Wikipedia:Village pump (idea lab) - Wikipedia


8 people in discussion

Article Images

The idea lab section of the village pump is a place where new ideas or suggestions on general Wikipedia issues can be incubated, for later submission for consensus discussion at Village pump (proposals). Try to be creative and positive when commenting on ideas.
Before creating a new section, please note:

Before commenting, note:

  • This page is not for consensus polling. Stalwart "Oppose" and "Support" comments generally have no place here. Instead, discuss ideas and suggest variations on them.
  • Wondering whether someone already had this idea? Search the archives below, and look through Wikipedia:Perennial proposals.

Discussions are automatically archived after remaining inactive for two weeks.

« Archives, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60

As a bot-author, link-rot saver, plain-text -> CS1 converter etc.. I am often aghast by the creative ways editors deploy nowiki and in the process create link rot, badly formed citations, and trip-wires for bots and scripts to snag on. Maybe it's just me, but I rarely see legitimate reasons for nowiki vs standard citation tools and templates. Suggest we could benefit from tracking categories and cleanup crews or tools for migrating these when it makes sense, which is probably at least 80% of the time. -- GreenC 18:29, 29 April 2021 (UTC)Reply

I actually use nowiki a lot when in combination with <code> (I also used it here) so that it doesn't actually turn into Wikitext formatting. Blaze The Wolf | Proud Furry and Wikipedia Editor (talk) 20:12, 29 April 2021 (UTC)Reply
The {{code}} template works too, is less typing and is itself pure wikitext — <code>. Plain text without the background is also possible using the {{codett}} template — <code> — GhostInTheMachine talk to me 21:09, 29 April 2021 (UTC)Reply
Example poor usage:
Last, First. Title. Publisher. <nowiki><http://example/com></nowiki>
(Note the surrounding <> which many tools pick up as being a part of the URL ie. http://example/com> ). They don't want the URL to be linked, yet to remain visible in full, sort of like paper. It's not a style exactly, typically found in pages mixed with other styles. Maybe it is a usurped domain (most cases not) - there are better systems for that too |url-status=usurped. It's rarely a good idea to nowiki in mainspace citations, there are better options. -- GreenC 21:21, 29 April 2021 (UTC)Reply
Why... why do people do that? Yeah, of course we shouldn't. Elli (talk | contribs) 23:37, 29 April 2021 (UTC)Reply
User:GreenC, some actual diffs of this would be useful, so we can see what's going on. I'm familiar with nowiki tags, but I've never seen them as having any connection with citations. Is this possibly something done by the visual editor, which I have often seen introducing unnecessary tags? Phil Bridger (talk) 21:47, 29 April 2021 (UTC)Reply
I've been under the impression that nowiki should be avoided in articles the same way directly invoking modules is. Is that wrong? I do think it'd be reasonable to track that, if we don't. Elli (talk | contribs) 23:37, 29 April 2021 (UTC)Reply
GreenC, as usual, WP:Visual Editor is to blame. [1] [2] [3] [4] — Alexis Jazz (talk or ping me) 02:45, 30 April 2021 (UTC)Reply
User:Alexis Jazz, Ack for type 1. This looks like an active bug as there are some old closed Phab tickets from 2015 that address a similar problem. Solution might be convert to "link", or simply plain text URL. It would be an easy enough search-replace, could be AWB. Guess first step is open a Phab and verify it's a bug (known or not). Do you know any further info? -- GreenC 03:38, 30 April 2021 (UTC)Reply
GreenC, here's type 2: [5] [6] [7]. As usual, VE is to blame. I use Wikiblame. For type 3, some uses are valid (like examples on pages about HTML) and 290 or so had empty nowiki tags. (nowikitag opened and immediately closed again) The only semi-valid use of that was on McCune–Reischauer and a related page where t' was written as ''t'<nowiki></nowiki>''. — Alexis Jazz (talk or ping me) 18:49, 30 April 2021 (UTC)Reply
Alexis Jazz. This got waylaid, just opened a Phab with your diffs: T282322 -- GreenC 19:40, 8 May 2021 (UTC)Reply

4 types:

  1. <nowiki>http = 6,432
  2. <nowiki>' = 8,715
  3. <nowiki>< = 705
  4. <http = 5,165

More to be discovered. -- GreenC 00:56, 30 April 2021 (UTC)Reply

(based on discussion [8] with Headbomb)

Headbomb created a great tool that highlights potentially unreliable sources based on community consensus (such as WP:RSP). Would it be possible to create warnings (through edit filters maybe?) when a user's edit contains one of those sources?

For example something saying: "The community regards this source as questionable (see discussion here). Please make sure the standards for reliable sourcing are being met or you edit might be reverted" so that inexperienced users might be helped in discovering the past consensus by the community?

