (talk) 00:46, 23 September 2022 (UTC)

Interesting, glad you linked me to that. It unfortunately doesn't seem to include very many editors, though it is a step in the right direction. I expect WMF may be somewhat averse to hiring editors themselves, mainly due to fear of community backlash; but frankly I'd expect the community to be relieved, based on the commonly reasons behind profilic editors' departure (burnout, overwhelm at poor quality of important articles, etc). Maybe making these hires subject to community recall would ease the WMF's concerns re: a potential community backlash. DFlhb (talk) 07:50, 23 September 2022 (UTC)

The movement has traditionally separated its organizational side from its editing side. For example, m:Affiliates can host events that encourage people to edit, but they can't make "official" edits.

I could be wrong, of course, but I have the impression that a separation suits the WMF's Legal team at least as much as it suits the editors here. Whatamidoing (WMF) (talk) 21:22, 28 September 2022 (UTC)

I don't think sprints are a good model to follow for the type of volunteer writing work done by the English Wikipedia community. I appreciate, though, that some editors are motivated by seeing continual progress towards some goal. Perhaps tools to automate updating pages such as User:Ravenswing/Hockey Mountain would be helpful in encouraging ongoing improvements. isaacl (talk) 15:52, 16 September 2022 (UTC)

How do people work together on a Vital Article? Also are inexperienced editors useful? That is they know how to write/cite/discuss/npov etc), (I am still going through communication links in WP) , What do you do if the article is likely to be high conflict? Wakelamp d[@-@]b (talk) 14:26, 15 October 2022 (UTC)

Iterative improvement and consensus, just like Wikipedia's olden days. It seems that we've lost that spirit years ago when nowadays, quality content is usually written by one person only. CactiStaccingCrane (talk) 03:11, 16 October 2022 (UTC)

I personally think that people are as important as articles for WP

So, I have been researching options for the last few weeks, and these are some thoughts

• Too long : Can we make VT article creation process more efficient? Maybe, but not quickly
• Hard ; Incentivising experienced editors on pump
• Tricky : Editathons. create few long term editors.

So what do you think of any of these options :

• Recruit Editors externally directly.
• Signpost for all :-)
• Recruiting Template on the VT articles, but link it to the VT project page, Ask them to sign up for a newsletter, and saying if they are working on a VT article, and have a revert, or an issue, or a question then ask on VT project, not the page.

Wakelamp d[@-@]b (talk) 13:10, 19 October 2022 (UTC)

All, I'm greedy :) In my honest opinion though, I think that the project should do a Signpost issue, which basically said "The project do this and that, this is why the project is important, and you can join us too by improving X (whatever that the project is working at the time)". Recruiting templates does not leave a lasting impact, so do recruiting editors externally. It's a really tricky thing. CactiStaccingCrane (talk) 13:27, 19 October 2022 (UTC)

Greedy is good :-) (Uncle Scrooge is still my hero). I am uncertain about Signpost as ot has only 500 (?) subscribers, and the rule of thumb is 1/1000 to 1/100 response is good (that's in my head and I am not sure why).

The suggestions about supporting new editors is based on this very interesting [of WMF research from 2017] which has pain points involving being lost, new article frustration - My thought was that if we pointed them towards a purpose and a community it might provide Beagle Boys for filling your Money Bin of Vital articles.

So we could add a Template to ten articles - "Have you always wanted to edit Wikipedia? Believe in being neutral and fair, like writing, finding references, and understand that it is a collaboration? Then We have 10 articles that we think are vital, and we will work as a group to fix them. Join @Sensei Cactis Legion today :-) Wakelamp d[@-@]b (talk) 16:16, 19 October 2022 (UTC)

The recruiting template could take the form of a Vital Article topicon, similar to the GA-class topicon; have the level be indicated on the icon, so it's easy to see if it's vital-1 or vital-5; click it, and you go to the WikiProject. People should feel that they wouldn't be the only ones contributing: "hey, I'm a professor in this; no time to edit much, but feel free to take a look at these 4 books that I feel represent the field well; I'm willing to do an expert review of the article once you guys are done with it." Even if that only played out a few times a year, it would be a vast improvement over the current state. DFlhb (talk) 17:41, 19 October 2022 (UTC)

Regarding the ideas above regarding motivation via pageviews/impact. Don't we kind of already do that? That form of motivation already exists, but it doesn't push people towards, say, The arts, it pushes them towards current events, sports, movies, video games, youtubers, internet memes, actors, musicians, etc.

Regarding high-level vital articles in general, while an article like food or the arts may get a lot of pageviews, I have a sneaking suspicion that people aren't going to the article with the intent to read those articles. I suspect that more often they're looking for something more specific that the higher-level article brings them to, or perhaps even an outline of articles about that subject. It seems to me that something which might benefit readers and editors would be to try pushing for people to work hard on outlines of big topics (on a talk page), with the idea that implementing it will ensure readers get what they want while at the same time providing structure for editors to fill in without overhauling the whole article. YMMV. — Rhododendrites ^talk \\ 12:29, 16 September 2022 (UTC)

Personally, I think that a good article would also serve as a good outline of the topic. Sure, I'm all for making a comprehensive outline for further reading purposes, but outlines should not be an excuse for poor content. CactiStaccingCrane (talk) 13:56, 16 September 2022 (UTC)

If I understood the suggestion correctly, it was to make an outline of how the article should be structured, rather than an outline of the article's subject. It would help interested editors co-ordinate their work on the article. isaacl (talk) 15:36, 16 September 2022 (UTC)

Right. So if one of the primary functions of high-level articles is as an outline of the topic, it shouldn't take the form of an outline but a good first step might be to come up with an overall outline. That way the outline function is satisfied and those who want to contribute have a structure to address parts of. A big issue with the big articles is that they're often clumsily organized, with big parts missing and others lumped into an overly broad header (for example)... — Rhododendrites ^talk \\ 19:03, 16 September 2022 (UTC)

I have four points here:

1. I am strongly opposed to the idea of using pageviews as any kind of metric of worth. It works against improving vital articles, not for it. For example, the article for Gigi Hadid has almost 20 times the pageviews than the article for Hippocrates. The article for the pornographic act of creampies has the same amount or more pageviews than the article for diabetes. Chasing pageviews was an abysmal idea for the media, and we do not need to adopt their abysmal idea.

1a. Using GPT-3 or any form of AI for this is an even worse idea; it should be self-evident why. A lot of machine learning is trained in part on Wikipedia anyway, as the content is free.

2. The problem has to do in part the fact that a lot of these things are very abstract, high-level concepts. Take "the arts," for instance. People have tried for millennia to define what "the arts" are and its scope; it's a known difficult problem. Given that many generations of the greatest thinkers the world have produced have, thus far, failed to settle the question, I don't think individual Wikipedia volunteers, or even Wikipedia volunteers en masse, can be faulted for also doing so. By contrast, an actor (for instance) has a narrow existence and a clear biographical framework. There is also the issue of sourcing. The larger the topic, the more difficult it is, paradoxically, to source, because coverage is likely to be piecemeal or flooded by hundreds of thousands of possibilities.

3. Given my name it's probably unsurprising what my stance is on this, but I do not think "minor changes, grammar, etc." are invalid edits. The average person reading an article is much more likely to see bad writing, shoddy formatting, and the like as an indicator of poor quality than an article being a short but clean stub or start-class. I've even found multiple cases of blatant, juvenile, easily findable vandalism existing in vital articles, and persisting through large content additions/overhauls; if we truly had such an imbalance then I wouldn't keep finding shit that's lasted 10 years ago in high-profile pages.

4. Finally, at the risk of upsetting people, if we are going to talk about wasted energy by active editors, then we should talk about the supposed "crisis" some people have made of inobtrusive stub articles that already exist -- i.e., are no longer wasting anyone's time, if they ever did -- which sucks up seemingly bottomless time and energy by active editors that could go toward actually improving content. Either these articles matter or they don't. Gnomingstuff (talk) 16:17, 23 September 2022 (UTC)

I don't agree with the first point. Obviously in looking at pageviews you have to compare like with like, and not worry about popular culture, sex and geography being waaay high. But page view comparisons between articles one might expect to have similar figures often reveal very different numbers, and are useful for directing editing effort. As someone who has been editing The arts since 2008 (I see, without apparently adding much) I'm fairly happy with the article being a rather simplistic listy mess that gets readers to more specific pages reasonably efficiently. "What is art?"-type debates belong at Art, Visual art and other pages. Johnbod (talk) 17:02, 23 September 2022 (UTC)

Note on 4 I forgot -- It also disincentivizes people to work on vital articles by creating an artificial sense of urgency elsewhere. As much as Wikipedia touts WP:DEADLINE, this is not evenly applied. The existence of an article like the arts is not up for debate, and its content is generally "good enough" to not seem to need immediate attention. On the other hand, if a possibly notable subject (or unquestionably for that matter; there was recently a Jorge Luis Borges book up for deletion!) has its article up for deletion, the research, which is non-trivial (and routinely isn't done well), then needs to be done as soon as possible, within a week and ideally early in it. That research may be wasted/demoralizing effort if it turns up nothing or if the article is deleted anyway. It incentivizes only partial research and improvement -- usually solely online and English-language sources, exacerbating recentism and Western-centric bias -- and the process attracts generalists: it's impossible to specialize in the hundreds of topics that come up, subject-matter experts for those hundreds cannot realistically be expected to show up on a procedural project page within a week, every week, and people have lives and schedules and can't drop everything to access library archives on short notice. As another example, I recently started sourcing articles from the Unreferenced Articles wikiproject, which have been in that state for over a decade and are often quite expandable -- but the only reason I did that was the recent proposal to nuke the whole thing. Gnomingstuff (talk) 17:08, 23 September 2022 (UTC)

• Thanks for your thoughtful reply; I'll try to address your points methodically.

1. Fair; I think my "metric" idea should overwhelmingly prioritize Vital articles over popular ones. By definition, vital articles are vital. They are of highest importance to the encyclopedia audience, from schoolchildren to the elderly, regardless of pageviews. Other types of contributions are already implicitly incentivized without needing to be incentivized further. People are already editing popular articles (people want their edits to be seen; and further; further, the higher the pageviews, the more editors the article will have, since some of these viewers will go on to edit). My idea of having a single metric that doesn't just reward Vital articles, but also copyediting, popular article editing, etc stems from a desire to have that metric become the metric, i.e. a metric that can completely replace raw "edit count" when it comes to gatekeeping Wikipedia:User access levels, or when it comes to users judging their own contributions and trying to gamify their Wikipedia experience, like trying to compete for highest edit count. But this new metric should absolutely incentivize Vital articles explicitly, since, again, people are already creating articles on each random high school or road without needing to be incentivized, and no metric will change that.

1a. I would never suggest unleashing GPT-3 onto Wikipedia. But it's clear that, no matter how long it takes (5 years? 20 years?), one day, AI will be able to read every book and study in existence, synthesize them, and make high-quality encyclopedia articles about them. I was merely suggesting that the Wikimedia Foundation either fund R&D into such AI themselves (likely too early), or at least collaborate with private organizations who have the funds to do so. Meta Platforms has already created an AI for Wikipedia that can check if a citation truly backs up a claim[4]. I think this stuff is coming faster than people think, and it would be good to make sure these AI get built with input from the community and WMF rather than get "thrust" upon Wikipedia.

2. Vital articles may be more difficult to edit, sure; I think that strengthens my point that they should be strongly incentivized. Harder to source (and scope) doesn't mean impossible; and with more eyeballs, these articles could truly thrive.

3. Also fair. But wouldn't you say that this constitutes content edits/copyediting, as opposed to minor formatting changes, fixing link-pipes, and script-aided MoS compliance? I think my idea would actually help address what you point out. Taking the Technology article as an example, there were dreadful passages that remained for years or even a decade because most edits barely ever touched the actual content. And to others reading this who might be concerned: people will still use scripts; the ease of making mass edits is already an implicit incentive. But there's "friction" to making substantive copyedits or content contributions, which is an implicit disincentive; and I think that should be addressed.

4. Definitely. I think it's a problem when stub or niche articles get deleted along with their contents; if they remained, many of them might eventually get merged into a more major article, or at least help editors with content contributions to other articles. A good recent example is the Singularity (system theory) article; basic but decent enough, and quite a bit of effort went into it. Deleting it, instead of merging or expanding, just means duplicating efforts later on when having to recreate some of that content from scratch in other articles. AfDs should have a somewhat high bar unless the topic is clearly not notable, copyvio, or an advertisement, to avoid wasting editors' time. DFlhb (talk) 20:48, 23 September 2022 (UTC)

Translators' noticeboard

I've been thinking about good-faith editing from non-native speakers who have trouble because of language barriers. Sometimes the edits are good enough, sometimes they get removed, and sometimes they're disruptive (up to and including ANI cases) even with translation algorithms as good as they are now. It's also not easy to call attention to cases where translation help is needed. Apparently, the local embassy was supposed to handle these issues, but it got removed from the main page because it was considered dead; that doesn't seem to have helped. With that in mind, I think we should try to develop a translators' noticeboard proposal. As well as calling attention to specific pages where translators can help, non-native speakers could be directed there when it's needed. There are a lot of Babel boxes on user pages, so it should be worth having a central discussion page for this. RAN1 (talk) 14:49, 19 October 2022 (UTC)

I just found out Template:Noticeboards has a "Translation" link that goes to Wikipedia:Pages needing translation into English, which has a backlog of just under 300 articles, so this idea can wait until that's done. RAN1 (talk) 03:20, 20 October 2022 (UTC)

PNT isn't working well, and hasn't been working well for more than a decade now. Don't let its existence stop you from thinking about better ways to contact multilingual editors. —Kusma (talk) 05:59, 20 October 2022 (UTC)

Wikipedia should be redesigned into a much more user-friendly and dynamic site...

Currently, Wikipedia is a plain static site that can be expanded in a pre-determined format. It would be more helpful to the reader, if Wikipedia is redesigned in such a way that it can attract more people to read and contribute.

1. Complex articles, that have lots of technical terms, should be simplified much further (I'm well aware of the Simple Wikipedia project, the problem is that it does not cover all information in all languages). We must also remember that a vast majority of users are students who might not know the terms but have to cover upon the specified section. (For eg:- The Human Brain)
2. A small summary of the article should be provided at the top of each (Proposed title: In a nutshell...), which should be more descriptive than the one-line summary but not bigger than 4 paragraphs. It can be in a card that is just a curved rectangle which stands out from the rest of the page.
3. I love the MinervaNeue interface. If it's functionality is expanded, then it could prove to be a better interface than the Vector (2022). Don't take me wrong, but not everyone likes detailed views.

Please feel free to voice your opinions below! Sugeeth Jayaraj (talk) 17:38, 11 September 2022 (UTC)

For 1, there's 6 millions articles out here. Feel free to start.

For 2, that's what the lead section already does.

For 3, what skin is best is subjective. That's why we have preferences.

