Hosta

Better coverage of internet addiction related to Wikipedia

There was recently a thread about this somewhere but I can't find it. It's a bit of a shame WP:ADDICTED only really exists as a humorous page and there's very little helpful guidance or explanation of the topic. What do? What's the best way a serious essay could be written on the topic? — Ixtal ^{( T / C )} ⁂ Join WP:FINANCE! 22:33, 18 June 2022 (UTC)[reply]

There's tons of research on internet addiction. A quick search for reviews shows a lot of free pdfs. Also remember anything NIH-funded will have a free link. Could we be so lucky as to have research specifically on Wikipedia addiction? Click here to find out!

I've not written policy, but I imagine you could mock something up pretty quick from all this, similar to as you would write an article. I also imagine that it's best to keep it short and sweet. Everyone's heard of internet addiction, so maybe just give the rundown, note the research, and link related WP essays. As for me, I can quit anytime I want. SamuelRiv (talk) 18:10, 20 June 2022 (UTC)[reply]

Also remember anything NIH-funded will have a free link - feels like as good a time as any to recommend the excellent Unpaywall browser extension, too. I didn't find the article massively revealing, especially since its case report concerns itself with anormal reading rather than editing habits, and the four Cs of Wikipedia were also new to me. 5P I've heard about, but this link's red...

However, I agree with the gist of your comment, it would be well to have some kind of overview to link to. One problem that shouldn't be underappreciated, though, is designing such a page in a way that wouldn't lead to people routinely interpreting its linking as a personal attack. I imagine that many editors one could justifiably suspect of having a genuine addiction to this site are at least mildly controversial and used to having passive-agressive aspersions cast their way. Dr. Duh 🩺 (talk) 09:40, 21 June 2022 (UTC)[reply]

I think that any serious page would be an improvement over the current linking to a humor essay, so while the problem might still exist I think it will be minimized compared to the current state. Nonetheless, a very valid point to raise Doctor Duh. — Ixtal ^{( T / C )} ⁂ Join WP:FINANCE! 10:25, 21 June 2022 (UTC)[reply]

@Ixtal, were you looking for Wikipedia:Village pump (proposals)/Archive 189#We should find a way to increase awareness of internet addiction among Wikipedians ? WhatamIdoing (talk) 02:14, 23 June 2022 (UTC)[reply]

• We have an article on Internet addiction disorder. Remember that our purpose is to provide information, not to increase awareness. Blueboar (talk) 12:56, 23 June 2022 (UTC)[reply]

• Talking of Int. Addiction, here s a proposal of mine a couple of months ago.[1] Cinadon36 13:09, 23 June 2022 (UTC)[reply]

GA and QPQ

Should the WP:GA process contain something similar to the WP:QPQ nominations for Did you Know? This would certainly help reduce the backlog that is currently at WP:GAN while also possibly giving nominators the opportunity to branch out and gain experience in different subjects. Thoughts anyone? CollectiveSolidarity (talk) 02:52, 21 June 2022 (UTC)[reply]

CollectiveSolidarity, as someone who’s reviewed 30 GANs and submitted one article for GA, this is certainly an appealing idea. We should be careful that inadequate GAs don’t slip through the cracks though as a result. ~ 🦝 Shushugah (he/him • talk) 15:56, 21 June 2022 (UTC)[reply]

I'd say the backlog (which I admit sucks) is preferable to people unwillingly and incompetently doing quick reviews just to satisfy the QPQ requirement. There are not enough eyes on GA reviews currently; any initiative that makes people who are not good at reviewing do more reviews needs to have something that guarantees quality standards are upheld. —Kusma (talk) 16:07, 21 June 2022 (UTC)[reply]

We could place recent promotions on a list, where a user with a track record of page quality control can do a quick skim (refs, images, prose, etc.) and decline the promotion if there are any outstanding issues. Think of it as similar to WP:NPR. Acting as a member of this “promotion patrol” could also count as an automatic QPQ. Perhaps a bit grandiose, but it could increase the efficiency of the system by giving prolific creators the ability to spend less time QPQ reviewing and more time improving their own GAs. CollectiveSolidarity (talk) 17:07, 21 June 2022 (UTC)[reply]

• No. I love writing Good articles, but I have no interest in reviewing articles submitted by others. If I were required to review other articles in order to nominate another for Good status, I'd be less inclined to expand and nominate entries. ---Another Believer _(Talk) 17:12, 21 June 2022 (UTC)[reply]

Does anyone know why we show at WP:GAN how many reviews people have, but not how many GAs they have? (Can we shame people into doing more reviews??) —Kusma (talk) 17:26, 21 June 2022 (UTC)[reply]

Given the crap that routinely gets through at DYK, it doesn't seem to be doing anything beneficial over there and I fail to see how it would work better for GAs.--User:Khajidha (talk) (contributions) 17:33, 21 June 2022 (UTC)[reply]

I'd be pretty wary about that for a few reasons. First, GAs should be one of the primary goals of the encyclopedia. Unlike DYKs, GAs are long lasting, and provide in-depth, well sourced, and neutral information to the reader. Second, the quality of GA reviews will likely be reduced, because the two skills are not the same. I feel qualified to write articles well enough where they make it through GAN fairly smoothly, but I am certain I would miss a fair amount that should be taken care of if I were to review an article. Third, each editor only has a certain amount of time and effort they can devote to Wikipedia. If someone is, for instance, writing GAs and doing NPP, are we willing to say that to have any more articles promoted to GA they have to cut back on NPP? ScottishFinnishRadish (talk) 17:47, 21 June 2022 (UTC)[reply]