It could be a great learning tool and a great way of saving a lot of time for a lot of users and improve article quality fast.

Even better if the warning was incorporated in the "cite tool" of the visual editor.

-- {{u|Gtoffoletto}}  talk 13:10, 4 May 2021 (UTC)Reply

We do have some warnings like that in Special:AbuseFilter/891. It's been a while since it's been expanded to cover more sources, but this one specifically deals with predatory sources. We could create other filters for different situations, but the threshold of crapness should be pretty high, not just "well.. it depends on a lot of things". YouTube is generally unreliable, but we wouldn't want to have a warning every time someone adds a YouTube link. Headbomb {t · c · p · b} 18:38, 4 May 2021 (UTC)Reply
There's also the problem that the filters don't highlight which source is unreliable. They have to page back and forth between a list of unreliable sources, and their added text, and try to figure which is the one. Suffusion of Yellow (talk) 19:24, 4 May 2021 (UTC)Reply
We do have a Wikipedia:Spam blacklist, and links from that list already get blocked. As for the filter warnings about unreliable sources, I think that any such warnings would have to be phrased rather conditionally. "You are trying to add a potentially unreliable source; please review it more closely" or something like that. Determining the reliability of a source requires a substantial degree of judgement and often depends on circumstances. Consensus regarding these issues is also complicated and should not be farmed out to bots or scripts. E.g. the outcomes listed at WP:RSP are themselves somewhat controversial, reflect varying degree of consensus, some of them are quite dated, some explicitly depend on how a source is used, etc. For sources that should never be used at all (I would put most junk predatory journals in that category), we should probably be trying to add them to Spam blacklist. Nsk92 (talk) 12:16, 5 May 2021 (UTC)Reply
@Gtoffoletto, IMO you should be saying "Headbomb created a great tool that highlights potentially unreliable sources". If reliability were as simple as 'anything from this website is always bad, no exceptions' then it would be much easier. However, that's not reality, and WP:RSCONTEXT still matters, even for a source that is "generally" (aka "not always") unreliable. WhatamIdoing (talk) 17:32, 5 May 2021 (UTC)Reply
I would agree with what WhatamIdoing says here. I'm grateful that on the odd occasion when I inadvertantly try to cite a predatory journal I get a warning message (I've no idea where from) but in general it takes human judgement to determine whether a particular source is reliable in context. We don't yet have enough advances in artificial intelligence to replace that. And, with the people here who develop code and templates seemingly being unable to cope with a few synonyms for parameters (something that was already old-hat in 1981 when I went into IT), we are unlikely on Wikipedia to even get close to what can be done elsewhere with AI. Phil Bridger (talk) 17:44, 5 May 2021 (UTC)Reply
Agree with all of the above. This is why this is just a warning and not a complete block from using those sources. The editor will be aware of the previous discussion and can then make an informed decision himself. -- {{u|Gtoffoletto}}  talk 17:54, 5 May 2021 (UTC)Reply
I'm not sure if that's true (when the Wikipedia:ABUSEFILTER is set to 'warn', lots of people just abandon their entire edit). Also, we have a problem with editors mindlessly removing 'bad' sources, so even if you know that the source is highly reliable for the particular point (e.g., Chinese state media on the official title of a Chinese politician) or actually authoritative (e.g., the original source of any quotation), there is still a risk that "your edit might be reverted" even if the source is being correctly and properly used. WhatamIdoing (talk) 21:01, 5 May 2021 (UTC)Reply

The actual reliability of a source is it's objectivity and expertise with respect to the text/item that the cite is supporting. I'd recommend advice like that in the notice rather than further entrenching the over-generalization that a particular source is always reliable or always unreliable. North8000 (talk) 21:13, 5 May 2021 (UTC)Reply

I would oppose this proposal. It would only put new Wikipedians off making edits to Wikipedia. Rollo August (talk) 15:28, 8 May 2021 (UTC)Reply