Headbomb {t · c · p · b} 19:00, 11 September 2022 (UTC)

Well, according to the Wikipedia:Make technical articles understandable guidance introduction articles could solve the problem with complex topics. So probably Introduction to the human brain could solve the problem. D6194c-1cc (talk) 19:53, 11 September 2022 (UTC)

I agree. Too many math articles on Wikipedia are just making me crazy by overly-difficult explanations. CactiStaccingCrane (talk) 10:54, 13 September 2022 (UTC)

If anyone is interested, a WikiProject:Introductory articles (doesn't exist yet) would be very beneficial given how much Wikipedia is used in school and to discover new topics. I don't have the time for that though, but that addresses point 1. DFlhb (talk) 09:23, 14 September 2022 (UTC)

Also an "Introduction" section might be a good idea in complex articles. Such a section may lead to the main introductory article. D6194c-1cc (talk) 15:15, 14 September 2022 (UTC)

CactiStaccingCrane (talk) - It's not just math articles, but also many, many medical, science, and engineering-related articles require a high degree of understanding before even approaching them. When someone doesn't know the jargon flashed around, they can quickly go down a deep and frustrating rabbit hole trying to understand link after link until they leave in frustration.

So far, I have not seen a technical article linked to a "Simple Wikipedia" version. Instead of a Shadow-pedia, perhaps articles could be written in easier to understand language in the beginning whilst about halfway through or so, the technical stuff comes out for the adventurous?

Thank you, Wordreader (talk) 06:43, 8 October 2022 (UTC)

Sugeeth Jayaraj, I've already created such a discussion in the past here and I doubt I've been the first. I hold a lot of wikisessions on real life and this was a request brought up by a lot of students we interacted with which had "left Wikipedia for Britannica" because of these "problems", which Britannica didn't have because of the many ways it allowed to manipulate information parsing for students. Unfortunately nothing was done about it in the end because such proposals usually get automatically piped to "you can create a new WMF Wikipedia if you want" and good luck with that. — Klein Muçi (talk) 11:50, 13 September 2022 (UTC)

I’m not looking for a separate Wiki… The thing is that Wikipedia’s main aim is to make information accessible for everyone. This is just an expansion on that…

Since it’s inception, Wikipedia has been very busy creating new articles. So, it is natural that not all articles can be quality-checked and simplified.

Most contributors, like Indians (just an example), are non-native English speakers, including me. We are ‘taught’ English rather than it being an natural inclusion in our lives. (We usually speak our native knowledge for example)

No other Wikipedia, other than the English version, has all the info we need. Every other version is usually very basic / stripped down.

If we can somehow raise this issue to the core team at Wikipedia, then maybe we can achieve our request.

Millions of people will be saving atleast 5 minutes to an hour of research, hunting for in-depth knowledge. Sugeeth Jayaraj (talk) 16:20, 13 September 2022 (UTC)

Yes, it's exactly the same request as mine for the exact same reasons only changing India with Albania. The problem I believe is that generally speaking Wikipedia is mostly concerned with how content is written, not consumed. If it is formatted properly or if it is citations are correct, for example. For as long as the content is correct, not much more is done for articles. We generally "don't care" about how that content is consumed, by who is consumed and why and we certainly don't provide any facility for that apart from the occasional accessibility for people with visual impairments (spoken articles). I, same as you, think we should be able to provide the same information styled differently for different types of "consumers", targeting their needs. But not many people think the same apparently.- Klein Muçi (talk) 16:47, 13 September 2022 (UTC)

@Sugeeth Jayaraj "Every other version is usually very basic / stripped down." Would it help to have the ability to have a default preference order of language language x -> simple en (or simple en auto translation)-> auto-translate en article -> english? (The N language bar dropdown seems more advertisement than useful.) Would the ability to click on any word to get a wiktionary, or to translate that one word help? (I would also like it to auto-translate WP short codes)

@Wordreader I too would like more linking to Simple, wiktionary For some of the technical articles, there are various papers in google scholar by medicos about whether information is misleading or comprehensible for patients. Wakelamp d[@-@]b (talk) 16:24, 11 October 2022 (UTC)

There should definitely be a section on the front page, like with featured articles or image of the day, that promotes introductory articles to complex ideas. Since so many people like to go down the rabbit hole and learn new things from Wikipedia, it would encourage people to participate in writing those articles, and might even influence some young people’s career paths. The fact that there aren’t many Intro articles currently would get fixed over time by increasing attention to this Wikipedia initiative, would cater to a younger audience that might otherwise leave Wikipedia, and would reallyhelp underserved countries where the education might not be the best. Maybe a partnership with Khan Academy wouldn’t be a bad idea either. DFlhb (talk) 15:55, 14 September 2022 (UTC)

• I think Headbomb sums it up best with his answer to #1. Kudpung กุดผึ้ง (talk) 12:29, 13 September 2022 (UTC)
• If the article is too difficult, maybe there is a reason for this? Not everything can be explained in easy terms (ELI5), but simplicity would kill all the meaning that will be perfectly understandable for an appropriate audience. (I wouldn't understand an article about chemistry/abstract math/etc, but my unpreparedness is mot the reason to cut the article and explain it shallowly). Artem.G (talk) 19:47, 13 September 2022 (UTC)

  Artem.G (talk) - I whole-heartedly disagree. WP is not a textbook, but many articles act like they are and set themselves above the average user when they should be helping as many users as possible to understand the topic, in a cursory way, at least. WP frequently solicits donations. Wouldn't somebody who comes to an article and leaves with at least some understanding be more likely to contribute than somebody who quickly finds themselves up to their neck in complex jargon and leaves in frustration? Any topic can be broken down into bite-sizes. It's always exciting watching the light come on in a learner's eyes rather than to watch that light go out. WP just needs to want to do that. Thank you, Wordreader (talk) 06:58, 8 October 2022 (UTC)

With regards to point 1. It is true that some articles' lead sections (and sometimes the whole article) may benefit from simplification and pruning in general, but this simplification should not be at the expense of removing technical information that will be useful to experts within that particular field. Often, there is simply no way to compress an article any further without losing crucial technical precision. Additionally, many extremely technical articles (for example, articles dealing with genes, specific organic/inorganic molecues etc) are almost exclusively accessed by readers who have at least some expertise in their relevant field, so there is little need to simplify the article for the general public. Rob3512 (Talk) 07:02, 14 September 2022 (UTC)

We already have a guideline for point 1: WP:TECHNICAL. CMD (talk) 11:21, 14 September 2022 (UTC)

• Two statements above by Sugeeth Jayaraj: "a vast majority of users are students" and "Most contributors, like Indians (just an example), are non-native English speakers..". I'm pretty sure neither of these is true overall, although in both cases it will be true for some articles. Johnbod (talk) 12:57, 15 September 2022 (UTC)

  Yes, I did mean for some articles, not all of them! Thanks for pointing me out! Sugeeth Jayaraj (talk) 07:49, 17 September 2022 (UTC)

  I suspect that there are more student-readers in India than any place else in the world. Some reader-focused research a few years ago showed that in most of the world, people wanted shorter articles with more pictures. In India, far more than anywhere else, readers wanted long, detailed articles. I suspect that many Indian readers were using Wikipedia as an opportunity for in-depth learning. Whatamidoing (WMF) (talk) 17:26, 22 September 2022 (UTC)

• I love this idea. I really wish Wikipedia could know what I'm interested in, based on reading history and edit history, and better help me navigate areas I'm interested in. I think there's a lot of social good in keeping people in the Wiki longer, IMHO it's better than getting sucked in by the Instagram or Facebook algorithm. I really agree that the Wiki is too static, and represents a one-size-fits-all approach which is rather too rigid compared to it's contemporaries. I don't think it needs to look and function like a social media, but I really wish it at least partially represented 'me', when I am logged in.

On the edit front, we could make it much easier with dynamic tutorials, or tools to point out really easy spots in the articles that could use some help. Using a native language ML we could point out to users what sections are too difficult to read, easily present what other closely related articles look like, and suggest edits that could be made. Right now, the barrier to entry here is rather high, with all kinds of disparate tools, plugins, scripts, etc. We could do a much better job of embedding tooling directly into the platform to make the learning curve much better. I'm thinking like a 'help out' button of sorts on an article, that can provide a list of ways a non-seasoned editor can dive in and assist. Think "These top 5 paragraphs have no references, heres an inline tutorial on how you can add a reference", "There's no short description, click here to add one", "This area has a poor readability score, can you help reword it?", "These are the potential spelling errors"...Mr.weedle (talk) 01:18, 18 September 2022 (UTC)

Yup, that's a great idea. Do you think that I should proceed to request on the main village pump? Sugeeth Jayaraj (talk) 12:43, 21 September 2022 (UTC)

The whole point of Wikipedia is to not collect user data. I don't think that is going away anytime soon. Sungodtemple (talk) 00:37, 22 September 2022 (UTC)

We can try to implement this in a user friendly way without data collection. Sugeeth Jayaraj (talk) 12:49, 24 September 2022 (UTC)

User path data is already kept on article and talk, but is anonymised. meta:LZia_(WMF)