In the past, I think I suggested this myself, but I've come around to the position that QPQ for GA would be a mistake. I've seen enough substandard GA reviews with the current process. If we required QPQ, that would just drive quality in the wrong direction. Yes, the backlog sucks, but cleaning up the backlog at the expense of quality would be worse. -- RoySmith (talk) 18:52, 21 June 2022 (UTC)[reply]

PS, what I would really love to see is some better reviewing tools. There's lots of good code (ie software) review tools that let you click on a line of code, enter a comment, and get an easy-to-follow threaded conversation. The current GAR process of manually editing a review document, manually pinging people, manually copy-pasting text you want to refer to, etc, is an impediment. It's what the software world calls "friction". -- RoySmith (talk) 19:02, 21 June 2022 (UTC)[reply]

Maybe something that says "Please fix ... due to criteria _____", with an easy way to flag "By the way, these parts are all just my personal opinion, since Wikipedia:Good article criteria doesn't even mention them (e.g., reference formatting, categories, templates) or says I'm not supposed to decline articles on these grounds (e.g., dead links, 95% of the MOS, details that I'm curious about)." WhatamIdoing (talk) 02:50, 23 June 2022 (UTC)[reply]

Yeah, the above replies show that this idea is getting thrown in the Bit bucket. A better process for reviewing is badly needed, but I doubt that a solution will ever be worked out. Apologies for not coming up with a more developed concept. CollectiveSolidarity (talk) 20:14, 21 June 2022 (UTC)[reply]

PS, I am at least going to take the high road for this and try to review a nom for every one I make myself. It should be at least encouraged to pay it forward for all the reviewers out there who don’t get a Green blob for all their unsung work. CollectiveSolidarity (talk) 04:08, 22 June 2022 (UTC)[reply]

• I've reviewed 18 GAn and written about 60. I'm sympathetic to those who want to the clear the backlog and yes, partly for selfish reasons I don't want this rule, but there are more consequences to a shoddy GA review (e.g. bad contest masquerading as good content) than an ill-considered DYK. I've often thought to myself, "I should do more reviews" but look through the noms and not really feel qualified to review any up for consideration at that moment. An exception for the QPQ for the first few noms would also be necessary in this scenario, since undergoing only one GAn is, in my opinion, not sufficient to train someone to perform a GA review. -Indy beetle (talk) 16:32, 23 June 2022 (UTC)[reply]

Addendum to 10YT

The WP:10YEARTEST seems to miss an issue that is present on some current-events-related articles I've been scanning – where recent events far overshadow the coverage that was deemed most notable when written in the article several years ago. I borrow the trope name "Arson, murder, and jaywalking" (draft essay) to describe it (for now). It's difficult to address with bold edits because the fix necessarily means removing, condensing, and replacing a great deal of well-sourced content, often in articles about politically-hot topics. Therefore I think an addendum to the RECENTISM#10YT essay section is warranted.

In searching past VP discussion there is one on RECENT news media being UNDUE that makes a few relevant points to the topic, but it's a long thread and mostly unrelated. See also a somewhat-related counterargument essay and the notability subpolicy it counters. If the substance of this addendum is worthwhile I know the writing style, length, and title will have to change significantly. I am posting here prior to taking it to the essay's talk page because this is my first essay attempt, so feedback on all issues, including fundamentals, is welcome. Thanks all. SamuelRiv (talk) 21:11, 21 June 2022 (UTC)[reply]

@SamuelRiv, I'd suggest that you think about giving editors a path forward. You say they should think about whether the past matters. Maybe also suggest "do your best, and give grace to editors whose best guess differs from yours"? Maybe "mark your calendar and come back in a year to adjust things"? Sometimes it's obvious (we knew at the time that the most important thing that would ever happen to the Chernobyl Nuclear Power Plant was the Chernobyl disaster), but sometimes it's just not. WhatamIdoing (talk) 02:55, 23 June 2022 (UTC)[reply]

I don't think I understand. This draft would ideally be added to the RECENTISM essay following the 10YT section or as a subsection, and what you're talking about all seems covered in 10YT. The diplomatic caveats are usually at the beginning of the long-form essay (in this case, RECENTISM). Did I completely miss your point? SamuelRiv (talk) 04:00, 23 June 2022 (UTC)[reply]

Oh, I thought you wanted to create a new/separate page. If you want to add this to the existing one, then I encourage you to Wikipedia:Be bold. The worst possible outcome is that you will discover someone who is equally interested in improving the page. (You'll know who it is when they revert your bold addition.  ;-) WhatamIdoing (talk) 16:36, 23 June 2022 (UTC)[reply]

The entire WP:10YEARTEST probably needs to be better written too. The "comprehensive rewrites" paragraph already states that citations to breaking news reports written at the time of the event "could be replaced by those to more scholarly, historical, or retrospective references created later on". And the "Just wait and see" paragraph mentions "Editors writing today do not have a historical perspective on today's events" - and that's probably true with those that will be writing tomorrow's historical and retrospective references. Zzyzx11 (talk) 15:48, 24 June 2022 (UTC)[reply]

Vital biographies

Hello, on the contents page, there doesn't seem to be a way to view the most important biographies for Wikipedia except vital articles or list of articles every Wikipedia should have. We do a good job at covering the other major areas of Wikipedia and was hoping to develop some ideas besides the vital articles projects to figure out how we can implement the biographies in our contents page. Interstellarity (talk) 00:24, 23 June 2022 (UTC)[reply]

How do you decide which ones are the most important? WhatamIdoing (talk) 02:57, 23 June 2022 (UTC)[reply]

@WhatamIdoing: If you go into the talk page of the vital articles project, WT:VA and WT:VA4, there has been a lot of discussion regarding which biographies are most important. The list is here. That might give us an idea on where to start with including biographies in the contents page. Interstellarity (talk) 16:33, 23 June 2022 (UTC)[reply]

Don't those conversations usually result in editors concluding that it's very difficult to identify "important" articles, and that the best we can really expect is "important to me"? For example, the list you linked has 27 politicians. It includes three monarchs from England, but it doesn't have Winston Churchill. From France, it has Napoleon and Joan of Arc, but it doesn't have Louis XIV. From the US, it has Washington and Lincoln, but it doesn't have anyone from the 20th century. There is only one person listed with Central or South America heritage; what about José Francisco de San Martín? Bernardo O'Higgins? Cesar Chavez?