The proposal has merit... but only if we're highly selective in which sources we warn in the edit filter. Everything deprecated at WP:RSPSOURCES would be fine to warn against, I feel. Below that (generally unreliable), we might want to log, but not to warn. Headbomb {t · c · p · b} 16:05, 8 May 2021 (UTC)Reply
@Rollo August: on the countary I feel it would help new Wikipedians understand how Wikipedia works and have a more positive experience. Being reverted is never fun. -- {{u|Gtoffoletto}}  talk 09:06, 13 May 2021 (UTC)Reply
@Headbomb: I agree. A simple example: the amount of time we waste on medical articles removing arXiv sources is staggering... and very often those are missed and remain in articles with unreliable information. Warning even just for deprecated sources would reduce this issue significantly. Do you think the best implementation would be an edit filter? I have no experience with those tools. -- {{u|Gtoffoletto}}  talk 07:58, 14 May 2021 (UTC)Reply
But that's a bit the thing. Much like YouTube links, arXiv links are used for many reasons other than MEDRS claims. Giving an edit warning for it would be too much, because the majority of arxiv citations are relatively unproblematic. Unlike a citation to say, the Journal of Obvious Nonsense. Headbomb {t · c · p · b} 19:50, 14 May 2021 (UTC)Reply
One of my favorite journals, along with The Journal of Obvious Results. Levivich harass/hound 20:11, 14 May 2021 (UTC)Reply
@Headbomb: Non peer reviewed articles are considered non-RS in all areas don't they? I see the only "exception" is for papers by proven SMEs. But that's why we warn and don't block. I mean: 99% of the times an Arxiv source will need to be rejected. The risk of a false negative (proper use being warned) is pretty low here and if you can make that determination (this is an SME so it's fine) then you can probably read and understand the warning. On the other hand we have a very high rate of true positives (99% of the time the arxiv link is inappropriate) and would be helping out new users and avoiding time waste by experienced editors. Normal editors don't know if a source has been discussed in the past (and should have to!). We can help them out and give them the contextual information to learn and discover past consensus rather than letting them "make a mistake by design" and cause the community to collectively waste time. Also: nobody likes being reverted. What would be really great is if we could give to the users the exact entry into the WP:RSP table. That would be very helpful but probably not technically doable with edit filters. -- {{u|Gtoffoletto}}  talk 12:57, 17 May 2021 (UTC)Reply
"99% of the time the arxiv link is inappropriate" Really no. You can source plenty of uncontroversial statements to the arxiv. And a good chunk of the time, the citation is to papers like arXiv:1309.0102, which really is a lazy way of citing doi:10.1063/1.4869589. Headbomb {t · c · p · b} 19:10, 17 May 2021 (UTC)Reply
  • Oppose Adding citations to Wikipedia is difficult. If you make it even more difficult then users will tend to add the same content without the citation. Also, it is well understood that Wikipedia is itself not a reliable source and that's because it is written by literally anyone with an axe to grind. And that especially includes the determinations made at WP:RSP. Andrew🐉(talk) 08:22, 14 May 2021 (UTC)Reply
  • I would generally oppose this. There are quite a few museums out there that use blogspot (Wikipedia already flags up a warning message) and YouTube as channels for relaying reliable info/evidence. I think how we are is good enough, and people power is the best way to check refs.Davidstewartharvey (talk) 14:09, 17 May 2021 (UTC)Reply

This namespace is quite useless. You can't actually get to read one of these books without either buying it from PediaPress or using the volunteer maintained tool MediaWiki2Latex. While it's great that MediaWiki2Latex exists it takes hours to actually get the book if at all and if the book is excessivley long like most our books you will have to download MediaWiki2Latex, but since that is only available for Linux with support at German WikiBooks that is a quite grueling process. And then after you've gone through all that you may realize the book contains don't actually work as intended due to a navbox someone slapped onto the book or errors after hours of rendering. In practice the pages are just simple lists of articles.

In short the user experience is absolutely terrible. Books in the book namespace are also called "Community maintained" and the namespace is supposedly reader facing (although that is debatable after the systematic removal of links from templates and mainspace). The namespace contains a ton of material that seems to be individual editors testing out the book creator of very low quality and much of the rest isn't feasible to print or export as a PDF due to it simply being too large.

The pageviews are also absolutley abysmal which can be seen with the massviews analysis tool. The graphs are completley wack due to the links being resently removed, one book (Book:Full Form) ocassionally getting many times more views than the rest of the namespace combined and some editors (me included) opening a large number of books to assess the status of the namespace or try to make them less bad. I think it's fair to say that when the views stablize the views for the entire namespace will be under 500 a day spread out over 6000 books. The majority of books don't get a single view a month and very few people would miss them.

We also have PediaPress to consider which prints some Wikipedia books. They link to some of our books and it would be courteous to at least contact them about any changes that happens.

What can we do about this then? There are a lot of options here ranging from absolutely nothing since basically no one sees it any way to removing the namespace entirely (probably with a refund system and keeping a handful of books in some other namespace for PediaPress). Other options would include something like disabling saving new books in the book namespace using Special:Book (which is a simple configuration option), userfication of most books, large scale deletion or protection of the entire namespace.

I'm really unsure where I stand on this question and have been flipping back and forth. There is a non-zero value in the books for a handful of users but there is also a substantial negative in leading users to broken material they can't use. I think I would like to see us officially dropping support for community books at least with no saving new books there, no more WikiProject tagging, no more "community maintained" language in documentation pages and so on. Probably followed by a likely very controversial mass MfD of unused books which comprise most of the namespace for possible soft deletion. I wouldn't be opposed to going further either though.