Which projects/areas are the main concerns/? I have cross posted this on WIki-project Maths. "Suggest people comment. I think the points are well made.I like the way this site handles it (Note its called skeptical science because it's skeptical about anti-climate change science - confuses everyone), They have basic, advanced, and intermediate levels as tabs. [[User:Wakelamp|Wakelamp d[@-@]b]] ([[User talk:Wakelamp|talk]]) 04:37, 18 October 2022 (UTC) :Could you pick an article and show us a rewrite of the lede.( i would prefer in maths as it is cross posed?) [[User:Wakelamp|Wakelamp d[@-@]b]] ([[User talk:Wakelamp|talk]]) 00:05, 21 October 2022 (UTC) == Organization for AfD discussions == I don't know if I will formally propose this, but: When an AfD discussion becomes very long, it's hard for any administrator to read all of the !votes in a row. It may be a bit easier if when ten or more people !vote on AfD, Wikipedians place their !vote under a header, like in RfA. For example, this is ten votes or under: *'''Keep''' Notable enough *'''Keep''' Passes [[Wikipedia:Notability|WP:GNG]] *'''Delete''' Does not have many sources come up in a web search Ten votes or higher is: ===Keep=== *'''Keep''' I've heard of this *'''Keep''' Plenty of reliable sources *'''Keep''' Passes [[Wikipedia:Notability|WP:GNG]] *'''Keep''' We've got entire articles from reliable resources on this subject, all sources fine, C-class, it's good. *'''Speedy Keep''' Per above, why are we deleting this in the first place? It's got reliable sources, and my web search shows a lot of results on this topic, all of the sources reliable. We've got articles on this in different languages, we're definitely fine. ===Delete=== *'''Delete''' Facebook is cited as a source. We can probably change it, but the users above say all of the sources are valid, and this is not. I checked another one of the sources, and it also is unreliable. *'''Soft Delete''' It doesn't look reliable, but maybe the topic will become more reliable later on. *'''Strong Delete''' Fails [[Wikipedia:Notability|WP:GNG]]. That and unreliable sources. *'''Speedy Delete''' per [[Wikipedia:A7|A7]]. No evidence of notability, just bare URLs. ===Other=== *'''Merge''' to [[Just Another Example Page]]. They're about the same topic, but both articles can benefit each other if merged. *'''Redirect''' to [[A different kind of example page]]. They've got the same info, same sources, but the other article has more to offer. <span style="font-family: Blackadder ITC; background-color: green; padding: 2px 3px 1px 3px;">[[User:Helloheart|<span style="color: white"> Helloheart </span>]] [[User talk:Helloheart|<span style="color: white">(talk)</span>]]</span> 03:26, 1 October 2022 (UTC) :This would just exacerbate the endemic problem of people treating AfD discussions as votes rather than discussions leading to consensus. Much better would be to discourage people from prefacing their comments with '''Keep''' or '''Delete''' unless they are absolutely sure. The change from "votes for deletion" to "articles for deletion" was made 15 or 20 years ago, but the former model still seems to have a hold on people. [[User:Phil Bridger|Phil Bridger]] ([[User talk:Phil Bridger|talk]]) 20:19, 1 October 2022 (UTC) :I think this proposal will lead to more comments <s>stricken</s> or moved and the proposal does not lend itself to better discussion. While there may be some value to administrators in counting !votes, it would be must harder to determine how consensus evolves during a discussion. One of the best features of the usual template of discussion is that you can easily see if the concerns of participants are addressed (such as consideration of newly identified sources). --[[User:Enos733|Enos733]] ([[User talk:Enos733|talk]]) 21:23, 1 October 2022 (UTC) [[File:Tom Lea - 2000 Yard Stare.jpg|thumb|right|"I must have closed at least a thousand AfDs, including the [[Wikipedia:Articles for deletion/Mass killings under communist regimes (4th nomination)|largest one in the project's history]]" -- [[User:Joe Roe|Joe]]]] ::Almost never is even a long AfD as difficult to assess for a closing admin as a long RfC. Additionally, I concur with the others above. I like to read the AfD from start to finish, because the natural flow of discussion makes it easier to identify rebuttals to arguments, counter-rebuttals and so on. That matters more given we judge policy and reasoning, not a numerical count. I dislike the trend to threaded discussions elsewhere on the project, as imo it shouldn't be used unless the participants simply can't keep the discussion under even a semblance of civility. While AfDs do flare up, time to time, rarely beyond the ability of the standard toolkit to return to acceptable norms. [[User:Nosebagbear|Nosebagbear]] ([[User talk:Nosebagbear|talk]]) 22:39, 1 October 2022 (UTC) :I must have closed at least a thousand AfDs, including the [[Wikipedia:Articles for deletion/Mass killings under communist regimes (4th nomination)|largest one in the project's history]], and not once have I found it difficult to "read all of the !votes in a row". – [[User:Joe Roe|Joe]] <small>([[User talk:Joe Roe|talk]])</small> 10:46, 5 October 2022 (UTC) *I too prefer to read comments in chronological order; I tend to give slightly more weight to trends; especially since articles at XFDs can be edited during the process, and new sources can be found along the way, and things like that, so if I see that some initial objections have been ameliorated by editing, the chronological ordering makes that clearer. I would oppose breaking this ordering to collate the keeps and deletes together; that kind of ordering also tends to emphasize vote counting, which we shouldn't do either. --[[User:Jayron32|<span style="color:#009">Jayron</span>]][[User talk:Jayron32|<b style="color:#090">''32''</b>]] 14:44, 18 October 2022 (UTC) *The key part of "AFD" is the "D" which stands for "discussions" -- this measure (although easier to assess) would slow down the discussion portion. And since AFD is not a voting process, it would fly in the face of that. It's easier to read, I'll give it that! Worth a prelim discussion.--[[User:Paulmcdonald|Paul McDonald]] ([[User talk:Paulmcdonald|talk]]) 19:59, 27 October 2022 (UTC) ::A quibble… While I agree that the “D” in AFD ''should'' stand for “discussion”… it does not in fact do so. Perhaps that needs to change. [[User:Blueboar|Blueboar]] ([[User talk:Blueboar|talk]]) 20:11, 27 October 2022 (UTC) == Make it easier to find User contributions == Currently when a user page and talk page doesn't exist, it's not intuitive for the typical user to distinguish whether such user exists, or whether someone hasn't created a user page yet. For user [[User:Example]] my proposal would be to modify the error/not found page, with a link to [[Special:Contributions/Example]]. A more aggressive approach could be to automatically create a default User/talk page, and allow people to remove/redirect as they please. Thoughts? ~ 🦝 [[User:Shushugah|Shushugah]] (he/him • [[User talk:Shushugah|talk]]) 21:59, 29 October 2022 (UTC) :@[[User:Shushugah|Shushugah]]: When a user exists, irrespective of whether or not their user(talk) page exists, there will be "User contributions", "User logs", etc. under the "Tools" section of the sidebar. Alternatively, you may use [[WP:MoreMenu]] (available as gadget at prefs) that creates a drop-down menu besides Twinkle, giving access to important links about a certain user. When a certain username ''does not exist'', none of these user-specific links are shown. <span class="nowrap">—'''[[User:CX Zoom|CX Zoom]]'''[he/him]</span> <sup class="nowrap">([[User talk:CX Zoom|let's talk]] • {[[Special:Contributions/CX Zoom|C]]•[[User:CX Zoom/X|X]]})</sup> 08:15, 30 October 2022 (UTC) == Guidelines for use of video == Video is an option whenever adding media to an article, but it seems that there's very little discussion of when the use of video is appropriate. There's no manual of style page for videos, as [[MOS:VIDEO]] just takes you to a disambiguation page for other things. [[WP:VIDEO]] describes the technical aspects of making and uploading videos for Wikipedia, but again, very little about their use. Is there a good reason why there's no guideline for video use, or is it just that no one has gotten around to it yet? Surely there are important considerations when using videos in articles that warrant a proper MOS page. What are these considerations? [[User:Thebiguglyalien|Thebiguglyalien]] ([[User talk:Thebiguglyalien|talk]]) 06:15, 30 October 2022 (UTC) :My guess is, the need for guidelines for good ways to use video will arise when we get a significant amount of bad use of video. Until then, we can use the ideas we already use for still pictures, with commonsense departures when they seem appropriate. Not much need to codify what is to be encouraged or discouraged, as long as we're getting so little anyway. [[User:Jim.henderson|Jim.henderson]] ([[User talk:Jim.henderson|talk]]) 04:38, 1 November 2022 (UTC) :My sense is that most considerations for videos can be easily extrapolated from the existing guidance for images, and that creating a separate page for videos would therefore be [[WP:CREEP]]. If you have in mind any video-specific style considerations, it'd be interesting to know them so that we can discuss! <span style="color:#AAA"><small>{{u|</small><span style="border-radius:9em;padding:0 5px;background:#088">[[User:Sdkb|<span style="color:#FFF">'''Sdkb'''</span>]]</span><small>}}</small></span> ^{[[User talk:Sdkb|'''talk''']]} 07:31, 2 November 2022 (UTC) ::I've been thinking about how situations where a video might benefit an article. I think it would be interesting if biographical articles had a video of the person whenever possible, which would give a better visual to supplement images (images can't show demeanor, physicality, voice, etc). Or maybe more videos could be used in articles about movies, especially for movies that are in the public domain. And then of course there's everything listed under [[Commons:Commons:Video#Importance of video|this list]] on Wikimedia Commons. I figured I'd come here to ask if there's precedent, or if there are any constraints that apply specifically to video use. And of course there's the discussion of whether they should be embedded or used as external links. I guess my point is that there's a lot of uncharted territory with videos that might be worth exploring. [[User:Thebiguglyalien|Thebiguglyalien]] ([[User talk:Thebiguglyalien|talk]]) 17:48, 2 November 2022 (UTC) :::Where in an article to put movies that have fallen into the public domain is a somewhat open question, and one that's going to become increasingly salient as more enter into it. I [[Wikipedia_talk:WikiProject_Film/Archive_78#Where_should_videos_of_public_domain_films_go_in_the_article?|raised it at the film project talk page]] a little while ago, and I'd love to see it considered more formally. <span style="color:#AAA"><small>{{u|</small><span style="border-radius:9em;padding:0 5px;background:#088">[[User:Sdkb|<span style="color:#FFF">'''Sdkb'''</span>]]</span><small>}}</small></span> ^{[[User talk:Sdkb|'''talk''']]} 21:11, 5 November 2022 (UTC) :Per Jim.henderson, ''policy follows process'' and not the other way around. Best practices evolve organically in the community, and [[WP:PAG|policies and guidelines]] are written to describe those best practices; however the only time we really need written policy is when there are situations where there is significant deviation from best practices. The low implementation of videos means that 1) consistent practices have not had time to evolve and 2) there's no exemplar set of "best practices" to even determine what is a "bad practice". General principles from other similar [[WP:PAG|policies and guidelines]], such as [[MOS:IMAGE]], is likely sufficient to deal with videos, by analogy. --[[User:Jayron32|<span style="color:#009">Jayron</span>]][[User talk:Jayron32|<b style="color:#090">''32''</b>]] 11:39, 2 November 2022 (UTC) == Learning best from the best a.k.a. robust database about successful people. == Hello, my dream is to be very knowledgeable well rounded person and to become successful artist/storyteller one day. I also big fan of James Cameron, George Lucas, Steven Spielberg and many other successful artists. I try to find info about them and read/watch their interviews and talks, but its all randomly spread across internet and its hard to find all information. And in current sea of information i think its best to learn from the best and I'm interested in things like what books(fiction, non-fiction), movies, documentaries, shows, music, games, podcasts, sites and maybe even specific knowledge James Cameron might of mentioned/recommended. For example in his conversations with Otago museum he mentioned books like "The Soul of a New Machine" by Tracy Kidder, "Deep Survival: Who Lives, Who Dies, and Why" by Laurence Gonzales. I already got the books and am planning to read them. And i want easily browsable user generated database where you could search through all mentions and recommendations with sources where they mentioned it(and timecodes for YT videos). And switch modes of that database and just see list of all of their youtube videos, text interviews, podcasts, books they made themselves, etc. I think it could be very educational for all fans and just curious people. For example other Cameron fan showed me video with Jim where he mentioned documentary "Forks over Knives", i watched that doc and decided to become vegan after that. And being vegan is one of the most effective measures to help the environment and slow down climate change. Also very beneficial to health. Can you imagine if many people had access to such database i am describing and learn even more useful knowledge? Can we please make something like that? [[User:Artfart7|Artfart7]] ([[User talk:Artfart7|talk]]) 21:04, 28 October 2022 (UTC) :Can someone please help me figure out how to do it? I always read Wikipedia, but only now created an account. Do i, like, create a new article with tables and start making data entries or do i need permission for that? Also i don't know how to make most efficient tables for robust database i need. Please help. [[User:Artfart7|Artfart7]] ([[User talk:Artfart7|talk]]) 00:49, 2 November 2022 (UTC) ::@[[User:Artfart7|Artfart7]]: Encyclopedias just summarize stuff for quick reference. This isn't really the place to document every interview James Cameron's ever had. [[User:RAN1|RAN1]] ([[User talk:RAN1|talk]]) 02:31, 2 November 2022 (UTC) :::I see your perspective. But maybe Wikipedia shouldn't limit itself to being just encyclopedia. If we really want for humanity to advance then knowledge should broaden, deepened and be available for everyone for free. If everyone is smarter world would be a better place. And my question to you what is essential function of encyclopedia - make short summarizations? or share and educate knowledge? ::: I had fantasies before that if i were one of the richest guys on the planet i would build wiki 2.0 where subjects like animals, transport, architecture and lot of other stuff that would be presented in quick summarizations AND multiple deep levels of knowledge. For example page on transport - cars - specific old car where would be presented photos, videos, blueprints, rotatable 3D with all parts, complete info on how it was build, materials, logistics, stories with that car in news and pop culture and everything else. Same for animals, architecture, industrial design and of course other knowledge like psychology, neurobiology, physics etc. There are already youtube channels from scientists, universities that share knowledge for free. Just search Stanford Online or Andrew Huberman on youtube. ::: And my idea of database about successful people just a little glimpse of that. Its a bit more focused than trying to learn everything, its trying to learn from the best to try make similar success of your own. [[User:Artfart7|Artfart7]] ([[User talk:Artfart7|talk]]) 07:44, 2 November 2022 (UTC) ::::I don't think we have a way to do all of that quickly right now, even with wikis, or else Wikipedia would have a lot more articles. [[User:RAN1|RAN1]] ([[User talk:RAN1|talk]]) 00:03, 6 November 2022 (UTC) == Reducing Conflict, increasing civility <!-- - editor leveling, restrictions on high conflict articles and abuse. WMF privacy--> == <!-- I have been convinced that we have to live with Ip editors and trolls. It's part of who we are, but we have a duty of care to editors, we want new editors, and WMF's continued media campaign is turning new and existing editors off. The below is about my 7th version (and I thank people for their patience). TL:DR New editors to have levels, At each level there are restrictions on avenues they can abuse other people, as they level up these are removed, Restrictions include Word based abuse filter, not able to do to change nationality, no access to edit high conflict pages, can't see other user's pages (they use information there to create attacks), All short codes are expanded out, which allows editors can use this expanded text option to use community created canned answers Log of editor stalking type behaviour to be automatically kept - checking contributions, check user tools, WMF behaviour that creates privacy risks , or deters editors are to be stopped We issue our own community media release after rfc. We advise WMF of a metric of toxicity. --!> How can we reduce the real/perceived conflict/anger in Wikipedia in certain processes and areas? Profanity/Curses/Slurs are just one problem Changes : 8/11/22 changed idea name to be "reducing conflict, increasing civility" and renamed sections to make it clear what they were about. [[User:Wakelamp|Wakelamp d[@-@]b]] ([[User talk:Wakelamp|talk]]) 12:32, 25 October 2022 (UTC) <!-- :Details :'''Leveling of editors - make them part of the community''' :* Different levels? There is no way a new editor can ever catch up on barn-stars or edits,. Leveling up gives them a status or goal :* Levels based on edit/char x on the last n days (but excluding all the ways you can cheat) Opt in for experienced editors, For new editors, they have no way of opting out until they each a certain level? But it's gamification/social. Yep. But it doesn't apply to you, and will reduce other experienced editors stress :* they deserve low conflict. :'''Early levels - But they have responsibilities''' :* No access to edit or see other user pages? People attack based on the content :* No ability to change gender, nationality or country, or death? Lots of conflict :* No access to edit conflict pages (any history of sanctions, high number of reverts, or top 5 % of pages with most editors? Conflict :* Key abuse Word filters. If one editor has a filter in place it applies to all including admins? Conflict :* No access to add images to another user's talks ? Porn :* Only Standard Sig? Conflict based on flags :* Full visibility (not rollover) of level / admin status on talk as part on sig ? Don't bite the newbie :* At each level, they are given a chance to subscribe to spotlight, or help on Vital Articles ? Community :* When they first can create pages, they have to watch a video, are only allowed to create based on wIkidata (but no AfC), next level by choosing an info box (The info box copies the relevant notability rules across)? Reduce NPP/AfC work, Rage Quits :* Default to Expand short codes (but not short codes within short codes), so editors can use canned responses eg "Wow, unfortunately your love of Yorkshire Weather Reports 1923-1927 needs to be backed by references. This video often helps people to understand. Have you thought of helping on Vital articles? Lowers aggression level. The intensity of aggression is related to core beliefs/place , personalisation of issue, and participants social skills, Vital articles are awesome :-) :'''Existing editors - and so do we''' :* viewing an editor's contribution log, maybe checkuser tools, using queries to access goes on a log viewable by a very few anti-stalking :* Cabals - don't know. But biased articles are reducing editors, and risk preaching to the converted :* A measure of toxicity, agreed in the community, and advised to WMF and press beforehand (so they can not claim that there has been no change). This is also to show the number of health and safety involvements. Their websites advise they are on call 24/7. Preferably this measure matches reddit, and some diversity focused sites.--> <!