It would not be especially difficult to build a list of 27 leaders that matter to people from my country. It is very difficult, and perhaps impossible, to build a list of 27 leaders that are equally important to everyone in the world. A list that sounds fair and balanced to someone from the US or the UK will not seem at all fair and balanced to someone from Africa or Asia. WhatamIdoing (talk) 21:27, 24 June 2022 (UTC)[reply]

@WhatamIdoing: Would you recommend against adding biographies to the contents page based on your statement above? Perhaps we could have 500 or 1000 people articles. Interstellarity (talk) 12:59, 26 June 2022 (UTC)[reply]

I don't think it makes sense to have a "List of People who are More Important Than Other People". It would make sense to have a "List of Articles That Probably Need Improving Now" (e.g., a copy of Wikipedia:WikiProject Biography/Popular pages), but I'm not sure what the point is of deciding which people are permanently more important than others. WhatamIdoing (talk) 21:19, 26 June 2022 (UTC)[reply]

Is your thought then to take, say, VA3 People and give them all a little blurb and some categorization (like the Contents-Tech page) and have that be another Contents page? Blurbs shouldn't be too terrible to write or even script-generate from the short summary or the lead sentence or two. One potential issue of course is representation politics, but obviously the other Contents pages (and VA itself) have dealt with it. Off the top of my head you could have 100 names drawn a from a pool of VA4 People under regular rotation, which might mitigate some foreseeable complaints. SamuelRiv (talk) 17:39, 26 June 2022 (UTC)[reply]

Let's not use the phrase "convicted felon" as often in the start of a Wikipedia article, while still acknowledging and being clear about what actions they have taken and what courts of law have said

I don't edit Wikipedia often, and this is my first time posting to the village pump. However, I have a criticism about how some articles start in describing someone as a convicted felon. While this might be a true fact, I think it is on one hand vague and on the other hand stigmatizing.

There are things that people are, and there can be stigma associated with it. I don't know that Wikipedia has much of a mechanism to respond to this. In the United States for example, being a convicted felon is something that can be hard for people and prevent access to all sorts of social support and job support. The phrase convicted felon is rarely put into a hyperlink so we can know more about what that even means. Is it appropriate to come out and brand someone with as a convicted felon, especially if they are famous for other things?

I don't think it is right or just. But I don't know how Wikipedia deals with questions of rightness and justice. It is really great to try to be neutral and not necessarily be into something that is justice related. I wouldn't want people to not have stigmatizing truths about them on their page because it can hurt. But one thing I think Wikipedia does do is care about is being specific and relevant.

I don't know that "convicted felon" as a phrase means much. Is it unduly creating a negative perception in the mind of the reader? I don't know. But take the example of the wikipedia article for say Annie Dookhan. It starts like this, "Annie Dookhan (born 1977) is an American convicted felon who formerly worked as a chemist at the Massachusetts Department of Public Health Drug Abuse lab[1] and admitted to falsifying evidence affecting up to 34,000 cases.[2]"

Is the notable thing about here that she was convicted of the felony, or is it the falsifying evidence affecting up to 34,000 cases?

Wouldn't it make more sense to say, Annie Dookhan is an American chemist who formerly worked as a chemist at the Massachusetts Department of Public Health Drug Abuse lab who admitted to falsifying evidence affecting up to 34,000 cases. Dookhan was sentenced to three to five years of imprisonment and two years of probation by Judge Carol S. Ball in Suffolk Superior Court, after pleading guilty to crimes relating to falsifying drug tests.

I think this is more helpful because it says that she is notable for this thing, and it was something she was sentenced to prison for. Convicted felon doesn't mean that much. there are lots of ways to be a convicted felon. Why not be specific and more precise instead of using a somewhat load but ultimately more vague phrase like convicted felon?

Can I just edit articles for clarity if I see that, or is that not a good idea? — Preceding unsigned comment added by Hockeydogpizzapup (talk • contribs) 07:30, 23 June 2022 (UTC)[reply]

• These things should be decided on a case by case basis. In the Dookhan case, yes I think it's more informative to lead with "chemist" than with felon. Some people become notable (or otherwise very much-defined) by the crimes they commit. I do believe its better to substitute the word "felon" with more exact descriptive, for example saying "convicted fraudster" or "convicted murderer" to be more specific. Wikipedia is not concerned with American stigmas broadly speaking (or real world impact, per WP:CENSORSHIP), but yes, I think "felon" is often an unhelpful label which offers little in the way of information. -Indy beetle (talk) 16:23, 23 June 2022 (UTC)[reply]

  It might be better to put the "claim to fame" up front: "Annie Dookhan falsified evidence affecting up to 34,000 cases while working as a chemist at the Massachusetts Department of Public Health Drug Abuse lab". WhatamIdoing (talk) 16:38, 23 June 2022 (UTC)[reply]

I would agree that unless the absolutely only notable thing the person has done is for being a convicted criminal , this phrase should be avoided in the early part if the lede, favoring neutral terms of their occupation. How and why they were convicted likely still belong later in the lede. Masem (t) 16:53, 23 June 2022 (UTC)[reply]