What do the pump think? I'm like I said quite conflicted on what to do here. --Trialpears (talk) 14:33, 6 May 2021 (UTC)Reply

Trialpears, I don't get it. I looked at Book:Full Form, Book:Fringe (season 2) and Book:The Funk Brothers. When I pick "Wikitext in "Edit this book: Book Creator · Wikitext", there's nothing but a list of articles. For books, we have b:Main page, right? What even is this Book: namespace? I didn't even know it existed. — Alexis Jazz (talk or ping me) 15:06, 6 May 2021 (UTC)Reply
Alexis Jazz At this point it is basically just lists of articles. There used to be a simple way to download these but due to bugs and security issues this was removed in 2017. No replacement is planned and the prospects of it ever working again is very low. There is like I said above a workaround which isn't great and the possibility to order a printed book consisting of these articles. --Trialpears (talk) 15:46, 6 May 2021 (UTC)Reply
Trialpears, inform the printing company, freeze the business for 3 months, create a new username for this and move the business to the userspace of that user in case someone has a use for these lists and to avoid a potentially controversial XfD, kill the zombie. In that order. https://mediawiki2latex.wmflabs.org/ doesn't seem to care about namespace, I copied Full Form to my userspace and I was able to run the tool. When it finished it asked me to click some arrow in the corner of my browser.. no idea what it's talking about, so I never got the PDF. With Full Form in the Book: namespace, it did exactly the same thing. Looking at the HTML, it seems to try to redirect to https://mediawiki2latex.wmflabs.org/file/1230.pdf (number is a job number so varies) but that just loads the home page of the project. — Alexis Jazz (talk or ping me) 17:27, 6 May 2021 (UTC)Reply
  • Amazing! I turned on the book gizmo, added some articles to my book, made book chapters, shuffled the articles into chapters, saved the book, previewed it at Pediapress and could have ordered a hardback copy delivered to me in the UK for about 17 GBP or 24 GBP printed in colour. Instead, I then used the MediaWiki2LaTeX magic to create a PDF. It churned away for over 30 minutes and then spat out a 42Mb PDF with my 7 articles presented nicely, pictures tidily resized to fit the pages, all of the links as footnotes, a full set of the references and a full list of contributors – all spread over 244 pages. Lovely. Impressive stuff. Awesome technology — GhostInTheMachine talk to me 18:30, 6 May 2021 (UTC)Reply
    — And arguably quite pointless. Wikipedia is on the Interwebbly. Why do I want to make a book of frozen pages? However, not everybody has a live network connection like I have and some people may want a hardcopy and it does work, so why worry about it? How much is the namespace costing us? — GhostInTheMachine talk to me 18:39, 6 May 2021 (UTC)Reply
    I'm glad you liked it! Experiences like these are what makes it difficult to deal with the book namespace. Books can work well but more often than not the experience is something more akin to Alexis above. It is worth noting that what you did (starting the book creator at Special:Book, creating a user page storing the book and using MediaWiki2LaTeX and PediaPress to render it) will still be possible if the Book namespace goes away. It would perhaps not be possible to go to Book:Elements and do the same with that book. I doubt that is much of a loss since that page is like most other books in the namespace not that great (what happened with groups 7 and 8, why is Mixture included, what's up with the stray hyphen and so on). Books that people wished to be retained would of course be refundable to userspace or potentially somewhere in Wikipedia space.
    There really isn't that large cost to retaining the namespace, my main concern is simply that it's a very bad user experience. Not being able to download you PDF after a long rendering session, or buy books where the reference section is "<templatestyles src="Reflist/style.css">" isn't great.
    Finally it's worth pointing out that we do have a download as PDF feature in the sidebar which is good at dealing with PDFs of single pages. --Trialpears (talk) 19:45, 6 May 2021 (UTC)Reply
    • Not sure that I exactly liked it, although the technology did seem rather cute. When I tried the book gizmo, it did not seem to allow me to save the book directly to my user space. I made the copy from the Book namespace myself. If the resulting plan is to kill off the Book namespace, then the save option needs to be checked and possibly changed/fixed — GhostInTheMachine talk to me 20:52, 6 May 2021 (UTC)Reply
      I don't think it likes subpages which it should be able to handle. Did you try to use a / in your book title? I'll file a phab ticket shortly if so. --Trialpears (talk) 21:00, 6 May 2021 (UTC)Reply
      GhostInTheMachine I am not able to reproduce this. I noticed the save book button doesn't light up when you enter something in the box but rather when you press away from the box. I remember that I had mild problems saving a user book once though possibly because the button didn't light up as expected. Could you try it again and see if you encounter any issue again? I'm not gonna report the graphical issue since the extension is only bugfixed maintained. --Trialpears (talk) 20:39, 8 May 2021 (UTC)Reply
    GhostInTheMachine, However, not everybody has a live network connection like I have and some people may want a hardcopy Try Kiwix? — Alexis Jazz (talk or ping me) 21:07, 6 May 2021 (UTC)Reply
    Yes, and the Internet-in-a-Box project, but even a limited real paper hardcopy still has advantages for some situations — GhostInTheMachine talk to me 21:42, 6 May 2021 (UTC)Reply
  • Wikipedia is good at trying out experiments with new ideas, but we are not good at cutting our losses and retiring those ideas that don't work out (see also: Simple English Wikipedia). The maintenance burden of books is greater than any benefit it will ever bring us in its current manifestation. {{u|Sdkb}} talk 09:21, 7 May 2021 (UTC)Reply
  • I would fully support retiring the Book namespace and any associated tech that we have. If removing the namespace would cause issues or be a waste of dev time (probably!), then I'd support deletion of all pages in the namespace and preventing the creation of any new ones. They serve precisely zero purpose any longer. ƒirefly ( t · c ) 15:01, 7 May 2021 (UTC)Reply
  • "In short the user experience is absolutely terrible." Indeed. The solution here is to improve it, not to kill it. Headbomb {t · c · p · b} 20:58, 8 May 2021 (UTC)Reply
    That would be ideal, but I can't see that happening. The extension is under bugfix only support and there have been no developer intrest for years and I don't see any developer prioritizing it in the future. I don't blame them either: The entire namespace has about the same amount of views as a popular portal like South Africa. Whether it's better or worse to keep or delete something with this bad a user experience is up to opinion but I don't feel like the issues are solvable without significant amounts of developer attention which would be much better spent on any item of the community wish list. --Trialpears (talk) 21:13, 8 May 2021 (UTC)Reply
Chicken and egg issue. There were a lot more view before {{Wikipedia books}} got deleted. Fix book rendering, {{Wikipedia books}} can be restored, and the views will skyrocket. Headbomb {t · c · p · b} 00:00, 9 May 2021 (UTC)Reply
  • Other publishers make books out of Wikipedia content so it's reasonable for us to have an official way of doing this. Reasons one might do this include:
  • Commemorative or festschrift works such as the collected writing of SlimVirgin.
  • Collections of WikiEd or WIR type activity such as the ongoing Coventrypedia.
  • Collections of a body of work under attack by deletionists
  • Thematic work such as articles written during and about the COVID-19 pandemic.
Andrew🐉(talk) 08:43, 14 May 2021 (UTC)Reply
There are definitely use cases for printing Wikipedia books. Regardless what happens with the Book: namespace it will still be possible to use the Book creator and save said books in user space (or move them to Wikipedia: space). I also wouldn't feel that it's appropriate to keep most of your suggestions in a reader facing namespace which the Book: namespace supposedly is which would make user or Wikipedia space more appropriate. --Trialpears (talk) 13:47, 14 May 2021 (UTC)Reply