-- :'''WMF changes - Privacy Protection - and so do they''' :* WMF to not out editor user names on their media release. Regardless, An external lawyer of community choice is to explain risks. WMF to take on responsibility for any consequences, mental health support. :* New editors are not to be surveyed unless they opt in when creating account? Users leave after survey :* No tracking of Ip address or user paths or tracking pixels, of editors, readers or donors for fundraising, research, or in dumps, or in test databases. In particular no vanity banners for large donors ? We are voter database heaven, People working for WMF have worked for voter database companies. The anonymity is easily broken by using comments and date/time tags to link back to the original editor. Researchers are using info boxes, edit data, to work out other editors location, interests, [https://arxiv.org/pdf/1712.06414.pdf Politics] [https://dl.acm.org/doi/abs/10.1145/3442442.3452350 Gender , and maybe religion and other socioeconomic data] --> :[[User:Wakelamp|Wakelamp d[@-@]b]] ([[User talk:Wakelamp|talk]]) 12:42, 25 October 2022 (UTC) ::There are a lot of ideas there, including some I've supported in the past. On other sites such as Quora your drafts can be viewable only by yourself. I think if we enabled editors to have some unpublished userpages to sandbox in it would reduce some of the policing work, and it would then be a clear process of changing your sandbox from private to public with the implication that others could then make "helpful" edits to it. ::I'm not sure what you are trying to achieve by saying the WMF should stop including usernames on press releases. The UCOC controversially prohibits this by moving from a no doxxing rule to no attribution rule off wiki. In the past the WMF has erred on the side of not giving credit to specific volunteers, even where it is very much due. I get that we don't want the WMF doxxing people by giving out more personal info than the individuals choose to link to their username. But our CC-BY-SA culture is very much about attributing people by their preferred username. I wouldn't want to see the WMF press releasing about one of us without consulting that volunteer, or at least giving a heads up when it is a minor mention such as an image attribution. But it would be good if they at least made the UCOC compatible with us honouring the BY part of CC-BY-SA when we use content from these sites ourselves. ''[[User:WereSpielChequers|<span style="color:DarkGreen">Ϣere</span>]][[User talk:WereSpielChequers|<span style="color:DarkRed">Spiel</span>]]<span style="color:#CC5500">Chequers</span>'' 09:53, 26 October 2022 (UTC) *This is very confusingly arranged, even as an idea list. Each of these needs a demonstration of why it is an actual problem - diffs or links to ANIs and RfCs that show the issues actually do occur and at a rate making it advisable to make these changes - currently I'd be surprised if a number of these were happening at any appreciable rate and thus the evidence disproving my position is core to include. Especially with this added, I'd then suggest breaking it up further. <!-- Template:Unsigned --><small class="autosigned">— Preceding [[Wikipedia:Signatures|unsigned]] comment added by [[User:Nosebagbear|Nosebagbear]] ([[User talk:Nosebagbear#top|talk]] • [[Special:Contributions/Nosebagbear|contribs]]) 11:00, 26 October 2022 (UTC)</small> === 1.0 Outstanding Questions ==== # Fixed - profanity is different from toxicity and should be separated out- # Is Civility more important than no rules/no censorship/anyone can edit? # Should an editor have to tolerate poor behaviour? # How do you determine what is acceptable/friendly? # Does conflict matter? === 2.0 Does Profanity/Curse/Slurs occur at an appreciable rate? === There are some [[Wikipedia:Administrators' noticeboard/IncidentArchive982|examples]] from ANI, but there are far more incidences based on the statistics below. I think many editors don't know about ANIs, RfCs, and some may have issues with [[WP:CIVIL]]. ==== 2.1 Profanity : Fuck /Fuckwit/Fuck off Statistics (still in progress) ==== Using fuck as an example *ANI 1262 cf. fuck off 496 fuckwit 49 shut up 1380 *Article page title only 498 *Article 12,394 cf. Fuck off 5798 Fuckwit 26 Shut up 51299 *Article revision history TBA *Article talk 9466 cf. Fuck off 6281 Fuckwit 95 Shut up 17697 *Article talk revision history comments TBA *User page title 79 *User page 8636 cf. Fuck off 2223 Fuckwit 34 Shut up 7096 *User page edit history TBA *User talk 15960 cf. Fuck off 8547 Fuck wit 204 Shut up 17991 *User talk edit history TBA *Article talk 9466 cf. fuck off fuckwit 95 shut up 17678 *Edit history pages TBA *Edit history talk page TBA *Speedy Don't know *Anti-vandalism - Don't know[[User:Wakelamp|Wakelamp d[@-@]b]] ([[User talk:Wakelamp|talk]]) 15:56, 1 November 2022 (UTC) :As raw numbers, these are pretty meaningless. I mean, let's say that there are 6281 uses of the phrase "fuck off" in article talk pages. There are about {{NUMBEROFARTICLES}} article talk pages. That means that less than 0.1% of article talk pages (on average) have seen the phrase "fuck off". I don't know if that's a problem or not, I'm just saying that proportions are more meaningful than raw numbers. We'd have to decide if 1 in a thousand is an acceptable level of times to have that phrase written in an article talk page or not. I mean, I wrote it twice in this thread, but [[Use–mention distinction|certainly not in an uncivil way]]. You yourself wrote it 6 times just in compiling your stats. So, given that a non-negligible number of uses will be not-uncivil, that should also be taken into account. It would need a human level of understanding, not just a database scrape, to classify a particular occurance of "fuck off" as uncivil or not. I'm not saying your research is not worthwhile, it likely is very worthwhile, but if you're going to do it, you need to make sure that your statistics have meaning. --[[User:Jayron32|<span style="color:#009">Jayron</span>]][[User talk:Jayron32|<b style="color:#090">''32''</b>]] 16:19, 1 November 2022 (UTC) ::I've typed the words 'fuck off' lots of times on Wikipedia. I am very confident that I have never used it in an uncivil manner. If anyone finds an example of me behaving like a fuckwit, I will shut up shop and go home. [[User:Girth Summit|<span style="font-family:Impact;color:#006400;">Girth</span> <span style="font-family:Impact;color:#4B0082;">Summit</span>]][[User talk:Girth Summit|<sub style="font-family:Segoe print;color:blue;"> (blether)</sub>]] 16:24, 1 November 2022 (UTC) :::@[[User:Girth Summit|Girth Summit]] If you never misbehave, please stay to increase the tone, but everyone is entitled to some [[wiktionary:fuckwittery]] As an Australian, my nation prides ourselves on on our ability to use fuck in every fucking sentence, but seriously does using fuck off on another user's talk page or on an edit summary make WP better? (as a total aside and attempt at humour, wouldn't (looking at your editor page), the same thing said in Gaelic "Taigh nam gasta ort" be classier?) [[User:Wakelamp|Wakelamp d[@-@]b]] ([[User talk:Wakelamp|talk]]) 12:38, 2 November 2022 (UTC) ::::My Gaelic is non-existent, but Google Translate tells me that means "you are a guest house"; how dare you?! Seriously, I can think of lots and lots of ways 'fuck off' could be said innocently, and usefully. ::::Example 1: I used to do a lot of training over at CVUA. If I am giving someone advice about how to deal with people, I might write something like ''"People often get pissed off when their edits are reverted. They might even get abusive and tell you to fuck off. The best thing to do is to stay professional and explain in simple terms why you reverted them."'' What you've got there is an example of someone discussing the term itself, not using it against someone; that happens a lot, and will show up in raw searches just the same. ::::Example 2: I sometimes write articles in collaboration with other editors. Some of these people I have become friendly with; some I've even met up with in the real world for a drink or a walk. If I'm writing with one of these people, and I make a suggestion they disapprove of, they might tell me to fuck off, which between us would just be a friendly, jovial way of saying 'I disagree with that proposal'. What you've got there is an example of the phrase being directed at someone, but between two people who know each other well, know that it will not offend, and would speak to each other face to face in exactly the same manner. That's not any kind of problem. [[User:Girth Summit|<span style="font-family:Impact;color:#006400;">Girth</span> <span style="font-family:Impact;color:#4B0082;">Summit</span>]][[User talk:Girth Summit|<sub style="font-family:Segoe print;color:blue;"> (blether)</sub>]] 12:58, 2 November 2022 (UTC) ::@[[User:Jayron32|Jayron32]] Thank-you for your reply I agree that we are both using fuck, :-) Your point about proportions is well taken, but my guess is that only 5 % article/user talk has ever been edited by a non-bot assisted human editor. An issue with both counts and proportions, is that it only looks at the current page (so no deletion/reverts/etc). I will do a separate section with more background [[User:Wakelamp|Wakelamp d[@-@]b]] ([[User talk:Wakelamp|talk]]) 12:40, 2 November 2022 (UTC) :{{tq|"Does conflict/profanity occur at an appreciable rate?"}} Two different things. And if there is verifiable evidence that reducing the latter would reduce the former, I'd like to see it. [[User:AndyTheGrump|AndyTheGrump]] ([[User talk:AndyTheGrump|talk]]) 12:52, 2 November 2022 (UTC) ::Actually, what tends to happen at Wikipedia, paradoxically, is as long as a personal attack or other uncivil behavior is ''accompanied'' by profanity, it tends to be ''less likely'' to result in a sanction against the user in question. Profanity-laden abuse gets a free pass where abuse that lacks profanity is more likely to get an appropriate response. Basically, as long as you say "fuck" while attacking another user, enough users will come to your defense so that you don't get sanctioned. It's so important that we ''don't'' sanction profanity that profanity becomes a magic shield that protects those who use it from ''other'' consequences to their actions unrelated to the profanity. --[[User:Jayron32|<span style="color:#009">Jayron</span>]][[User talk:Jayron32|<b style="color:#090">''32''</b>]] 13:08, 2 November 2022 (UTC) === 2.2 Any Confirmation of profanity/curses/slurs ? === # I found about 200 K user talk or article talk discussion pages currently with 20 problem phrases. Shit seems popular, as were common slurs. Profanity embedded in Editor names is common (we block them with bots, but the vandal gets a permanent user name to point at). Other vandals appear to have the apparent aim of getting a vandalism warning on their editor page for a profanity # A separate study that I have misplaced found that a common pattern was conflict on a prolific editors talk, a drop to minor changes than exit. I also think that editors get very stressed, and conflict starts. This conflict is often from trolls # ( WMF and the Media describes the community as toxic, and there are [https://scholar.google.com.au/scholar?hl=en&as_sdt=0,5&q=wikipedia editor toxic OR or OR racist OR or OR sexist OR homophobic OR transphobic OR angry OR uncivil&btnG= 169 K] academic papers that may portray a negative view of editors or the community *Any confirmatory proof? Many editors complain about [[Wikipedia:Civility]] on talk/quora/facebook/reddit, and a Google Jigsaw/WMF project aligns with that opinion, The project manually classified a 100K sample (from 2015) ten times 10 times, and then AI the other 68 million comments. It is a very good filter, but not good enough (I think high 90s) to auto detect. The research found that *9 % of toxicity was created by 34 editors *80 % were by 9500 editors with less than 5 comments, which include 40 % by anons. * 3 toxic attacks in a 12 month period gave you a 50 % chance of being blocked or warned. My guess is that new editors are targeted * that user talk had a [https://github.com/ewulczyn/wiki-detox/blob/master/src/figshare/Wikipedia Talk Data - Getting Started.ipynb 3 times] higher proportion of attacks than article === 2.3 Any Ideas to reduce profanity/curse/slurs these areas? === The main areas seem to be * Ip editors especially on user talk * Editors with less than 5 comments * New editors (profanity by them, or at them), * The very small group of high toxicity editors, and * Attacks at prolific editors on user talk * Editor Names - we blot them with bots, but the vandal gets a permanent user name to point at. * Vandalism with the the apparent aim of getting a vandalism warning on their editor page * High conflict articles * and maybe Abusive edit summaries [[User:Wakelamp|Wakelamp d[@-@]b]] ([[User talk:Wakelamp|talk]]) 13:26, 2 November 2022 (UTC) :Profanity usually isn't a problem, because Wikipedia's not-censored policy makes this a very moot point. Civility, on the other hand, requires professionalism that people get paid for at work, but not here. [[User:RAN1|RAN1]] ([[User talk:RAN1|talk]]) 16:10, 2 November 2022 (UTC) ::The level of profanity that is acceptable is very much dependent on culture. As an Englishman I am often surprised that many Americans seem to treat things as profanity that I consider part of normal discourse, but am equally surprised by the high level of profanity still considered civil by many Australians, Glaswegians and Irish people that I have known. A bit of understanding is needed by everyone on a multicultural project such as Wikipedia, and certainly not this equation of profanity with conflict and toxicity. [[User:Phil Bridger|Phil Bridger]] ([[User talk:Phil Bridger|talk]]) 16:32, 2 November 2022 (UTC) ::I'd agree that we have a problem with IP editors and other throwaway accounts, and I'd add that some people see them as part of Wikipedia's civility problem, while others see them as the toxicity that we instantly reject from Wikipedia. Two very different ways at looking at a group of editors who we all agree get quickly blocked and reverted. I'm not sure there is so much agreement on there being a "very small group of high toxicity editors" or who gets into that group and on what basis. Or rather we can all agree that there have been such editors, but which currently banned editors meet that criteria isn't so easy to agree, just look at various ban appeal discussions. Describing any current members of the community as high toxicity would be more contentious, and unless it was accompanied by an arbcom case it is hard to see how you could do it without making a personal attack. As for dodgy editor names, I'm not sure how much of that is bot based, I thought it was still mostly a manual process. How is it a problem when we create a blocked userpage? As for the idea that civility belongs in a paid environment and only in a paid environment, I beg to differ. I've worked in a professional environment where the language could some times strip paint off the walls, and I've volunteered in environments where politeness was de riguer. I grew up in a culture where profanity was a normal part of all-male conversation, but was instantly dropped when in mixed company. I suspect part of our problem is that some people don't pick up on the cues that mean that all wiki conversations are in "mixed company", such as that non trifling business of them being recorded and publicly shown indefinitely. ''[[User:WereSpielChequers|<span style="color:DarkGreen">Ϣere</span>]][[User talk:WereSpielChequers|<span style="color:DarkRed">Spiel</span>]]<span style="color:#CC5500">Chequers</span>'' 17:26, 2 November 2022 (UTC) :::The high toxicity editor description is based on the WMF/Jigsaw research (see above) - personally I was concerned whether this was a misinterpretation of a template comment . Some of the IP editor comments are pretty [[User talk:92.78.207.186|extreme]] (although aimed at a bot is pointless), and there is an [https://en.wikipedia.org/w/index.php?title=Special:AbuseLog&wpSearchFilter=12 abuse filter] with about 140 K hits at the moment :::The problem with the dodgy user names is that is its not good for the people who deal with these sort of things; they are defending Wiki from people that are insulting/attacking it. interestingly the AI researchers who did subsequent work had ethical research about the crowd source workers' mental health). Other reasons are that it wastes experienced editor time fixing, and the vandal still gets their jollies form having their name visible, and new editors may have created an innocent name (A company I worked for for had a Herr Fucker) and then been told it was deleted, :::I grew up in a similar environment, and civility was treating people equally and with respect if they respected you and others, so I agree with this [[User:Beeblebrox/fuck off|essay]]. [[User:Wakelamp|Wakelamp d[@-@]b]] ([[User talk:Wakelamp|talk]]) 01:06, 3 November 2022 (UTC) ::::The research was flawed - it lacked nuance and was also using definition of toxicity that I don't believe the community would accept. The same runs true for your bits here - not all (indeed most) wiki profanity doesn't add to project toxicity. "9 % of toxicity was created by 34 editors" was also never going to be the case - toxicity requires an audience who can see it (otherwise, at most, it'll be attempted toxicity) ''and'' it actually has to be problematic ''and'' it has to be genuinely scaled (one toxic edit isn't another). ::::I'd actually say it's offensive to suggest that, say, swearing at a user is as toxic as doxxing them or threatening to murder them or contacting someone's employer. [[User:Nosebagbear|Nosebagbear]] ([[User talk:Nosebagbear|talk]]) 09:29, 4 November 2022 (UTC) ::::: Good point. I have changed the section heading to "Any Ideas to reduce profanity or toxicity or incivility in these areas". Your comment that toxicity requires an audience, made me realize that Wikipedia is very different from other forums in that user talk is more like a private chat room. If an editor runs into problem on user talk, the current process [[No personal attacks]] is basically don't feed the trolls. This section was for ideas about reducing the issues. What do you think we should do to reduce? [[User:Wakelamp|Wakelamp d[@-@]b]] ([[User talk:Wakelamp|talk]]) 13:25, 4 November 2022 (UTC) :I really think we need to split profanity out from this - it isn't the problem. Telling someone that you don't trust them, or that they don't understand something because they're a <insert class of person> is offensive, regardless of whether or not you say 'fucking' when you say it. :We don't have many tools at our disposal to make people less unpleasant to others though. We have policies that require us to be civil to each other, and we have human editors who sometimes fail to abide by those policies - whether that's because they're trolls, or because they're used to a different kind of discourse in different places, or because they get frustrated, or simply because they're cantankerous so-and-sos. We also human have administrators who sometimes fail to enforce the policies strictly enough, or who enforce them too strictly (probably both at the same time depending on who you ask). I don't know that there is any more that can be done - it's the encyclopedia anyone can edit, so long as they're not enough of a dick to piss everyone off. [[User:Girth Summit|<span style="font-family:Impact;color:#006400;">Girth</span> <span style="font-family:Impact;color:#4B0082;">Summit</span>]][[User talk:Girth Summit|<sub style="font-family:Segoe print;color:blue;"> (blether)</sub>]] 17:03, 5 November 2022 (UTC) : If we can't make people less unpleasant to others through a tool, then what about if we let the offendee decide on their user talk/page? We could give each user the power to control access to their user talk/messaging them/email, by blacklisting and editors. If it's an admin issue, then another admin can deal with it. [[User:Wakelamp|Wakelamp d[@-@]b]] ([[User talk:Wakelamp|talk]]) 05:18, 6 November 2022 (UTC) ::What you describe is essentially the current situation. Some things you can do for yourself - you can switch off the ability for editors to email you, and you can switch off the notification system. Also, if Alice tells Bob not to post on her talk page, long-standing convention is that Bob needs to stay away from there, except in a very limited set of circumstances (e.g. posting mandatory notifications of a noticeboard discussion). Bob would be blocked if he continued posting. Same goes for pinging someone, etc - essentially, if it looks like an account is trolling, we block it. If lots of people are trolling a particular person on their talk through changing IPs, we would apply semi protection so that IPs can't edit it. We also have one-way and two-way [[WP:IBAN|IBans]], which require that people stay away from one another if they just can't get along. [[User:Girth Summit|<span style="font-family:Impact;color:#006400;">Girth</span> <span style="font-family:Impact;color:#4B0082;">Summit</span>]][[User talk:Girth Summit|<sub style="font-family:Segoe print;color:blue;"> (blether)</sub>]] 16:50, 6 November 2022 (UTC) :::I am not sure whether new editors know about any of the ways to resolve disputes; do new editors turn up complaining about user talk of the admin boards or teahouse? Even if they did the [[Wikipedia:Blocking policy]] states that it is for "persistent personal attacks" (which means that a one off attack which makes an editor leave doesn't count), encourages not imposing blocks, and there no extra penalties for unfairly attacking new editors. I get the everyone has a bad day/week/month logic, but it means that there is no feedback to encourage better behaviour. [[User:Wakelamp|Wakelamp d[@-@]b]] ([[User talk:Wakelamp|talk]]) 12:12, 7 November 2022 (UTC) :::What issues do you see with an editor being able to block another editor from their user talk? Especially if it was for a set period - fuck off for a week. One thing that it would do is allow bad behaviour by an editor with multiple editors to be visible. :::For me, user talk is like a pre [[Eternal September]] type unmoderated small IRC chatroom; the user talk pages are "invisible" in that no one else sees them and it is not clear what to do if flamed. So maybe, we get with the 90s and move towards personal blacklists :-) (Apologies if I have my IRC history wrong) , or give them access to more email-like features - unsubscribe from newsletters, mark as spam. block, delete flames. or archive it. (The arses that revert deleted comment (non admin templates) also should go to [https://en.wikipedia.org/w/index.php?title=Wikipedia:BLANKING&redirect=no <nowiki>WP:BLANKING]]] hell :-)) Wakelamp d[@-@]b (talk) 13:39, 7 November 2022 (UTC)