You might find some arguments from a previous discussion on moralization in article leads interesting, though your point is probably better made on its own merits. You're right – "convicted felon" is pretty useless on its own. (To address EC above) Even if only the consequences of conviction are important, every jurisdiction has very different lifetime penalties for felons that have been changing over the decades, so specificity probably would still be essential in the lead. I'll start looking out for the phrase too. SamuelRiv (talk) 17:02, 23 June 2022 (UTC)[reply]

"Felon" on its own is not the best, though there are some people whose notability is simply that they are career criminals. For some people (like Bernie Madoff), it's easy to lump all of the sorts of crimes together by characterizing him as a "fraudster", but it might be harder for people like James Galante where there are 93 crimes that span a wildly large range of activity. I don't know else to lump together racketeering, racketeering conspiracy, Hobbs Act extortion, mail and wire fraud, witness tampering, tax fraud, and conspiracy to commit arson charges; surely we wouldn't start our articles off with "James Galante (born January 5, 1953) is an American racketeer, racketeering conspirator, extortionist, fraudster, witness tamperer, kidnapping and arson conspirator, and associate of the Genovese crime family" in the first sentence where "James Galante (born January 5, 1953) is an American convicted felon and associate of the Genovese crime family" works fine.

I do see problems in which this phrasing can be abused to attack living people who committed a felony like fifty years ago or as a child (provided that it's a relatively minor point in their biography), but also I don't really think that this phrasing is a problem at all for biographies on subjects who are notable principally for having committed a large number of felonious crimes. — Ⓜ️hawk10 (talk) 17:41, 23 June 2022 (UTC)[reply]

Easy: "James Galante is an American gangster/criminal/mobster (pick one), an associate of the Genovese crime family, and a real jerk." SamuelRiv (talk) 18:45, 23 June 2022 (UTC)[reply]

Doing such is a bad idea on several levels. First, using such as the "noun" that the person is is a huge POV choice. Second, putting it in the first sentence of the lead is also a huge POV decision. Third it can be a POV "spin up". The common meaning of the term is pretty bad / serious, but in various places (even limiting it to the US) adultery, throwing something at a bus station and cutting off more than half of a sheep's ear are felonies. North8000 (talk) 17:48, 23 June 2022 (UTC)[reply]

I agree with Hockeydogpizzapup. And I would add that people are rarely notable for having been convicted, but of having committed an offense. It's not as if for example no one knew who Annie Dookhan was until she was convicted of a felony. Also, the term felony is not commonly used in many countries outside the U.S. TFD (talk) 18:14, 23 June 2022 (UTC)[reply]

I like the idea in general that when using subjective or hostile terms, they should be spun up to explain why they apply. This is nearly always the case for successful/positive-labeled people, and should be same throughout. Masem (t) 18:58, 23 June 2022 (UTC)[reply]

The problem with trying to regulate "subjective" terms is that whether a term is subjective seems to be subjective. (Specifically, any word that I disagree with is merely the subjective opinion of whichever WP:BIASED sources you were reading.)

The problem with "hostile" is that sometimes hostility is actually warranted. Murderer, rapist, torturers, enslavers/human traffickers: there are people who will subjectively feel that these are all hostile terms and will try to tell us that it's far better to call them something "neutral". This is a kind of neutral that's unknown to the NPOV policy, but maybe we'll rename Category:Murderers to use the non-hostile term "non-consensual population control technicians". Or, you know, maybe we'll just Wikipedia:Call a spade a spade and stick with the "hostile" name. WhatamIdoing (talk) 21:43, 24 June 2022 (UTC)[reply]

Using the term "felon" seems to be too US centric to me: the term is rarely used outside USA now. For the majority of readers it will be difficult to understand what it means (except that this is something ugly). It should be replaced with "criminal" or dropped altogether from the lead. Ruslik_Zero 20:38, 23 June 2022 (UTC)[reply]

On this one count, 'convict' may be better, as this does not discriminate between US or any other terminology, e.g. 'murder convict' is clear to all I think. As to the prominence of such attributions, it must logically depend on the relevance to the biography. If the person is known primarily for their conviction, it should appear prominently, but if conviction is a minor part of their identity, not so much. Iskandar323 (talk) 06:55, 25 June 2022 (UTC)[reply]

Though I do see a problem with felony being a specific category of the crime, so perhaps it is appropriate in that the crime is specific to the US is outlined in US terminology. Iskandar323 (talk) 06:57, 25 June 2022 (UTC)[reply]

I think the phrase is problematic in its application. We could, accurately, say "Tim Allen is a convicted felon and actor..." Of course, we don't. What rule can be articulated to say where the cut-off is for convicted felon status to make it into the lede? BD2412 T 07:09, 25 June 2022 (UTC)[reply]

We do say Ted Bundy "was an American serial killer." That's because that is what he was notable for. And note we don't call him a convicted felon. TFD (talk) 13:40, 25 June 2022 (UTC)[reply]

Private RevDel requests

Right all, we've all known for a long time that it makes zero sense that there's a quick way of making an Oversight request (the email), but RevDels either need to be: requested where everyone can counterproductively see them; be made on IRC; or emailed to an individual admin.

Unlike OS, it's likely not the case that all RevDel requests should ultimately be made privately. But certainly a somewhat more professional structure would be a positive.

That said, there are multiple options with corresponding negatives and positives. There is also likely scope for discussion on what advice we should give for when/how to utilise it (must/should/can/don't). Please feel free to add options. Nosebagbear (talk) 18:19, 23 June 2022 (UTC)[reply]

Options

1. A VRT channel - have a queue that anyone en-wiki admin can request access to, and just add an email address that individuals can email requests to. Possibly with a set-up that enhances the odds of being given the information needed to carry out a revdel.
2. Something in the same vein as UTRS - a private structure designed to handle it, again with access provided on request.