how about a new namespace, called "Community:"? pages here would be for specific community ventures, efforts, organizations, etc. within specific geographical communities, neighborhoods, etc. .

in creating this name space, we could use the "subpage" feature to our advantage; namely, each "community" would have a single main page as a hub or central node; then, all pages for individual groups, items, issues within that community would be subpages of the main hub page.

Please note, items within this namespace would be subject to the same criteria as regular entries, as entailed by WP:notability, WP:RS, etc etc. it is simply a way to encourage, induce and motivate greater editing activity, within various communities that may perhaps need greater coverage than they have currently.

I know this is a long shot, as is any proposal for a new namespace. However, i wanted to post it here, at the Idea page, just to get some overall feedback, input, brainstorming, etc. I welcome all comments. thanks! ---Sm8900 (talk) 🌍 20:05, 7 May 2021 (UTC)Reply

We already do this – see WP:PROJECT. Geographical communities are also supported by chapters and language versions. Andrew🐉(talk) 08:50, 14 May 2021 (UTC)Reply

 

I made a HistoryHelper userscript. Am I at right place to request a feedback? In short the script can extract diffs from «View history» and «User contributions» pages» into a wiki {{diff2}} tags. -AXONOV (talk) ⚑ 21:07, 12 May 2021 (UTC)Reply

User:Alexander_Davronov, looks cool, installed. Any problems will let you know. -- GreenC 18:30, 15 May 2021 (UTC)Reply

Would it be possible to create a bot that updates billionaires' net worths based off of Bloomberg's index? ~ HAL333 05:36, 15 May 2021 (UTC)Reply