• See, the issue is, you've lumped three things together, one of which is unrelated to the other two. Toxicity AND incivility are major issues that need to be dealt with. Profanity is entirely unrelated to either of them. --Jayron32 13:42, 7 November 2022 (UTC)

  Profanity is always uncivil. Look at the start of Profanity, where we define it as "a socially offensive use of language". There is no daylight between "being socially offensive" and "being uncivil".

  The question then becomes: Was that use of language really profanity, when properly and fully analyzed? I'm old enough to remember boys writing "Hello" and using a finger to cover up the last letter, and thinking themselves very daring indeed. The social meaning of certain individual words have changed a lot since then, but if it's actually profanity, rather than something on some mid-century list of Seven dirty words, then it's automatically uncivil.

  As for whether it's okay, I suggest applying the Robustness principle to yourself. I emphasize those last two words because it's unfortunately common for people to invoke this while claiming the opposite of the actual meaning. We need people who say "These words are offensive to a subset of people – including, disproportionately, women, older people, people from non-English-speaking countries, and people from more 'formal' cultures – so I won't use them myself, even though they don't bother me". The internet does not need any more idiots who believe that Postel's Law is only binding on the person they're cussing at and not on themselves. WhatamIdoing (talk) 19:53, 7 November 2022 (UTC)

  Why is it always the most offended person who gets to define what is socially acceptable or not? Yes, I generally try to assess the situation, and avoid language likely to enflame. However, there are words that are offensive to some and not to others; who gets to define, in the context of Wikipedia, which words are or are not "socially acceptable" here? So, perhaps I can concede that profane words exist, I will agree there are words we should be not using at Wikipedia. Let me revise my contention then. Fuck is not one of them. The OPs focus on that word is what the problem is. --Jayron32 13:24, 9 November 2022 (UTC)

  I remember some experienced editors at this wiki being offended some years ago when they were called power users. I think it's silly to be offended by people saying that you have above-average knowledge and skills, but a couple of them genuinely were. So what do we do?

  I think that if you follow Postel's Law yourself, then you stop using that term, at least around the people whom you expect to be offended (e.g., because they have previously expressed discomfort with the term, or because they share some similarities with people who have previously expressed discomfort with it).

  If you don't embrace Postel's Law, you might tell them that they're being unreasonable/overly sensitive/too easily offended, or you might try to dictate their emotions ("you shouldn't feel offended when I say that").

  If you're a jerk, you probably use the offense-provoking word as often as you can, just to show everyone that you refuse to be constrained by factors such as considerateness, tactfulness, or tolerance for other people's views/quirks/feelings. Or, you know, civility. Being a jerk is the opposite of civility. WhatamIdoing (talk) 22:25, 9 November 2022 (UTC)

  I see your point about why does the offended party get to decide ; people are becoming more easily offended. But profanity/curse/slurs is an accepted taboo in society. It's purpose is to reduce stress, bond, express anger. or to show dominance; free speech does not mean free audience. Wakelamp d[@-@]b (talk) 07:37, 10 November 2022 (UTC)

  Wikipedia Profanity#Types states it much better

  According to Steven Pinker, there are five possible functions of swearing:

  • Abusive swearing, intended to offend, intimidate or otherwise cause emotional or psychological harm
  • Cathartic swearing, used in response to pain or misfortune
  • Dysphemistic swearing, used to convey that the speaker thinks negatively of the subject matter and to make the listener do the same
  • Emphatic swearing, intended to draw additional attention to what is considered to be worth paying attention to
  • Idiomatic swearing, used for no other particular purpose, but as a sign that the conversation and relationship between speaker and listener is informal

  Wakelamp d[@-@]b (talk) 08:13, 10 November 2022 (UTC)

  Some of the stuff is definitely profane/used in ways meant to wound, especially by very new editors - pick your least favourite swear word and have a shot! I also think the anti-vandalism people may delete a lot, but I am not sure, Other vulnerable groups are people who avoid conflict, those in less than robust mental and physical health, and the young. ( I hadn't heard of the hell/hello one, but I thought 800873SS was very risque as child when I received my first calculator, although I was/am still in awe that 12345679 *9 = 111,111,111) Wakelamp d[@-@]b (talk) 14:16, 8 November 2022 (UTC)

Proposal: Mark retired editors

Moved from Wikipedia:Village pump (proposals)

— xaosflux ^Talk 16:32, 8 November 2022 (UTC)

I propose creating a bot and adding Template:Retired to the top of talk pages of users who have made no edits in the last 2 years. This will help editors to easily know whether someone has been recently active or not. Loewstrios (talk) 16:15, 8 November 2022 (UTC)

I propose making a bot to add Template:Find something useful to do instead to posts suggesting we waste our time placing potentially-misleading templates on other peoples talk pages. AndyTheGrump (talk) 16:20, 8 November 2022 (UTC)

I cancelled the RFC on this, the idea needs some more development first. "Retired" itself probably isn't the best option, as that is generally an affirmative declaration that you will not be around, vs options where you just happen to have not edited as being discussed above. Feel free to discuss more ideas here. — xaosflux ^Talk 16:35, 8 November 2022 (UTC)

I think helping users quickly identify whether another user will respond to messages is more useful than harassing people for using the Village Pump as it's intended. Thebiguglyalien (talk) 17:13, 8 November 2022 (UTC)

Without commenting on the merits of having a bot do this, you should probably employ {{Not around}} instead. 74.73.224.126 (talk) 00:14, 9 November 2022 (UTC)

Related? Wikipedia:Village_pump_(idea_lab)#Is_there_a_way_to_visually_identify_inactive_editors? ---Another Believer _(Talk) 16:35, 8 November 2022 (UTC)

If you're going to click through to the user page, you might as well install User:PleaseStand/userinfo.js (or similar), which will give you the time since the last edit plus other useful information. WhatamIdoing (talk) 01:05, 9 November 2022 (UTC)

• Oppose any and all attempts at this or anything like it, no matter how it is phrased or what modifications are made. There is no meaningful use to this; if you can find someone's user page, you can find their contribs list. Come to your own conclusions about their editing history. There's no good reason to put "retired" templates in people's user space against their will. Just no. --Jayron32 14:36, 9 November 2022 (UTC)
• Oppose as redundant, there is already a gadget setting that gives data on when someone last edited. It even works on popups.· · · Peter Southwood ^(talk): 02:11, 10 November 2022 (UTC)
• Oppose There are about 45 million registered accounts. Per Registered editors by edit count about half of these have made an edit but the number of still-active editors is comparatively small. So, the proposed update would affect over 20 million accounts and this seems too many to be done in this way.

If we're just talking about the most active editors then another issue is that, witb them, the {{retired}} template is often used as a gesture and such editors often keep editing even with the retired template in place.

What's really needed is a standard warning such as you get when an editor is blocked. If you try to communicate with an editor who has not edited for some time, then the standard interface should warn you of this.

Andrew🐉(talk) 09:53, 10 November 2022 (UTC)

Is there a way to visually identify inactive editors?

There's a setting which lets me easily identify banned/blocked editors because the usernames are faded, italicized, and given strikethrough treatment.

Is there a way to give a similar treatment to inactive users? I'd tired of stumbling on lists and categories of editors, only to find most of them haven't edited Wikipedia in years. Would make browsing a lot easier.

Perhaps this is already an option. Perhaps not. Would other editors see value in this? ---Another Believer _(Talk) 15:03, 3 November 2022 (UTC)

It would be rather useful. Some little symbol indicating if there were x edits in the last y period. Johnbod (talk) 15:27, 3 November 2022 (UTC)

Sure, or perhaps italicizing but not giving strikethrough treatment (to differentiate from a ban/block). Any way of identifying inactive editors without conflating with banned/blocked editors would be helpful. "Inactive" is subjective (we'd need to offer a definition) but we could go with something simple like "no edits in 5 years" or something. ---Another Believer _(Talk) 15:37, 3 November 2022 (UTC)

I'm assuming you're already using WP:POPUPS so you are seeing the last active info there, but you'd like something more readily apparent? I don't really know how the blocked user strikeout methodology works, so I can't speculate on what kind of backend might be needed to support this idea, but it is interesting. DonIago (talk) 17:19, 3 November 2022 (UTC)

@Doniago: I'm less familiar with POPUPS. I see names struck because under my user preferences, I've selected "Strike out usernames that have been blocked" under the Appearance section of the "Gadgets" tab. Here's where we could potentially add the option to italicize or otherwise change the appearance of the names of inactive users. ---Another Believer _(Talk) 17:29, 3 November 2022 (UTC)

@Another Believer I'd agree I find the popups tool is pretty useful for this - mousing over a name gives a very quick indicator of the key data (block status, rights, total edit count, date of first and last edit). I find it really interesting when looking at old discussions to see how often I think 'huh, I remember that name...' and then discover they a) left in 2012 or b) are still prolifically around. It's not so useful for looking at a longer category/list of names, though, since you'd have to scan over each one individually. Worth a try, perhaps.

The code for the gadget is at MediaWiki:Gadget-markblocked.js and it looks like it's querying the API to ask if the user is blocked; I guess the API should just as easily be able to give you the last-edit date, so it's technically possible? I'm not sure I'm technically competent to do it, but I can sort of see how it would work... Andrew Gray (talk) 18:04, 4 November 2022 (UTC)

This sounds a lot like the Who is active gadget that Community Tech put together. It's not available on enwiki at the moment (though an interface admin could copy it over), but you could use the following two lines in your common.js to enable it for yourself:

mw.loader.load( 'https://www.mediawiki.org/w/index.php?title=MediaWiki:Gadget-whoisactive.js&action=raw&ctype=text/javascript' ); // Backlink: [[mw:MediaWiki:Gadget-whoisactive.js]]
mw.loader.load( 'https://www.mediawiki.org/w/index.php?title=MediaWiki:Gadget-whoisactive.css&action=raw&ctype=text/css', 'text/css' ); // Backlink: [[mw:MediaWiki:Gadget-whoisactive.css]]
:

Vahurzpu (talk) 22:10, 12 November 2022 (UTC)

Thanks! Do folks think this is worth submitting as a more formal proposal here ENWP? I'm not too familiar with how village pump proposals work, or where else this discussion could be continued for further review. ---Another Believer _(Talk) 00:34, 13 November 2022 (UTC)

Time in music articles

Very few of the articles on individual songs give us much musical information. In particular, I would like editors to be encouraged to tell us what the time of a song is , i.e., is it 3/4, 4/4, 5/4, 3/3, 6/8 or what? For most songs, this is basic information, which is why I think song articles should include it.Kdammers (talk) 18:38, 10 November 2022 (UTC)

Not a music expert, but can't songs be transcribed in different ways, leading to different time signatures? Also, how would we account for songs that change time signatures? And how would we obtain a definitive time signature? That last part is most important to me, since most people would probably add time signatures as OR. RunningTiger123 (talk) 15:56, 11 November 2022 (UTC)

Thanks for the response. I'm not a music expert either, but [ or "that's why"] I would like to know the time signature of songs. Many of the song articles refer to an "iconic" version of a song, so the time signature for that version should be indicated. If there is another version that's discussed, that can be noted, e.g., 'in John Doe's 1958 cover, the 4/4 time is changed to a waltz (i.e., 3/4 time).' If the time changes, that can also simply be stated, e.g., 'although most of the song is in 4/4 time, the last verse switches to 3/4.' With sheet music, the time signature is given right at the start. For music which doesn't have sheet music (or, more likely, editors don't have it), I admit that this can be an OR problem: for musicians for most songs the time is apparently as obvious as the color of the sky, whereas for other people it is as opaque as mud. But there is secondary material about a lot of songs that indicate (albeit oft indirectly -- but, still, clearly meant) the time. Kdammers (talk) 16:28, 11 November 2022 (UTC)

To discuss time signatures of a musical piece - whether a song or something else, would require discussion in reliable secondary sources (and owing to the transcription issues mentioned by RunningTiger123, probably not primary sources such as scores).Nigel Ish (talk) 16:22, 11 November 2022 (UTC)

What do you mean by transcription issues? With movies, running times, cast etc. are taken from analogous sources. Kdammers (talk) 16:29, 11 November 2022 (UTC)

Why not take it from the primary sources?

@RunningTiger123, just so you don't need to worry about it, if you look at a printed source, and write in Wikipedia exactly what that document has printed on it, that is not original research. Original research could happen when you sit down with your favorite instrument, pick out a little of the song from your memory, and say "Huh, kind of waltzy, so that must be ³
₄." When material comes straight out of a source, it's not the editor making stuff up.

It sometimes helps to remember that the origin of our NOR policy was an Wikipedia editor, 'way back in the day, claiming that he had proven Albert Einstein wrong about special relativity and persistently trying to cram his personal "research" into Wikipedia. That's what NOR is meant to stop: editors making stuff up and using Wikipedia as the first place to post it on the internet. WhatamIdoing (talk) 17:08, 11 November 2022 (UTC)

It's very common to find sheet music for transcriptions of songs online, but many of those are unofficial and/or greatly simplified from the original recording. I could see editors citing those documents for the time signatures, and in those cases, I would argue the time signature is OR. Even when sheet music is released with the approval of the songwriter/label, it's often simplified and may not be entirely correct for the original recording. RunningTiger123 (talk) 18:39, 11 November 2022 (UTC)

OR = "made up by the Wikipedia editor". Copying information exactly from an erroneous source is not the same as "made up by the Wikipedia editor".

What you describe is the garden-variety problem of needing to find a reliable source. This is not an insurmountable problem.

Also, "the original recording" is not really a relevant concept for most music written before the mid-20th century. WhatamIdoing (talk) 07:26, 13 November 2022 (UTC)

Many songs use more than one time signature, and the same piece of music can be written in multiple different time signatures. Some songs are known for their time signature: e.g., Take Five gets its name from being written in 5/4, and we have a list of musical works in unusual time signatures; whereas some entire genres are almost always in the same time signature (waltzes are 3/4, house music is 4/4). I'm not sure if WP:RS about a waltz would mention the time signature unless it was something other than 3/4; same with house music and 4/4, because that would be assumed (I might be wrong about that). There's probably no harm in listing time signatures for articles about songs if there's RS to source it, but I'm not sure every song's time signature is WP:DUE for inclusion. Note also that many of our articles about "songs" are actually articles about notable recordings of songs, and the sheet music the recording artist played from isn't necessarily the same as the published or official sheet music for the song they recorded. The recording artist(s) might not even have used sheet music at all, and many recording artists don't always keep the time exactly as it's written for the whole song, so you can't necessarily tell what time signature a recording is in just by listening to it. And some songs have different time signatures for different instruments that are playing. So it's all rather complicated. It's not like the running time of a film. This is kind of like suggesting every song article list the key of the song: that doesn't work because songs can have key changes, and the same song can be written in different keys. Levivich (talk) 07:02, 13 November 2022 (UTC)

If you're talking about something like God Save the King/America (My Country, 'Tis of Thee), it would probably be unobjectionable to note that the time signature is 3/4 and that the meter is 6.6.4.6.6.6.4, but you'd struggle to pick "the" key because there is no definitive answer (though "at least a full step down from whatever key they sang in during the 19th century" is probably true). Wikidata would presumably want all the details; the Wikipedia article might or might not.