Discussion

• Personally, I'd prefer the VRT channel. It has some negatives, but would be familiar to many, and more user-friendly when set against the status quo. Nosebagbear (talk) 18:19, 23 June 2022 (UTC)[reply]

  To be honest, I normally just email OS and mention that it probably needs a revdel, rather than OS, and let them handle it. May not be the best way, but it keeps from advertising the content that needs a revdel on-wiki. ScottishFinnishRadish (talk) 18:22, 23 June 2022 (UTC)[reply]

• Would certainly want to make sure any new queues aren't encouraged for RD's that don't require privacy (like old copyvio revisions that have been reverted). — xaosflux ^Talk 14:01, 24 June 2022 (UTC)[reply]

Idea: Bot that tags "problem" user talk pages with {{whois}}, {{Shared IP}}, etc.

Summary: Bot adds whatever {{whois}} related notice it think's is most appropriate to the user talk page of users who are editing in bad faith. I've already written a good chunk of the code, most of this is based on how that code works now or how I expect it to work when done, but it can be changed.

Goal: It's my understanding that one of the purposes of templates like {{Shared IP}} is to deter vandalism by communicating something like "You are not as anonymous as you might think you are, here is proof that we know who to contact if you do something really bad." especially the {{whois}} template. I'm not aware of any automation to the process of using these template other than Twinkle having a form that needs to be filled manually similar to what it does for warnings and welcome messages. Manually finding and filling the information is tedious and, to my knowledge, rarely done. Automating the process within certain limits may deter vandals, especially those editing from a place like a school where the IP may be registered to the school.

Suggestions on all variables, timeframes, edit numbers, etc. below are appreciated

Determination of users who are editing in bad faith: There are many things that can be checked to make this determination. The bot currently "decides" a user is editing in bad faith based on if all these criteria are met:

• The user received a standardised user warning template on their user talk page which follows the format uw-warningname# and where # is 3, 4, or 4im. This indicates that someone warned the user in a way that assumes bad faith per Wikipedia:WikiProject User warnings/Design guidelines#Severity levels.
  • The warning user is checked to ensure they have made at least min_warner_contribs in the last warner_contribs_timeframe. This check helps prevent the bot from affecting users who are being vandalized using warning templates and users who may have been warned wrongly by less experienced users. This has not been implemented in my code yet, I am leaning towards 100 edits in the last 30 days.
  • The user is not warning themselves.
• The user has made a contribution in the last 7 days. A check like this helps ensure that warnings are valid, suggestions on the timeframe are especially appreciated.
• The user has made less than max_user_contribs in the last user_max_contribs_timeframe. Users with substantial contributions are probably much less likely to be receiving a valid bad faith warning and are probably very unlikely to be vandalism only accounts. This has not been implemented in my code yet, I am again leaning towards 100 edits in the last 30 days. Total edits cannot be used universally because of shared IPs but IPs and accounts can be handled differently.
• The bot has not edited the page in the last 36 hrs. This prevents edit warring while being short enough that if a shared IP removes a template it can be put back.
• A template from Category:Shared IP header templates does not exist on the current version of the page. The bot trusts anyone before it, including itself, to have done a better or equally good job.
• The user talk page can be edited per exclusion-compliance. Anyone who knows enough to do this probably knows enough to know what the bot is trying to communicate anyway.

Actions on IPs editing in bad faith:

• Perform a lookup to attempt to determine who the IP is registered to.
• Attempt to classify the registered owner to determine the best Category:Shared IP header templates. This code hasn't been written yet, I don't know how successful it will be but it will likely be based of regex rules in the bot's userspace which will be editable by administrators. Most rules will likely be made to match a specific ISP which has been manually classified.
• Add the determined template to the top of the user talk page. If classification fails, use a generic template like {{whois}} or possibly one made specifically for use by the bot. This code hasn't been written yet.
• Based on feedback here and consensus if proposed, perform actions on potentially related IPs

Actions on potentially related IPs: Nothing in this section has been coded yet.

• Do some checks based on the registration information and classification, maybe skip anything to do with potentially related IPs if there's some reason to think that nearby IPs might not be accessible to current IP's user.
• Scan an expanding range of the original IP, adding each meeting certain criteria to a list, until the range size reaches the registered range, the range is unreasonably big, or the list contains a certain number of IPs.

Example: 192.168.4.65 is the IP, 192.168.0.0/16 is the registered range in CIDR notation. Start at 192.168.4.65/31, or a sensible range based on the maximum list size, scan the range for user that meet the criteria for inclusion in the list, then scan the IPs added to the range when you increase the range to 192.168.4.65/30 or whatever until you hit the maximum list size, get to your range limit, or get to 192.168.4.65/16.

Possible limits:

• Max assigned range to do any scan: /20
• Max IPs to scan: 256 (note that these do not have to be each scanned individually, you don't need to make 256 requests to check 256 IPs)
• Max IP on list: 25
• IPv6, you can scan a whole range in one request but unless there's some way to unify 2⁷²+ talk pages and post just one message that they'll all see it's pointless.

Possible criteria:

• Has made w contribution(s) in the last year
• Has made less than x contribution(s) in the last 30 days
• Has a user talk page with at least one warning of any severity on it (limiting to bad faith warnings would likely be redundant)
• y of z most recent contributions have been reverted

• Once the list is done, tag the user talk page of every IP on the list with either the same template used for the original IP or with a new customized template specifically for this.