@HAL333: Is there an easy to use, free API for that data? SQLQuery me! 06:40, 15 May 2021 (UTC)Reply
Not sure... ~ HAL333 17:02, 15 May 2021 (UTC)Reply
It's a subscription site and this list (top 500) is one of their premier products behind a paywall - are they giving it away for free to be kept current on Wikipedia with an open source license? -- GreenC 17:23, 15 May 2021 (UTC)Reply
I'm not sure whether any source can be said to be reliable when it comes to net worth. I couldn't tell you what my net worth is (my main assets are about 95% of a house (the mortgage will be paid off next year) which I share with my wife and a pension pot) and nobody's net worth is actually determined unless it is all held as cash. Assets such as property and shareholdings only have a specific value when they are sold, and in the case of large shareholdings they often cannot be sold for the current value determined by the stock market. And that is all ignoring the fact that nobody's worth depends on what they own. Phil Bridger (talk) 17:42, 15 May 2021 (UTC)Reply
I agree with Phil; I think there would be V problems with that: a billionaire's net worth is only an estimate; estimates vary; and it changes every second (at least when stock markets are open). I feel that including net worth in an infobox is problematic for the same reason: whatever number is listed is always inaccurate; it's an example of false precision. Better to just state whether the individual is on a list, like Forbes 500. It's not really accurate to say, eg, that Gates's net worth is $150 billion (or whatever), because that can swing by a huge amount (sometimes an entire order of magnitude), as it did last year. It's better to say something like "Gates is consistently ranked in the top 5 richest people in the world, with his net worth estimated from $XXX billion to $YYY billion," than to put a single number on it. (Wikipedia should not, for example, state who is the richest person in the world, because it can't be said with any accuracy.) Levivich harass/hound 17:47, 15 May 2021 (UTC)Reply
@HAL333: ideally, we'd have that being updated in Wikidata, though we have a lot of editors opposed to including data from Wikidata in articles. Elli (talk | contribs) 09:58, 17 May 2021 (UTC)Reply
I appreciate the responses. ~ HAL333 22:41, 18 May 2021 (UTC)Reply

The idea is to make a section in wikipedia that is dedicated to welcome information. Example: an english speaking person is coming to live in Modena/Italy. He/she will have plenty of questions concerning schools, administrations, doctors, restaurants, sports etc. In the wikiyournewtown she/he will find the information they are looking for. — Preceding unsigned comment added by 2001:861:3204:B40:18B5:FDD0:36BE:26B9 (talk • contribs) 06:46, 15 May 2021 (UTC)Reply

There is a separate Wikimedia project called WikiVoyage for tourism. Also see WP:NOTGUIDE. Sungodtemple (talk) 01:46, 16 May 2021 (UTC)Reply

Sometimes I add "(fair use)" to filenames. Generally when it's either shadowing a file on Commons or it's foreseeable that a file on Commons may get shadowed by it in the future. Examples: File:Lego Worlds (fair use).jpg because an unrelated file named "Lego Worlds.jpg" could be uploaded to Commons. File:Denis Akiyama (fair use).jpg because "Denis Akiyama.jpg" could be uploaded to Commons for various reasons that don't have to invalidate our local picture. File:Mickey Mouse (fair use).png because c:Category:Mickey Mouse isn't empty and the copyright on the little vermin is expected to expire in a few years which will result in more uploads to Commons, but that won't invalidate our fair use picture because 1928 Mickey looks nothing like current Mickey.

There is another reason to add "(fair use)" to filenames: it's the only thing that is sure to survive when third parties hotlink it and survive in some cases when a site caches the image. For example, on Everipedia or Wikivisually or WikiZero, how can you tell which images are free and which are fair use?

I'd prefer to always add "(fair use)" when a file needs to be renamed anyway (I wouldn't rename any files just to add it), but I wonder how others look at this. I'm not sure anything about this could or should be written into policy, but perhaps we could make a recommendation for file names. For fair use media, "Article title[, section title] (fair use).jpg" seems to work generally, never clashes with Commons, doesn't require subsequent renames if a movie poster is overwritten with cover art, or if a screenshot of a character from one game is overwritten with a screenshot of the same character from another game. I see both happening every now and then, needing a rename because the filename specified the game or medium. — Alexis Jazz (talk or ping me) 11:38, 17 May 2021 (UTC)Reply