If, instead, you are looking at The Star-Spangled Banner, it'd be more challenging, as 6/4, 6/8 time, 3/4, and 4/4 are all known (the latter frequently being the choice of marching bands). In such a case, I'd be inclined to skip it unless there's a source talking about why these variations matter. WhatamIdoing (talk) 07:42, 13 November 2022 (UTC)

For many styles of Western music, rhythm is a key component, and thus a discussion of a song's timing for its metric beats is important for analysis, even if it follows a common pattern for the genre in question. If it varies during the piece, all the more reason to describe it in the article. If the song's genre is one that lacks regular beats, that's also important to note. A song's rhythmic characteristics or absence thereof are essential elements to describe in some form. If that can be summarized with one or more time signatures, it's reasonable to include this info.

On a side note regarding keys: it's relevant for a specific song recording or performance. The choice of key can be related to the specific instruments used, and for songs with vocals, it will depend on the range of the vocalist. However for the song itself, the key doesn't matter; it's the relationship between the notes used that's important. isaacl (talk) 08:38, 13 November 2022 (UTC)

Articles with parenthetical disambiguation

Is it possible for articles with parenthetical disambiguation to use the title without the disambiguation, but keep the URL the same? I'll give you an example of what I'm talking about: The article on the country of Georgia is titled at Georgia (country). The URL https://en.wikipedia.org/wiki/Georgia_(country) would remain the same, but the display title would be just Georgia, but without the parenthetical disambiguation. Britannica and Fandom do something similar. Examples: https://www.britannica.com/place/Georgia-state and https://shrek.fandom.com/wiki/Shrek_(character). I understand that there purposes are different from Wikipedia, but the concept is still the same. Has there ever been any discussion on whether this could occur? Interstellarity (talk) 23:48, 12 November 2022 (UTC)

@Interstellarity mediawiki supports this concept via DISPLAYTITLE, but we do not here on enwiki; the setting (mw:Manual:$wgRestrictDisplayTitle) is all or nothing, so can't just be for some pages and is off. In general we want a pages title to be it's name - so that people that want to link to it will use the right name in links. — xaosflux ^Talk 01:51, 13 November 2022 (UTC)

@Xaosflux: Thanks. It looks like based on what you said, it is better to use the article title that is linked through the URL rather than use a different display title. Interstellarity (talk) 13:52, 13 November 2022 (UTC)

Automatic enhanced protection on Trans and Nonbinary people articles

Tonight there's been an immense wave of transphobic edits - mostly editing pronouns repeatedly - and almost exclusively with unregistered users (IP Vandalism) at this point it would be prudent to give at least the first level of protection to these articles so that unregistered users would need to go through a pending changes process. — Preceding unsigned comment added by Marleeashton (talk • contribs) 03:56, 15 November 2022 (UTC)

Has this been reported at WP:ANI? It probably should be, if there is evidence of a coordinated effort. AndyTheGrump (talk) 04:28, 15 November 2022 (UTC)

No - I just flagged one of the articles for protection. Didn't know about this page. I'll send it in. I can mention the changes I've reverted but not sure how to prove it's coordinated - I was going through 'recent changes' and so many of them were pronoun changes or bizarre derogatory claims about trans/nonbinary people. So disappointing. Marleeashton (talk) 04:52, 15 November 2022 (UTC)

Wikipedia generally does not allow WP:PREEMPTIVE protection of articles. If there's a specific article that's being vandalized, then protection might be appropriate. Otherwise, the solution is just to WP:Revert, block, ignore. Thebiguglyalien (talk) 17:03, 15 November 2022 (UTC)

The use of Artificial Intelligence images for illustrative purposes

The source of the argument.

I recently came into this discussion in the article Sundiata Keita, I have been somewhat bothered for years by the complete lack of illustrations for what I felt was a rather important historical figure, but my lack of artistic talent meant I couldn't really do much to solve it. However, recently I was able to use an AI software to generate what I believe is a pretty decent interpretation of a portrait fitting the general description of the man when he ascended to the throne (a young west African, who wore white clothes and a white turban into battle). This lead to the argument about whether I am even allowed to make such speculative portrait, and truth be told I couldn't find any specific policy that talked about AI generated images. Maybe there is something I missed. I can find examples of many articles using speculative portraits, so there is precedent for that, but none that are done using AI. Since this may become a bigger issue in the future, it would probably be a good idea to talk about it -- Plank (talk) 23:29, 4 November 2022 (UTC)

@Plank there's actually at least two aspects in play here: copyright (or image ownership) and whether it is suitable for use.

The former alone is likely to be a palava - dalle2 and its counterparts have some oddly structured setup to attempt to end-run round copyright precedent that in US terms is likely (IANAL) to mean they're in the public domain (while in the UK they'd probably belong to the person writing the prompts...until their T&Cs you agreed to had you transfer copyright to them).

The latter is also a very good question (we have allowed certain drawings, and obviously accept portraits that have been made of individuals who weren't ever seen by the artist). Nosebagbear (talk) 00:33, 5 November 2022 (UTC)

I think there's both the speculative part and the WP:OR aspect (see Wikipedia:Manual_of_Style/Images#Making_images_yourself also I suppose). I personally am hesitant about speculative portraits in general but those that come out a research program at a university or museum if accepted by others in the field are fine. We do tend(?) to allow self-works for things like diagrams, charts, maps, drawings, photographs, etc; I don't know of any detailed drawings/paintings of persons for biographies are self-works on Wikipedia but I'd be interested to see some.

There are also copyright concerns which enwiki seems to care deeply about and I'm not sure if there have been any discussions about how the murky licesing/authorship of things from DALL-E (which your image seems to be, based on the watermark on the bottom-right) and Stable Diffusion models would fit in our licensing requirements. At least for DALL-E, apparently: "an OpenAI spokesperson said, 'OpenAI retains ownership of the original image primarily so that we can better enforce our content policy.'"[5] so my understanding would be that DALL-E images would be a no-go here, in general, for copyright reasons since they seem to only provide an exception for "Social media, livestreaming, and demonstrations policy" instead of a real license.[6] And it's possible some of the artists whose work was used in the model training may retain some copyrights! So at best it's very unclear. Skynxnex (talk) 00:43, 5 November 2022 (UTC)

I don't really think this qualifies as "research" does it? I really didn't seek any new information that wasn't already available in article's bibliography (Niane, D. T., 1965). Making a portrait would fall more in the category of an artistic endeavor (although there is probably some discussion to be had about AI art being art, but that's mostly irrelevant here). As for copyright, according to the OpenAI website "Subject to the Content Policy and Terms, you own the images you create with DALL·E, including the right to reprint, sell, and merchandise – regardless of whether an image was generated through a free or paid credit.". So I see 2 possibilities a) I hold the copyright, so I am free to put the image in public domain b) the image is public domain by default because it came from a machine, and in that case it's free to use--Plank (talk) 02:45, 5 November 2022 (UTC)

Thanks for link to their support page saying that. Somehow I only found... Other things when I searched. I don't know if it is only either of those things. If you asked for a book say that was "dragon vampires" and a machine spit back, literally, half of Twilight and half of A Game or Thrones that wouldn't be public domain or you own it. So there's some line somewhere. For OR, I'll have to think more and I might just be wrong and it is equivalent to painting a picture in this case (although much less input from the creator.) Thanks for bringing this up, since I think enwiki will have to gave guidelines about this. Skynxnex (talk) 12:29, 5 November 2022 (UTC)

• I think a lot would depend on how the image was captioned… we would want the caption to clearly indicate that the image is an “AI generated artist’s rendition” of the subject “based upon written descriptions” (or something like that)… so that the reader knows that the image is a guess as to what the subject might have looked like, and not necessarily accurate. Blueboar (talk) 12:58, 5 November 2022 (UTC)

  This WMF project sees similarish,

  "When BIPOC histories are told, too often they lack visual representation. Of the 30 articles in the African Royalty category on English Wikipedia, for example, only three pages picture their subjects. Visual aids and illustrations help people understand and retain information; their absence reinforces and perpetuates biases. " Wakelamp d[@-@]b (talk) 15:51, 5 November 2022 (UTC)

Probably worth mentioning, the Openai terms of use do forbid tricking people into thinking the image was done by hand. When I initially added it to the article I did caption it as artificially generated since I didn't want to deceive the reader--Plank (talk) 16:48, 5 November 2022 (UTC)

• There is a precedent with exoplanet illustrations. There are just the imagination of an artist and are not based on a real image or known reality. AI generated images could be similar, but it may be more based on other images than pure imagination. Graeme Bartlett (talk) 06:03, 6 November 2022 (UTC)
  • I agree that these should be permissible, with appropriate disclaimers. Frankly, we are at the beginning of this new reality, and it is only going to become a bigger issue going forward. BD2412 T 06:06, 6 November 2022 (UTC)

    @Plank, I'd ask the licensing/copyright questions over at Commons, because I think they are more likely to give you solid advice.

    In general, I think there's nothing wrong with creating an "Artist's impression" with these tools and uploading it to Commons. It would be nice to have half a dozen different options for editors to choose from. I gently suggest that you use the phrase "artist's impression" or something similar in the description. You might also be interested in c:Category:Artist's impressions, especially the subcategories for computer-generated images.

    In the specific case, since there is a credible minority POV that he might have been Muslim, then you should consider a color change: Muslim men do not wear gold jewelry. WhatamIdoing (talk) 07:18, 8 November 2022 (UTC)

    Copyright aside, this is not a good idea. A "young west African in white clothes" is incredibly generic. You could probably apply the same picture to a myriad of biographies. Creating a speculative image is WP:Original research (and likely specially WP:SYNTHESIS depending on how the algorithm works), and that's before you get into issues such as the one WhatamIdoing raises. We have a lower bar for image WP:V as a general culture, but that shouldn't extend to a generic and newly invented image. CMD (talk) 14:03, 8 November 2022 (UTC)

The unavoidable fact is that a masive percentage of articles about ancient and medieval figures are illustrated with artists impresions made centuries after he person's death based in very limited or even no information. For example, the article Avicenna is currently illustrated by an image made in the 1950s, and the only basis for it being Avicenna, and not let's say, Almanzor (who is illustrated with a statue made in 2002) is because the artist said so. Furthermore, WP:Original research also states that "original images created by a Wikipedian are not considered original research"--Plank (talk) 17:58, 8 November 2022 (UTC)

Those are depictions of the individuals in question that exist in use elsewhere in the real world, not inventions of Wikipedians. CMD (talk) 01:45, 9 November 2022 (UTC)

They are inventions nonetheless. I think the thing to do in those cases is to have a discussion about what such an image should look like based on the available data, generate a bunch of options, and then have a discussion about whether any of those options fits, or if further refinement of the image is required. BD2412 T 14:09, 9 November 2022 (UTC)

• I am having a hard time expressing why this feels so wrong; but I think I still prefer that humans make human decisions for all things done at Wikipedia. If a person creates their own graphic, that's one thing, but encouraging, or even allowing, AI created graphics seems like step down a road I am not comfortable walking down. --Jayron32 14:39, 9 November 2022 (UTC)
  • Surely you can't look at that proposed image and think that it doesn't look like the product of human decisions. To the extent that a human determined the prompts to be used, and apparently selected from multiple possible outcomes for those prompts, a human decision has been made. BD2412 T 14:44, 9 November 2022 (UTC)
    • Well, at some level, humans created everything we're talking about. Humans invented computers in the first half of the 20th century. It's that enough? No, the issue is the proximity between the end product and the human creation. If I have a picture of a person, I still want a picture someone made by looking at that person, or barring that, at least some historically significant image that has its own reasons for being well known. The notion that we can just type "Mansa Musa" into an AI art program and pick one of the things it spits out at us and say "Good enough" is a bridge too far. --Jayron32 14:49, 9 November 2022 (UTC)
      • Obviously we're not just typing "Mansa Musa" into an AI art program. We are (or the OP is) typing in a set of characteristics corresponding to what is known about the subject. BD2412 T 15:02, 9 November 2022 (UTC)
        • Somehow, that feels even less likely to be useful. "a young west African, who wore white clothes and a white turban into battle" is now LESS likely to be the subject of the article, not more. Again, my only two requirements would be 1) The actual image made by an actual person, who got to actually look at the actual subject or 2) Barring that, a speculative image that was also created by a person, where the image has some historical importance. For example, it is unlikely that the person who created the Bayeux Tapestry witnessed the entirety of the Battle of Hastings, and probably just as unlikely that the people so depicted are meant to be faithful representations of their appearance. Since we don't have any verified contemporary portraits of William the Conqueror, as a second-best option, I think the Wikipedia article is fine using his depiction from the Bayeux Tapestry because the image is so historically significant. Where neither is true (not a faithful depiction, not historically significant) we're running into some tenuous territory with encyclopedic value. If we add the third "whammy" of "AI-created", I am having a hard time seeing the image as useful in any way. "What some rando sitting at his computer thought he might have looked like" is probably not good enough for a portrait to be used in a Wikipedia article, and "What some rando sitting at his computer fed into an AI program" is even less so. --Jayron32 15:21, 9 November 2022 (UTC)

I am not too worried about the copyright issues (as far as I am aware our consensus is wholeheartedly in favor of such images being free and Commons has held so many times). However, one thing that is extremely concerning to me is that images like this typically make great haste in getting out into the "information ecosystem", or whatever the kids are calling it now. This is going to immediately start showing up whenever you look for that guy in a search engine. It is going to get fed into Wikidata, and returned by all kinds of automated things online; the fact that it looks convincingly like a painting means that it will basically be impossible to tell it isn't one. That's to say nothing of on our own articles -- sure, we can have a caption saying "this isn't actually the guy and somebody just made this up" but at that point why have a picture at all? I mean, we would not just put a photo of Matt Damon on Charlemagne's article because we're pretty sure he was a white guy with dark hair and we don't have a photo of him to compare with -- this would make no sense. To me an incorrect illustration is worse than none; exoplanets are a special case because it is physically impossible to have a photo of them (and so everybody knows it is merely guesswork). But many paintings of old dead people do exist, so there is a great potential for confusion. jp×g 11:34, 17 November 2022 (UTC)

Using AI to improve Wikipedia

There ought to be a central place to discuss the use of AI/algorithms to edit or improve Wikipedia. This would encompass the following:

• using summarization algorithms to summarize books/web sources, to help editors quickly parse high-quality but dense academic content and expand the article with it
• using AI to repair poorly formatted citations, repair incorrect citations, and scour the web and Google Books to add new citations for unsourced claims
• using language models to detect poor grammar, and propose copyedits in the Visual Editor
• using language models to create short or long-form text with citations, to improve help us expand articles and move more articles towards GA or FA quality, especially those that are currently neglected
• using more recent/advanced machine translation tools to redo previous machine translations made years ago, which would be a vast improvement.
• improving the Translation Tool's machine language translation integration, so editors waste less time fixing formatting and spend more time reviewing citations and content.
• improving the consistency and reliability of Wikidata

These AIs all currently exist in open-source forms, in more-or-less advanced stages, and have made enormous progress in just the last few years. But they're hard to use for untrained editors, may require editors to train the models on their own machines, and lack recommendations on their proper use. Some may feel these algorithms "aren't ready for prime-time", but they may save many editor-years of work, and we've crossed the positive benefits-costs threshold a while ago. The more we can use AI to assist with stuff it's good at, the more time editors have to improve the rest of Wikipedia. Wikipedia greatly lacks editors, so the judicious and forward-looking use of AI tools may help us vastly improve the encyclopedia, minimize editor workload, and get better results. This is an area where small-c conservatism just holds us back, with backlogs ballooning and editors feeling overwhelmed.