Actions on registered accounts editing in bad faith: Nothing in this section has been coded yet. Currently there is a check that excludes talk pages of all registered accounts. Having similar capabilities to what is done with IPs would require a checkuser bot and hopefully that's enough information to explain why I think it'd be a bad idea. An alternative would be to create a generic template basically saying that it is possible to connect you back to your real identity, allowing for real world consequences if you do something bad enough, but that probably won't have the same effect as seeing the name of your school, workplace, ISP, or whatever. I think it'd be worth trying that if someone was willing to analyze edits from before and during a test period to see what impact it has. PHANTOMTECH (talk) 11:16, 24 June 2022 (UTC)[reply]

Discussion (bot)

• I haven't thought much about the rest of proposal, but I think that a bot that watches a page to handle manual requests would also be a good idea. I'm thinking of the bot going through an IP range's contributions in a time frame and tag individual IPs with the suggested template. 0xDeadbeef 13:17, 24 June 2022 (UTC)[reply]

  @0xDeadbeef Just to clarify are you talking about a page where users would give the bot an IP range and template + data the the bot would go through the range tagging all IPs that have made contributions within a certain timeframe using the user provided template + data? PHANTOMTECH (talk) 14:15, 24 June 2022 (UTC)[reply]

  Yes, and the bot could also have a feature to actively monitor new contributions from that range to insert templates and/or notify editors if needed. Let me know if this is not something we would want. :) 0xDeadbeef 14:32, 24 June 2022 (UTC)[reply]

  I like that idea, thanks :) I think temporarily monitoring the range (for a day to a few weeks depending on circumstances) could be beneficial. Permanent monitoring would require more resources over time and might end up giving notices to unrelated users that could confuse them without much justification since, with permanent monitoring, it could be months or years since the range had problems. PHANTOMTECH (talk) 14:46, 24 June 2022 (UTC)[reply]

  I'm interested in working on that idea if you haven't started. 0xDeadbeef 15:06, 24 June 2022 (UTC)[reply]

  I haven't started work on anything specific to the manually being able to request the bot tags and monitors a range thing. PHANTOMTECH (talk) 15:45, 24 June 2022 (UTC)[reply]

• Any such bot would quickly become defunct when meta:IP Editing: Privacy Enhancement and Abuse Mitigation rolls out. — Preceding unsigned comment added by Ahecht (talk • contribs) 14:43, 1 July 2022 (UTC)[reply]

New Wiki Product

Originally posted on Commons:Village pump/Proposals

Hello,

I write today as I have an idea for a new Wiki service. Called "Wiki Archives" its main purpose would be preserving artifacts like Documents and Photos that others might not seen in saving. Wikimedia Commons is a great service, but I believe that some of what's in it should be split into this new product. When you search for something, you can find pictures in like old building plans and old photos which should be in an archive where they could be appreciated more. It would work like commons as people could upload things, but it would be stricter in what it lets in (e.g. New phots taken by users and other things). If you have any questions, feel free to contact me. Thanks! DiscoA340 (talk) 19:04, 25 June 2022 (UTC)[reply]

@DiscoA340: This is outside the remit of English Wikipedia. Have you heard of meta:? That is the place to suggest that kind of thing. --Redrose64 🌹 (talk) 22:22, 25 June 2022 (UTC)[reply]

Is our collection of old images and documents that are already uploaded to Commons not being appreciated? There are a number of issues that splitting the exiting repository would create, so I would suggest a lot of thought to the questions and comments posted on Commons already before taking this to Meta. Imzadi 1979 → 18:59, 26 June 2022 (UTC)[reply]

Unless someone thinks otherwise, I do believe that these old images and documents are not being used to their fullest extent. Wikimedia Commons is mostly meant to provide free and useable images the public, not act a dedicated archive. Doing a search on WM for images about "Bladen County", you will find in useful images like maps and pictures of landmarks, but then you will find images like this which would be hard to use and is ironically in the National Archives. WikiArchives would provide that research tool to people who want to find images lake that. DiscoA340 (talk) 20:29, 26 June 2022 (UTC)[reply]

So why not put the image in c:Category:Bladen County, North Carolina or even c:Category:Elizabethtown, North Carolina? --Redrose64 🌹 (talk) 22:15, 26 June 2022 (UTC)[reply]

I don't see where anyone else has mentioned it yet, but Wikisource sounds like what you are looking for. ~ ONUnicorn^{(Talk|Contribs)}problem solving 20:33, 26 June 2022 (UTC)[reply]

• This sounds like an unnecessary fork of Commons. It's explicit that Commons is not just for hosting images that are used on Wikipedia, though they are supposed to be of some potential use to the public. What problems you're encountering can be resolved by better categorization, not the creation of a new website. -Indy beetle (talk) 13:32, 27 June 2022 (UTC)[reply]

Catalogue numbers in {{cite}} templates

Hi all, am I the only person who finds the layout of catalogue numbers (eg isbn, jstor, oclc, doi) a bit intrusive? I came across James Leasor#Bibliography, which is a good example of how they can dominate the screen. I find that small caps ISBN 9780552105866 are much less wearing on the eye than ISBN 9780552105866, and are the same size as a standard url link. I'm sure no-one would advocate url links this size. Would there be a case for incorporating this into the {{cite}} templates? I imagine it would be trivial to implement, but what do others think? I should mention that my prefs/gadgets include "Disable smaller font sizes of elements such as infoboxes, navboxes and reference lists", but there is still an inconsistency in the relative 'importance' of the information. Cheers, MinorProphet (talk) 06:03, 27 June 2022 (UTC)[reply]

I believe you want to direct this suggestion to the Help_talk:Citation_Style_1 page (this is the combined Talk page for the Citation Template). It's a good suggestion fwiw. SamuelRiv (talk) 06:49, 28 June 2022 (UTC)[reply]

Done, thanks for the tip. MinorProphet (talk) 10:06, 29 June 2022 (UTC)[reply]

Outstanding 'to-do' notifications (userscript solution)