@Alexis Jazz: well, eventually, all fair use media will be copied to Commons when the copyright expires. So for long-term naming, I'm not so sure about including "(fair use)" in the name (more relevantly, the file could also be voluntarily re-licensed). Elli (talk | contribs) 12:44, 17 May 2021 (UTC)Reply
Elli, those are some valid points. But getting re-licensed is fairly uncommon and entering the public domain takes decades (if not well over a century) for most files. And when a file gets copied to Commons, the hotlink URL will change anyway so there is not much harm in changing the filename when copying to Commons, I think. — Alexis Jazz (talk or ping me) 03:40, 18 May 2021 (UTC)Reply
@Alexis Jazz: fair. I would be in favor of establishing an actual naming convention for certain NFCC criteria - and renaming all files we currently host to match - but I think if we do that well, we won't run the risk of being confused with Commons (for example, if we did, "Book Author Year (cover)" I don't think we'd ever get a naming conflict). Elli (talk | contribs) 05:09, 18 May 2021 (UTC)Reply
Elli, I generally prefer not to specify "cover", "poster" etc in the filename because a fair amount of rename requests involves overwrites that replace one type with another or posters that were incorrectly labeled as covers and things like that. Renaming all currently hosted NFC files will be a Herculean task, but I would support it if we can hammer out a good naming convention. — Alexis Jazz (talk or ping me) 06:01, 18 May 2021 (UTC)Reply
@Alexis Jazz: sure, I wasn't suggesting that be the naming convention. Just gave an example of something I think wouldn't clash.
And I agree - a Herculean task - but one probably worthwhile for improved categorization, clarity, etc. Elli (talk | contribs) 06:03, 18 May 2021 (UTC)Reply
Elli, I just thought of another rationale for including "(fair use)" in the filename: it would allow the creation of an edit filter to prevent those files from being used outside article namespace by matching \(fair[_ ]use\)\.[wgjps] or something like that. We currently rely solely on bots (and highly observant users) to revert such inclusions afterwards. I remember some recent drama that could have been prevented had such a filter existed. — Alexis Jazz (talk or ping me) 02:31, 23 May 2021 (UTC)Reply
@Alexis Jazz: interesting thought, but still not sure if it would be worth it overall. Elli (talk | contribs) 02:33, 23 May 2021 (UTC)Reply
Elli, tools to export files to Commons could be adjusted to remove "(fair use)" automatically from the filename. (should be a rather trivial adjustment) Just saying, maybe ways can be found to address any remaining concerns. — Alexis Jazz (talk or ping me) 02:49, 23 May 2021 (UTC)Reply
@Alexis Jazz: fair, I wouldn't be too opposed. I think a non-free content general naming convention would be more useful, though. Elli (talk | contribs) 02:52, 23 May 2021 (UTC)Reply
Elli, I fully agree, but these ideas are not mutually exclusive. I'd prefer to include "(fair use)" in a naming convention.   — Alexis Jazz (talk or ping me) 02:56, 23 May 2021 (UTC)Reply

I may be wrong, but in 20 years, Wikipedia's design has not changed much. The design was great for the online innovations of 2001, but... not so great now. Websites nowadays have simple, user-friendly buttons, accounts systems that have cross-platform "read later" lists (or better yet, folders too), personalised feeds, etc. Wikipedia still has a fairly limited UI, while the rest of the internet has moved on. The app has some modern intergrations, but these don't sync with the website so it's still limited in scope.

Now obviously a "For You" feed would include countless data and privacy concerns but surely there could be some sort of opt in system for a user-data driven, modern UI? Like a personalised feed that simply recommends new articles based on ones you have read would do wonders for this. Is there any reason why this can't happen? Squid45 (talk) 17:38, 17 May 2021 (UTC)Reply

In your user preferences, you can change your skin and even create your own. There is even a list of cool skins. 'Read later' lists can be bookmarked in-browser. And note that Wikipedia is an encyclopedia, not a social media website or dictionary. A personalized feed would defeat the purpose of Wikipedia. Sungodtemple (talk) 17:53, 17 May 2021 (UTC)Reply
One of the attractions of Wikipedia is that it doesn't make silly suggestions of what we might like in the way that many other web sites do. It is a way of selling you more or telling you what you should think rather than anything that benefits the reader. If that's modern, then I'm proud to be old-fashioned. Phil Bridger (talk) 18:10, 17 May 2021 (UTC)Reply

Can I also say that one of the positive attributes of Wikipedia is that, unlike many other websites, it does not ask us whether we want to accept cookies? Rollo August (talk) 20:12, 17 May 2021 (UTC)Reply

I'm really not too keen on the idea of a "For You" feed, or user data driven navigation. For starters there's the privacy issues alluded to in the op. Tracking and storing user movement around the site so that an algorithm can guess what readers might be interested in would be a massive divergence from current policy where even stuff like checkuser data is only kept for 60 days for user privacy, and if it's an opt-in system you'll never get enough data to produce meaningful results (most of our readers do not have accounts, and I imagine only a fraction of registered users would be happy with the WMF tracking their reading). How do you propose to separate user data from reading from user data from editing? Any tracking of where our registered users go is going to be massively polluted with nonsensical page jumps from things like Auto wiki browser, new page patrol, anti-vandalism work, sorting through cleanup categories, etc. Fundamentally though the idea of driving user navigation by page views should be unnecessary - navigation should be built into articles. If a paragraph mentions something that a reader might be interested in it should be a wikilink, if there's a series of articles on a topic that make sense to read together they should be presented together in a navbox, if there are articles that are a natural follow on from the article a reader has just read they should be listed in the "see also" section, and all articles should be sorted into appropriate categories to make it easy for editors to find related content. We also have a huge number of volunteers working on maintaining the front page as a showcase of high quality content from around the site, which has enough variety to serve\ well as a list of recommendations for things readers might be interested in reading. 192.76.8.91 (talk) 22:41, 17 May 2021 (UTC)Reply
“ The term "modern" makes sense from a descriptive perspective – for instance, solely to distinguish newer periods and ways from older ones.