This research is quite promising, but there ought to be a more prominent, centralized way for hobbyists, AI researchers, and editors to collaborate on creating and improving such tools, and making them easier for trusted editors to use. This could use some WMF backing as well, and it would be nice if the WMF would directly collaborate with AI research centers and universities to improve such algorithms. It would greatly help their mission to reduce systemic bias as well as encyclopedia quality and reliability. DFlhb (talk) 11:53, 4 November 2022 (UTC)

"we've crossed the positive benefits-costs threshold a while ago." Citation needed. While this may perhaps be true for a bullet point like "detect poor grammar and propose copyedits", I seriously doubt it for "move more articles towards GA or FA quality" or even the "summarize books" level. Perhaps link to these "already existing" AIs and examples of what they do, so we can discuss this based on some actual facts? Fram (talk) 12:07, 4 November 2022 (UTC)

• Summarizing: https://openai.com/blog/summarizing-books/
• Repairing citations: https://tech.fb.com/artificial-intelligence/2022/07/how-ai-could-help-make-wikipedia-entries-more-accurate/
• Creating text: GPT-3 or similar algorithms, manually trained on a corpus of reliable sources relevant to the subject picked by an editor or found by an AI (see above).
• More recent/advanced machine translation tools: DeepL is available on the web, just not built-in to Wikipedia's Translation Tool. It's far better than Google Translate, and I maintain that using it to redo previous machine translations would be a vast improvement and help clear that backlog.
• Improving reliability: the already-existing ORES, and see meta:Research:Community-centered_Evaluation_of_AI_Models_on_Wikipedia.

GA or FA quality won't happen through AI tools alone, but they can help bring an article from a stub to B or C-class, so editors can dedicate their efforts on reaching GA. At no point am I suggesting we should deploy machine learning tools without editor oversight, or that their use should be possible without editor oversight.

As these AIs become more competent, it's useful to have a central place to discuss their improvement (that are relevant to future use on Wikipedia), and where hobbyists and research institutions can get more direct community feedback on new proposed AI tools. DFlhb (talk) 13:09, 4 November 2022 (UTC)

Ugh, no, please. We had to ban an editor last year who used Deepl to create articles based on German Wikipedia articles. The results were not acceptable, way too many mangled sentences, incomprehensible or simply wrong results. The summary site you give: the very first page of the very first example they give of their work (one would assume they vetted it and are proud of it) is

• Original
• "‘Well!’ thought Alice to herself, ‘after such a fall as this, I shall think nothing of tumbling down stairs! How brave they’ll all think me at home! Why, I wouldn’t say anything about it, even if I fell off the top of the house!’ (Which was very likely true.)"
• AI Summary[7]:
• "She thinks about how brave she’ll seem when she gets home, and how she’ll never tell anyone about her fall."

Which, well, is not the same at all.

Despite all the claims to the contrary, AI is far from ready to be used for these things. Fram (talk) 13:54, 4 November 2022 (UTC)

Fair. Maybe this idea comes too early. DFlhb (talk) 13:58, 4 November 2022 (UTC)

I think so, sorry. E.g. the Meta (Facebook, not Wikimedia) article above discusses 2017 in classical music and how they find a better source (well, a blog, but who cares?) Too bad that there example is just wrong, the system identified somehow the Dallas source as referencing the LA announcement, when these were always two different things (in the current version of the article source 183 and 186). Letting loose this AI on that article would have replaced a correct source (presumably 186) with a blog source for a different fact, which would have made the article worse, not better. This is not some error from among hundreds of examples, this is the example they presented from their own work, after presumably having removed the more obviously wrong results from the AI. Using such tools is just an accident waiting to happen. Fram (talk) 14:07, 4 November 2022 (UTC)

These were the edits made after the AI article appeared: [8][9]. So there was a mix-up between the two press releases that predated the AI article. As a result, the AI solution was suboptimal, but it's probably fair to say that this error would have lain undiscovered for years otherwise. Andreas JN466 15:18, 4 November 2022 (UTC)

I agree DeepL is better than Google Translate and that it can be a great help but it is just that: a great help. You still need to have a fully competent human translator checking through the output. Andreas JN466 15:21, 4 November 2022 (UTC)

An algorithm is not a reliable source. No editor should be using AI-produced 'summaries' of material they cannot understand. AndyTheGrump (talk) 12:17, 4 November 2022 (UTC)

I agree that the WMF should collaborate more with university researchers on tools to assist with editing. In my opinion, it's a great match of needs: Wikipedia offers a large database of content and user interactions to mine, WMF has money to help with funding (which often can be leveraged by researchers with matching grants from other sources), and university departments have cheap grad student labour. isaacl (talk) 16:00, 4 November 2022 (UTC)

Definitely; while my proposal for a centralized discussion place may be premature, I still stand by my suggestion that Wikimedia become more proactive, if only to avoid all the work being done by Meta & Google Brain, two companies which many editors distrust; or which may avoid releasing AIs as open-source and attempt to charge Wikimedia for their use. DFlhb (talk) 16:12, 4 November 2022 (UTC)

To me, having a venue with a specific focus is up to those involved. If tool developers would find a new forum useful, they can go ahead and create it. (Forums for tool users will generally be focused on functionality and process, rather than underlying technology.) isaacl (talk) 16:21, 4 November 2022 (UTC)

This idea looks at things from the wrong end. The thought process should start with asking what needs improving and then look at what tools can help with this, rather than start with particular tools and ask how we should use them. Phil Bridger (talk) 16:10, 4 November 2022 (UTC)

If you're interested in this, you may be interested in m:Abstract Wikipedia. Whatamidoing (WMF) (talk) 18:47, 4 November 2022 (UTC)

The first thing I thought about this summer, when I saw NNLB, was the potential to one day end up with a single-language Wikipedia, that gets translated on-the-fly for readers and editors, and where the latter's contributions are automatically reflected across all languages. The constructor idea from Abstract Wikipedia seems more readily achievable, since it doesn't rely on machine translation being perfect. DFlhb (talk) 20:57, 4 November 2022 (UTC)

I may have been unclear, but that was actually my thought process: the problems I'm attempting to solve are the decline in active editors, the neglect of our vital articles, and the sustained backlogs. I don't see any systemic, non-short-term solutions except AI. DFlhb (talk) 19:59, 4 November 2022 (UTC)

I fail to see how replacing active editors with unreliable AI (presumably leaving the editors to clean up the mess) would reduce any decline. I'd have thought it more likely to exacerbate it. AndyTheGrump (talk) 20:15, 4 November 2022 (UTC)

While there was a decline in editing during the 2007/2014 era, part of that was the effect of the edit filters automating edits away. In any event there is more editing taking place now than in 2014, editors come and go, but talk of editor decline as if it is still 2015 is dated. ϢereSpielChequers 11:37, 6 November 2022 (UTC)

We should have a guideline laid out at Wikipedia:Artificial intelligence (separate from Wikipedia:WikiProject Artificial Intelligence, which is about our coverage of the subject in articles). BD2412 T 21:46, 6 November 2022 (UTC)

Ooh, Abstract Wikipedia:

• 2022: Launch Wikifunctions in production; start community discussions about the architecture
• 2022: Start development of Abstract Wikipedia proper
• 2023: Integration of Abstract Wikipedia into Wikipedia proper and other sister projects"

Should be here any minute now! I guess the community discussion is already planned and I just missed the announcement? Fram (talk) 14:02, 9 November 2022 (UTC)

It's only 1444, per the Islamic calendar. Give them some time! DFlhb (talk) 14:18, 9 November 2022 (UTC)

Christ we need to split enwiki off from this nonsense. The more years I spend here, the more it sinks in that the WMF is using donations to pursue goals that are completely different than the goals I am pursuing when I donate my time here. Levivich (talk) 15:43, 9 November 2022 (UTC)

This is an important issue, and one that will only grow in importance, and I think that any editor who does not try to read as much about it as possible and keep abreast of developments is doing themselves a disservice. Not wanting to be a man of talk and no action, I wrote GPT-2 (while the new one is more relevant, it has basically identical architecture to the old one). Also, we tried using the new one for some Signpost articles, about which I wrote at length in the August issue (Wikipedia:Wikipedia Signpost/2022-08-01/From the editors). Currently, there is this big shiny thing called "Galactica" from Facebook, which is the latest in a long series of efforts by said company to create "AI" that can "write" Wikipedia "articles" (with varying amounts of scare quotes on all three of those words). That notwithstanding, I think the problem is more societal than technological, and that if the Facebook efforts have not been useful it is likely because they are being designed and built by Facebook employees rather than Wikipedia editors.

"AI" is a bullshit phrase that I try to avoid whenever possible. The reason for this is that first of all it's poorly defined, and second of all the definition is constantly changing due to advertising and political reasons; if you want an example of this, look at this image, which illustrates our own article on "AI"..... and was generated using a single line of code in Mathematica! Simply put, "AI" means "using computers to do stuff that computers aren't good at". If we just say the actual thing that most "AI" is -- currently, neural networks for the most part -- we will find the issue easier to approach. In fact, we have already approached it: the Objective Revision Evaluation Service has been running fine for several years. I would recommend that some more attention be given to stuff like this, perhaps through a big RfC, perhaps through guidelines or supplements to existing guidelines, perhaps by longer conversations here, who knows. jp×g 08:41, 17 November 2022 (UTC)

Critically, these discussions and guidelines should under no circumstances be called something like "Wikipedia:Artificial intelligence", because nobody (including artificial intelligence researchers) knows what that means, and it will lead to strange discussions like the one above. It is all well and fine to say that we should "never use artificial intelligence" to write Wikipedia articles, except Google searches now use Bidirectional Encoder Representations from Transformers to parse queries and rank results (see here). Similarly, if we "didn't use algorithms" it would be literally impossible for the servers to process HTTP requests and send pages to our web browsers. The main controversy at hand here, I think, is whether large language models should be used for unsupervised mass editing, and the main issue at hand is to what extent they can prove helpful in daily editing life. jp×g 08:48, 17 November 2022 (UTC)

I would recommend that some more attention be given to stuff like this do you have any specific things in mind? DFlhb (talk) 09:22, 17 November 2022 (UTC).

Well, the first thing I would say is that everyone should go read some stuff to learn about the current state of ML models and their capabilities (maybe one doesn't exist and I will have to get off my ass and do something for once). On the side of proposals, it would be useful to get an idea of what exists and can easily be adapted to our purposes. A number of optimistic gee-whiz models purport to do very ambitious stuff like multimodal writing from numerous sources -- what is the current state of the art? Some of these tasks will be easier than others. jp×g 11:26, 17 November 2022 (UTC)

All of that sounds great; both forward-looking and realistic. I'd help if I had any CS background. DFlhb (talk) 11:51, 17 November 2022 (UTC)

Users' "Contributions" page search box

Tracked in Phabricator
Task T12788

I suggest we add an option for searching users' contributions by articles titles, adding it to the search box in users contributions page. - Joaquin89uy (talk) 11:34, 10 November 2022 (UTC)

@Joaquin89uy: this sounds like feature request phab:T12788 (which has been open since 2007). — xaosflux ^Talk 11:46, 10 November 2022 (UTC)

@Xaosflux: - That thread seems to "fizzle out"... Is that the case? Or am I misreading it? - Joaquin89uy (talk) 12:17, 10 November 2022 (UTC)

It's still open. As far as your idea above goes, we can't do this, because the software doesn't support it. Someone could update the software to do it, and there may be some notes on T12788. That it has been open for 15 years without resolution could be that it is very hard, or just that no one is interested in writing it. Just like articles, most of the software is also written by volunteers. — xaosflux ^Talk 15:20, 10 November 2022 (UTC)

This could also be done client-side with a script perhaps: Fetch 5000 entries, then hide all the entries that don't match your search string. — xaosflux ^Talk 15:22, 10 November 2022 (UTC)

@Xaosflux: I wouldn't have a clue of how to do that, lol. But thanks anyways. - Joaquin89uy (talk) 16:28, 12 November 2022 (UTC)

• Startist can do this (right now it is set up to work with new sections only, but it could be modified easily to work for any user-defined search string). jp×g 11:52, 17 November 2022 (UTC)

  Another tool that does a similar job is Edit summary search. A similar tool could filter a given user's edits by page title rather than summary text, though it would often be much more efficient to filter page history by user instead. Certes (talk) 12:13, 17 November 2022 (UTC)

Files, Templates, and Categories

I just realised that this is going to get confusing fast, so I have split this topics into sections, and moved the comments to match,

Hi, I was wondering about the number of edits being an incentive. We seem to have many pages in various namespaces that serve little purpose.