As a follow-up to Wikipedia:Village pump (idea lab)/Archive 41#Outstanding 'to-do' notifications. (participants in previous thread: User:Kieronoldham, User:Xaosflux, User:Certes, User:DannyS712)
As I said in that thread a month ago, I have a monopoly on non-WMF echo notifications. (echo emulation) I created a reminder scheduler using that some days ago. To activate it: first install Bawl. (needed because the echo emulation is part of Bawl, you can disable Bawl's main features if you don't like them)
In the settings for Bawl, enable "Notification scheduler (Memoria)" at the bottom of the "Advanced" tab. After the next page load you'll have a "Memoria" PortletLink. The reminder stores its entries as a global user preference.
Recurring reminders are possible. Only recurring every n days is available. (so not every month/year, though you could approximate that with 30 or 365 days)
You can also subscribe to reminders. This could function as a kind of mini-newsletter, possibly useful for WikiProjects. Documentation has not yet been written but here's an example JSON. — Alexis Jazz (talk or ping me) 15:07, 30 June 2022 (UTC)[reply]

Talk pages, headings of new sections, and manually changing the edit summary so it says something else

Suppose I attack someone on their talk page like the following fake message where I poke fun at myself to do this demonstration

"NEWSANDEVENTSGUY you are a stinky nincompoop"

In the classic editor, that would show up in version history and my contribs as

NEWSANDEVENTSGUY you are a stinky nincompoop (new section)

By using visual editor I can manually put the section behind some civil camouflage simply by changing the proposed edit summary from the offensive section heading to something innocuous such as

Barnstar of Awesomeness (new section)

Anyone just doing a quick scan of my contribs will see this and won't realize I was a jerk to someone.

We'll never plug all the ways people can be jerks and get around rules, but that doesn't mean we shouldn't try. Would it make sense to turn off the ability to tweak the edit summary for newsections on talk pages? Yes, I can think of other ways to post insults but make contribs look pretty, too. Fact that OTHERSTUFFEXISTS really doesn't have relevance to pros/cons of this possible proposal. NewsAndEventsGuy (talk) 14:08, 1 July 2022 (UTC)[reply]

@NewsAndEventsGuy So you want to enforce putting personal attacks in edit summaries where they will be visible to everyone unless you get an administrator to do a revdel (as opposed to just on the page itself where anyone can blank it)? --Ahecht (TALK
PAGE) 14:38, 1 July 2022 (UTC)[reply]

I suppose that's a downside, but the overall idea is to increase the visibility of problem behaviors so its easier to establish patterns and get blocks that will reduce future problems. NewsAndEventsGuy (talk) 14:51, 1 July 2022 (UTC)[reply]

In the classic editor the "edit summary" box isn't exposed, the "subject" line is what gets put in to the edit summary. So if you use new section in the classic editor you can just put "Greetings" in the summary, then put your derogatory message in the body and it already has this same situation. — xaosflux ^Talk 14:39, 1 July 2022 (UTC)[reply]

This would change nothing. You can always just add a new section manually by editing the entire talk page. The text within /* */ is what makes the hyperlinked edit summary. Anarchyte (talk) 14:40, 1 July 2022 (UTC)[reply]

WP:BEANS, y'all. WhatamIdoing (talk) 01:44, 5 July 2022 (UTC)[reply]

Idea: 3RR Bot

A bot that detects potential 3RR problems could work, although the bot might need an advanced algorithm for ruling out self reverts. 0xDeadbeef 05:06, 5 July 2022 (UTC)[reply]

@0xDeadbeef We have User:ProcBot/EW which does what you describe, though it is currently inactive pending a move to toolforge. 192.76.8.85 (talk) 11:46, 5 July 2022 (UTC)[reply]

Let's not go on fishing expeditions for 3RR disputes. If things resolve without escalation, even if some policy was violated somewhere, that's a perfect case of WP:IAR. Headbomb {t · c · p · b} 05:32, 7 July 2022 (UTC)[reply]

Idea: using JavaScript to display protection icons

A few years ago I wrote a script that automatically would insert protection icons to protected pages. This would completely eliminate the need for the addition and removal of protection templates from an overwhelming majority of pages.

Something similar is already used on Wikidata because for technical reasons you cannot have protection icons on data items. Here, you can have protection icons on Wikipedia pages through the addition of templates (which is okay), but it takes time to add these icons to articles that are protected.

I want to suggest using JavaScript to insert the protection icons into the corner. This would allow for these icons to be added to pages without the need to insert protection templates like {{pp}} or {{protected page}}. It would also allow for the icons to display in the "edit" and "history" modes. Finally, it would allow for the icons to be displayed differently if a user has permissions to edit. In my implementation I have it that if a user has rights to edit or move the page, the protection icons display in the unlocked state. I think this would be helpful so that new users can understand intuitively which pages they are almost certainly able to edit and which ones they cannot edit. Aasim - Herrscher of Wikis 04:38, 7 July 2022 (UTC)[reply]

Awesome Aasim, I think having javascript to display these icons is great, but only when used in addition to protection templates. There are noscript users and bots out there and we can't ignore them. 0xDeadbeef 05:22, 7 July 2022 (UTC)[reply]

Well yeah. I see the templates still a bit needed for categorization of the different protected pages. The goal of this though is so new users can know which actions they can make on a page and which ones are restricted. And about noscript, unfortunately almost all of the web relies on JS to work correctly. Bots can also ascertain that a page is protected through the API, not by screen scraping. All of the protected titles can be viewed in Special:ProtectedPages and Special:ProtectedTitles anyway~ Aasim - Herrscher of Wikis 15:39, 7 July 2022 (UTC)[reply]