1. Templates

• 740317 Templates of which 593177 have no redirects
• 475061 Templates of which 366069 have no redirects Wakelamp d[@-@]b (talk) 04:11, 20 August 2022 (UTC)

• The standard has always been that if you see a need for a template, make a template. As a result there is some amount of duplication and redundancy. Over time templates can also be superseded or lose their original purpose. Superfluous templates are mostly harmless, there are editors who work on merging, redirecting, or deleting templates where beneficial. 74.73.224.126 (talk) 14:52, 19 August

I'll note as I have recently started working on creating and consolidating certain templates, it's a difficult process without a bot to run with you. I haven't put in a request yet per just the scope of my project, but in the meantime the consolidation has required supporting aliases and features that I then deprecate with plans to quickly drop. (And these are low-use templates, but I'm not manually converting 100, 50, or even 12 page usages just to consolidate, especially when I know I'll need a bot later anyway). Also, there's a matter of the dormant template creators you may risk offending. An essay of recommendations of good behavioral/deletion/consolidation practices in template space might be worthwhile (I can't find one on quick search, but I should search harder to not contribute to essay bloat (you forgot complain about that one, OP!) SamuelRiv (talk) 15:56, 23 August 2022 (UTC)

Well done! 100 K templates definitely needs a bot, and that's one edit count increase I would be very happy with. :-) Essay bloat now added, although I expect the numbers are small, and I do like reading them as they give hints to past issues.Wakelamp d[@-@]b (talk) 01:27, 25 August 2022 (UTC)

Please note also that for historical reasons, DYK pages are created as templates, which would account for several new ones every day. My thumbnail estimate is that there are about 75,000 of these. BD2412 T 20:39, 25 August 2022 (UTC)

I count ~60,000 nompages, but six o' one :) theleekycauldron (talk • contribs) (she/her) 23:16, 12 September 2022 (UTC)

Could we move DYK and similar noms to WP: or even WT: namespace (since they're discussions)? That would tidy things up a lot. Certes (talk) 10:36, 13 September 2022 (UTC)

I think this is a very good idea, and that we should have done it years ago. The next-best time is now. WhatamIdoing (talk) 18:53, 13 September 2022 (UTC)

Can we move Main Page to WP: while we're at it? --Ahecht (TALK
PAGE) 03:52, 18 October 2022 (UTC)

Agree. Very good idea. Jason Quinn (talk) 14:10, 22 September 2022 (UTC)

@74.73.224.126. It should be that if you see a need for a template, create it but reality is that there are editors who seem to jump on the chance to delete newly created templates. So creating a new template is not a user-friendly process and can turn into opening a can of worms. Jason Quinn (talk) 14:14, 22 September 2022 (UTC)

So, as my template knowledge is very low, what would people suggest as a wording for the proposal. Also Is there a way we could reduce down the numebr of templates if we had more functionality? Wakelamp d[@-@]b (talk) 15:50, 17 November 2022 (UTC)

2. Files

• If they are a copy of Wikimedia commons, why do we have 705271 talk pages?? Wakelamp d[@-@]b (talk) 04:11, 20 August 2022 (UTC)

• Categorization, most of these have one or more templates that add them to tracking categories. This may not be all that helpful, but it's not harmful either so is mostly ignored. 74.73.224.126 (talk) 14:52, 19 August

Wouldn't the category be on the file page rather than talk? Wakelamp d[@-@]b (talk) 15:50, 17 November 2022 (UTC)

Category:Fair use images contains 665000 files Christian75 (talk) 06:59, 10 October 2022 (UTC)

Some wikiprojects actively tag files for tracking purposes similar to how they track articles. Imzadi 1979 → 03:42, 8 November 2022 (UTC)

3. User Talk has 6 time as many pages as User

• User 3,852,785 and User talk 17,976,469 Wakelamp d[@-@]b (talk) 04:11, 20 August 2022 (UTC)

• Not surprising at all. I don't have a link to the research handy but the vast majority of accounts never make a single edit on any wiki; the second most common number of edits per account is one, after that two, etc. If an account reaches autoconfirmed it's already part of a tiny elite. While it doesn't make sense to track IPs with zero edits a similar pattern emerges; IPs with one edit are most common followed by two etc. This was true even before the introduction of IPv6. Hence most of those user talk pages are just template messages, often added by semi-automated tools, welcomes, warnings, deletion notices etc. For accounts or IPs that will never edit again. At one time it was common to delete temporary user and user talk pages for accounts after a set period, but this has long-since fallen by the wayside. Deletion of old IP user talk pages was also carried out under an alleged CSD that never actually had consensus, but that too was halted. Since there's no value to deletion this really isn't surprising, though a bot is now tasked with blanking old IP talk messages. 74.73.224.126 (talk) 14:52, 19 August
• Yes, especially if these figures include archives & other sub-pages. I have 46 talk archives & at least 20 sub-pages for drafts etc. Johnbod (talk) 20:05, 25 August 2022 (UTC)
  • Same. BD2412 T 20:45, 25 August 2022 (UTC)

Aside from the points made above, many users (even very active ones) never bother to make a user page because they do not feel a need for one. These users may receive posts to their talk page and even respond to them, but still not make a user page. User:Khajidha (talk) (contributions) 18:27, 14 October 2022 (UTC)

SUMMARY - No change needed Wakelamp d[@-@]b (talk)

4. Categories

• In 2010 we had 638,834 Categories
• In 2022 we have 2,188,563 Categories
• Category talk is 1865015 pages
• I have raised a Quarry request for the breakdown by month. I would also like to find out how many times that readers actually click on a category Wakelamp d[@-@]b (talk) 04:11, 20 August 2022 (UTC)

• As with templates the idea is that if you see a need for a category you should go ahead and make it. Many categories are just for tracking or maintenance. The number of articles has more than doubled since then so the increase in categories isn't really that surprising. Category talk pages mostly function as file talk pages do (i.e. see above). 74.73.224.126 (talk) 14:52, 19 August
  • I would add to this that there are certain maintenance categories for which a new subcategory is generated every month, for instances of the categorized issue arising in that month. BD2412 T 20:46, 25 August 2022 (UTC)

Also is there any objection to me asking for the following changes to WIkimedia statistics or should I move it as a proposal

• Change Top Edits to Most Edits
• New Editor type semi-automated to pick up Hotcat etc. (They are currently part of user) Wakelamp d[@-@]b (talk) 01:56, 25 August 2022 (UTC)

  I have updated my query request to ask for chars, as well as counts. I have been advised is not possible to get the number of times a reader clicks on categories. Wakelamp d[@-@]b (talk) 04:57, 1 September 2022 (UTC)

Do any editors use compound non maintenance/hidden categories? Is it just for Wikidata? My interest is that in 2013, that the NYT complained that [https://www.nytimes.com/2013/04/28/opinion/sunday/wikipedias-sexism-toward-female-novelists.html American female novelists were being moved out of their American novelists), does anyone know the background? The root cause is we have compound categories (with immense amounts of Nationalistic ones). I think I suggested a similar change before, but my reason is totally different now. is this something volunteers could develop?Wakelamp d[@-@]b (talk) 04:14, 26 October 2022 (UTC)

5. New Stubs

• A lot of stubs would seem better suited to Wiki species or wikidata, and may be being created semi-automatically *Many have 1 ref to amateur web sites. Wakelamp d[@-@]b (talk) 04:11, 20 August 2022 (UTC)

• There's a lot of history and controversy here. Suffice it to say that the communities views on notability have shifted over time, and many current articles exist as a legacy of earlier times. Semi and even fully automated article creation was tolerated in the past, Rambot created a large number of American county and municipality articles. Wikispecies has a different purpose than Wikipedia, and some of the information here would not be suitable there. Wikidata didn't even exist until 2012 long post-dating many of the mass database style stub creations. 74.73.224.126 (talk) 14:52, 19 August

Are these past/current discussions consolidated/linked somewhere? It looks like all the minor asteroid and sky survey object bot-created stubs have been cleaned up into lists since the last time I checked many many years ago. Municipalities are somewhat different because people live there (though arguably it may be a greater incentive to try editing Wikipedia if your municipality had no page than if your municipality had a terrible bot stub, although the barrier to entry is far greater (the Draft/AfC structure did not appear to affect retention, though I'd like to see a well-designed comparative study on retaining new editors whose first experience is editing existing articles versus creating new ones -- and now I'm off-topic). SamuelRiv (talk) 15:38, 23 August 2022 (UTC)

I think the asteroids have been added to,

Going off topic is always allowed :-) My User_talk:Wakelamp#IP_editing_research has a list of the Portuguese/Farsi trial of no IP edits and I think the results of edit vs create. I think the failure rate on new articles is more then 95% for new editors. Most are killed off by speedy. Wakelamp d[@-@]b (talk) 01:49, 25 August 2022 (UTC)

6. Why Recent Changes are so big?

• Apart from the above, we also have some editors constantly doing multiple edits on the same page within seconds Wakelamp d[@-@]b (talk) 04:11, 20 August 2022 (UTC)

• I seem to remember a proposal in the past to combine multiple edits made by the same user to an article over a short period of time which was shot down for both technical and practical reasons. Anyway unfiltered recentchanges has moved too fast to be a useful tool since at least when Rambot got up and running; filtered versions are still quite useful. 74.73.224.126 (talk) 14:52, 19 August

7. Weird Stuff on User

• I think it might have to do with Bots, but I found weird pages that just update the time Wakelamp d[@-@]b (talk) 11:51, 19 August 2022 (UTC) Wakelamp d[@-@]b (talk) 04:11, 20 August 2022 (UTC)

• There's a lot of weird stuff in userspace, for the most part it's best ignored. 74.73.224.126 (talk) 14:52, 19 August

SUMMARY - No change needed Wakelamp d[@-@]b (talk) 15:50, 17 November 2022 (UTC)

8. Articles

• Ok - Namespace 0 (articles0 looks weird Wikipedia:Statistics The numbers on Namespace 0 (articles) do not match those on Wikipedia:Statistics Wakelamp d[@-@]b (talk) 12:38, 19 August 2022 (UTC) Wakelamp d[@-@]b (talk) 04:11, 20 August 2022 (UTC)

• Zero-indexing is the norm in the world of computation.74.73.224.126 (talk) 14:52, 19 August

• Apologies - It wasn't the zero index that confused me, it was the mismatch between the numbers. I have corrected it. I now realise that the official number is the one without redirects Wakelamp d[@-@]b (talk) 04:11, 20 August 2022 (UTC)

• Correct. Separately, try to remember to add an additional line break or use a template to create paragraphs for unindenteded text (see previous rendering here). Finally when revising your own comments after they've been responded to try to adhere to Wikipedia:Talk page guidelines#Editing own comments, I've fixed it for you in this instance, but some people may get irritated when those are not followed. 74.73.224.126 (talk) 06:09, 20 August 2022 (UTC)

  Thank-you for that information. After 13 years I am still learning how to do things, and I appreciate others tolerance. Wakelamp d[@-@]b (talk) 09:43, 1 September 2022 (UTC)

9 Other

Taking a stab at this 74.73.224.126 (talk) 14:52, 19 August

I have moved your comments to match the new topics, although I have 2sections in categories. Please feel free to correct. Wakelamp d[@-@]b (talk) 04:11, 20 August 2022 (UTC)

 Done; probably should've copyedited earlier for concision and ease of reading, but it is what it is. 74.73.224.126 (talk) 06:09, 20 August 2022 (UTC)

10 Portals

Wikipedia:Village pump (proposals)/Archive 157#Hiatus on mass creation of Portals Wakelamp d[@-@]b (talk) 08:45, 25 August 2022 (UTC)

11 Essays

Wakelamp d[@-@]b (talk) 08:45, 25 August 2022 (UTC)

12 Comparison with other WPs

The enWP ratio of article to other seems high.

https://meta.wikimedia.org/wiki/List_of_Wikipedias]]

Wiki	Articles	Total	Ratio	Edits	Admins	Editors
Vietnamese	1,275,834	19,379,100	15.189	69,010,746	20	878,846
English	6,547,077	56,453,893	8.623	1,103,443,021	1,033	44,103,225
Arabic	1,184,731	7,823,043	6.603	59,044,611	26	2,296,556
Chinese	1,301,037	7,136,541	5.485	73,182,372	66	3,257,536
French	2,451,972	12,172,803	4.964	196,184,858	158	4,456,210
Portuguese	1,094,746	5,409,706	4.942	64,143,753	55	2,799,931
Spanish	1,799,230	7,720,543	4.291	145,274,893	63	6,628,458
Italian	1,770,143	7,504,894	4.240	128,815,061	122	2,288,917
Russian	1,850,168	7,425,382	4.013	124,938,187	76	3,239,213
Ukrainian	1,191,225	4,134,757	3.471	36,916,450	45	641,531
Japanese	1,340,684	3,961,526	2.955	91,074,155	41	1,973,889
German	2,720,855	7,513,224	2.761	224,347,581	190	3,986,231
Swedish	2,552,658	6,123,318	2.399	51,047,790	66	833,574
Polish	1,534,468	3,560,050	2.320	67,762,113	101	1,189,135
Waray-Waray	1,265,944	2,881,790	2.276	6,287,722	3	53,040
Dutch	2,100,091	4,470,660	2.129	62,554,195	36	1,227,293
Cebuano	6,125,805	11,232,244	1.834	34,883,660	6	93,899
Egyptian Arabic	1,597,655	2,013,586	1.260	7,347,014	7	189,801

Wakelamp d[@-@]b (talk) 03:09, 3 September 2022 (UTC)

Your list says "readers" and "edits", but the table you've copied from says "edits" and "editors" for those fields. ;) AddWittyNameHere 19:40, 4 September 2022 (UTC)

Thank-you. Sorry for the delay

I'd be more worried about the lower ratios. Those would seem to indicate that no one is discussing anything. Which might mean that articles, even ones that might attract serious disputes on the English site, are being posted with little or no oversight. How many of the articles on those sites are the work of only a single editor? And how many of those single editor articles are biased because of that? --User:Khajidha (talk) (contributions) 15:07, 15 October 2022 (UTC)

13 enWP Edit- 4 sharpish peaks per day

Does anyone know why the edit peaks three or 4 peaks per day in en (enlarged picture) ?

It looks like six-hourly maintenance runs by one of the users named here, possibly ST47ProxyBot. Certes (talk) 19:48, 1 October 2022 (UTC)

14 Topics on Article Talk And Project Talk pages

Just randomly looking through article talk topics, Start/Stub/C are never answered or acted on, about 10 % of those on better articles are acted on, or answered before being archived If an editor instead goes to the project based on the templates at the top of talk, then there is a high chance of the project being inactive. (aside, if a project is inactive who reviews Article Class?) Could we centralise all the low quality article talk, so they get an answer? Or run a chat bot? Wakelamp d[@-@]b (talk) 14:13, 15 October 2022 (UTC)

Do you mean that if someone posts a question on the talk page of a stub, they are unlikely to get an answer, but if they post a question on a B-class or GA's talk page, they are more likely to get an answer?

It might be more pointful to do this analysis against article page views and number of active page watchers than against assessed class. Whatamidoing (WMF) (talk) 18:10, 17 October 2022 (UTC)

You are right - Views or Watchers would be better, as I don't think the manual article classification process is working. Do views include webcrawelers and wiki bot non-edit processes? And do the correct number of watchers show on quarry, or is it rounded up? Wakelamp d[@-@]b (talk) 15:50, 17 November 2022 (UTC)|

Page views attempt, within reasonable limits, to exclude web crawlers. Proper bots shouldn't trigger page views either.

I don't know how Quarry handles page views. You could ask at Wikipedia:Request a query. The key point is that you want to see active watchers, because people who created an account, watchlisted a page, and then lost their passwords have basically zero likelihood of replying, but their watchlist is stored in perpetuity, so they'll still be considered "watching". See https://en.wikipedia.org/w/index.php?title=Wikipedia:WikiProject_Contents&action=info for one of the most extreme cases: almost two thousand "watchers", but only six (6) of them have visited Special:Watchlist in the last 30 days. Whatamidoing (WMF) (talk) 05:18, 18 November 2022 (UTC)

15 What is the question again?

I am seeing a lot of data compiled above… but it is not clear (to me) why we are being presented with all that data. Could you summarize the central question? Blueboar (talk) 15:49, 15 October 2022 (UTC)

The initial question was are number of edits causing an incentive to create pages in various namespaces that serve little purpose, which was based on people discussing how to incentivise article improvement quality. The question about article talk was also about editor retention. Hope that helps Wakelamp d[@-@]b (talk) 15:01, 17 October 2022 (UTC)

In addition, I am looking at this is because I am going through all WMF and media complaints about WP editors or community, and trying to work out a possible fix for review here.( And FYI, The order of preference for through removing functionality, error proofing and simplification, automation, aligning responsibility and benefits, quality assurance. WMF monitoring, and user complaints). But that's done now :-) Wakelamp d[@-@]b (talk) 04:21, 26 October 2022 (UTC)

16 New Space History

Should we declutter Wikipedia space by moving pages kept for historical reasons to a space called history?Wakelamp d[@-@]b (talk) 15:50, 17 November 2022 (UTC)

The last time I looked, the single biggest type of page in the Wikipedia: namespace was old deletion discussions. Whatamidoing (WMF) (talk) 05:20, 18 November 2022 (UTC)

99 List of what should become a proposal

Wakelamp d[@-@]b (talk) 15:50, 17 November 2022 (UTC)