Having dynamic page content just to read can be a problem for all sorts of caching, additionally there are pages where for whatever reason we don't want to display an icon on the page. — xaosflux ^Talk 15:47, 7 July 2022 (UTC)[reply]

Using "Edit distance" instead of "Byte size difference"

Hi, in history pages like https://en.wikipedia.org/w/index.php?title=Main_Page&action=history there is a "Byte difference" in parentheses like

(cur-prev) 15:25, June 17, 2022‎ Cyberpower678 (talk - contribs‎) (2,964 bytes) (+3‎) Undid revision 1093586636 by Cyberpower678 (talk) Whoops. Tested on the wrong page (thank) (Tag: Undo)

Here (+3) denotes "Byte size difference of previous and current edition of this page". I really think that "Byte size difference" is wrong here, and we should show "Edit distance", for example by using Levenshtein distance. Suppose this scenario: someone only changes one word e.g., "on" to "in" in his edit. Here "Byte size difference" says that their difference is zero, because no change occurs in length of strings, but "Edit distance" shows the correct item to to the user, e.g. by Levenshtein distance the edit distance of "on" and "in" is equal to 1. So what should be shown to the users in parenthesis is the "edit distance" not "byte size distance" of editions. Thanks, Hooman Mallahzadeh (talk) 11:23, 7 July 2022 (UTC)[reply]

How do you propose this be reconciled with the page size value (unavoidably in bytes) right before it? It might be that the edit distance would be a more informative indicator in some cases, but it would also be very unintuitive to very many people, and certainly so with the way this information is currently being displayed. That is to say nothing about the technical feasibility of making such a change, which I'm not qualified to speak on. Dr. Duh 🩺 (talk) 11:36, 7 July 2022 (UTC)[reply]

It might not be perfect but it is only an indicator of a change. No matter what the value represents you will not be able to tell what the edit is by having a number (+2/-50) explain what a user has done. The difference here would mostly be for copy editing which I don't see an improvement of showing a +1 over a 0. Terasail^[✉️] 11:48, 7 July 2022 (UTC)[reply]

@Terasail For myself +1 (i.e., edit distance) is very more informative than 0. And I really think that (and hope) it would be more informative to others. Hooman Mallahzadeh (talk) 11:52, 7 July 2022 (UTC)[reply]

@Terasail Yes! A copy task makes "Byte size difference" equal to 0, but in such cases, "edit distance" should be equal to 1, and not bigger than 1, so implementation by Levenshtein distance is not a good choice for such scenarios. You are right! We should somehow modify the algorithm for such copy tasks, to return edit distance equal to 1. But for scenario of "in" and "on", Levenshtein distance is good enough. Hooman Mallahzadeh (talk) 13:16, 7 July 2022 (UTC)[reply]

@Doctor Duh You are right, for someone that does not know anything about edit distance, this data may be unintuitive. But edit distance is more informative for an expert. I really think that by using the idea of edit distance here, this concept becomes more frequent among people. I have an idea too. We can show both items, i.e., both "Byte size difference" for beginners and "edit distance" for experts, be shown here.

Technically edit distance is

The minimum number of operations required to transform one string into the other.

implementing that technically is not hard. Hooman Mallahzadeh (talk) 11:49, 7 July 2022 (UTC)[reply]

@Doctor Duh@Terasail Implementing for different scenarios like copy/paste scenario makes it more difficult, but I think number of scenarios is small and we can implement that. Hooman Mallahzadeh (talk) 13:24, 7 July 2022 (UTC)[reply]

@Doctor Duh @Terasail And I should note that "Byte size difference of editions is equal to -5", is so much ambiguous that nearly says nothing about what happened in the article, it says nothing about how much change have been occurred in the article. These are some probable scenarios:

1. a word omitted - in this case edit distance is equal to byte size change by a sign change
2. in article many changes occurred but size difference is equal to -5 - in this case edit distance is big
3. very small parts changed but size difference is equal to -5, - in this case edit distance is small but greater than 5
4. some copy/paste occurred and a word omitted - in this case, "edit distance" information is much more reliable

Among these scenarios, only in the first scenario, byte size has good information about what has been occurred in the article. Edit distance says how much (minimum) operations required to achieve the next edition of an article. Hooman Mallahzadeh (talk) 13:57, 7 July 2022 (UTC)[reply]

I find just remembering the bounds of a metric is often sufficient for practical intuition in this kind of thing. From the Levenshtein distance article:

• It is at least the difference of the sizes of the two strings. [Lower bound]
• It is at most the length of the longer string. [Upper bound]
• It is zero if and only if the strings are equal. [Null case]

Moving blocks of text wouldn't count, but if that's a desired feature it's an easy addendum to the code (appearing as "mv" text on the Watchlist or something). As for the issue of minor copy edits, they can still vary in byte difference, and that's what the "minor edit" flag is for. There is no non-linguistic algorithm that can determine if a "small" edit may actually be vandalism: it takes only 4 chars to do Jeffrey Dahmer was a jerk → Jeffrey Dahmer was not a jerk.

Of course my preference for quantitative metrics is determined first by the Rule of Cool (first search result is such a downer), but I imagine this idea would lose a lot more people if we brought in some stat mech spice. So regardless I'm on board 100%. SamuelRiv (talk) 14:10, 7 July 2022 (UTC)[reply]

• @Hooman Mallahzadeh: this idea has been proposed in phab:T10571 - I don't think this is anything that we can just do here on the English Wikipedia (having to run some script to rewrite every one of those values on each line for everyone is really unlikely to happen - but maybe you could do some initial trials with a personal user script) - and that it is something that would be best done server side. You can join in the discussion at T10571 to further develop it. — xaosflux ^Talk 14:06, 7 July 2022 (UTC)[reply]