It becomes a problem when it carries the presumption that older ways are "old-fashioned"; that is, presumed to be worse. In technological fields where businesses make the choices and impose them on users, the reverse is often true.

”
— Richard Stallman

When the "modern user" gets used to something which is 'convenient' but not actually in their interests and asks us to implement it too, our best option is to politely say "We won't give you that, but we're not refusing out of malice; here are the reasons why this change is actually not in your interest". It's a good opportunity to try to teach people to respect themselves, actually. DesertPipeline (talk) 05:53, 18 May 2021 (UTC)Reply

I have an idea I'd like to suggest. So recently, InedibleHulk commented on my talk page. I absolutely LOVE the sense of humor he frequently uses during discussions and in the edit summary, whatever the circumstances are he somehow has me laughing my ass off all the time whenever I encounter him. I realized that if I have a bad day, I can just go to his contribution hostory and see posts of his that are guranteed to make me laugh. This made me think, why not make it possible to watchlist contributions? Primarily it can be used for legitimate purposes (like monitoring a new user's contributions for mentorship purposes or a recently unblocked or otherwise scrutinized user to monitor their edits and behavior, or to keep track of edits and improvements for a particular topic if the user specializes in that topic), but it also has the potential to be used for personal pleasure too such as in the case I provided. Would this be a good idea to extend watchlist abolities to a specific user's contributions? DrewieStewie (talk) 03:17, 18 May 2021 (UTC)Reply

DrewieStewie, m:Community Wishlist Survey 2021/Admins and patrollers/Watchlist of users — Alexis Jazz (talk or ping me) 03:29, 18 May 2021 (UTC)Reply
phab:T2470 has been opened for this since 2004! — xaosflux Talk 15:36, 18 May 2021 (UTC)Reply
See also phab:T33105. — xaosflux Talk 15:37, 18 May 2021 (UTC)Reply
It's not quite what you want in this particular situation, but you can set up RSS feeds for individual user contributions as well as page histories. Not many people regularly use RSS readers any more, but I think I had a couple of these set up back when I did do so. Andrew Gray (talk) 16:47, 18 May 2021 (UTC)Reply
Anyone who actually wrote a script for this would have a very high likelihood of being SanFranBanned and having the source code oversighted, unless the person you want to follow has to approve you running such a script on them (and I can't imagine many people giving such approval, except in a few very limited training/mentoring situations). To quote the WMF on exactly this proposal:

Following and being followed must be a mutual agreement. The idea would be to have a way to give agreement on the fact that you can be followed. On the watchlist, when you add a user, that user will receive a notification to agree (or not). That notification would gather a message from the follower and a link to the follower's talk page. Resume that following may be possible on the watchlist or into preferences page.

 ‑ Iridescent 19:04, 18 May 2021 (UTC)Reply
This WMF statement sounds incompatible with the CC-BY-SA license that writers here agree to. The right to copy someone's work, must surely also include right to the awareness that the work exists. And the knowledge of each individual piece of work implies being able to know everything a user does under that license, from the time they release the copyright, ie when they save. Graeme Bartlett (talk) 23:12, 18 May 2021 (UTC)Reply
The licence does not include a requirement to notify the authors of all copies or derived versions, to have a central registry of all copies or derived versions, or other comprehensive awareness mechanisms. It would be a significant burden and potential privacy concern for distributing or modifying works in a free (libre) manner. That being said, the recent changes feed and the watchlist are awareness mechanisms for changes made on Wikipedia. User contributions are already fully visible, either through the special contributions page or through the recent changes feed. Anyone can build an interface to display them as they wish (they can create a client that tracks which contributions they have already clicked through to see, for example) and make use of it on their own computer. isaacl (talk) 02:25, 19 May 2021 (UTC)Reply
Um... no? I thought we discouraged hounding. Seems like this would just provide an easy tool to make it worse. 🐔 Chicdat  Bawk to me! 10:41, 19 May 2021 (UTC)Reply
(This is not about Special:Tags. — xaosflux Talk 12:46, 19 May 2021 (UTC))Reply

Do tags actually improve anything on Wikipedia? They really have no point beyond kicking the can further down the road for other editors to address. I don't need an orange tag for me to recognize that an article needs sources or has POV issues. Just look at this tag I removed from Walter Mondale dating back to 2011! It didn't accomplish anything in over 10 years beyond being an eyesore for the reader. If an issue actually needs to be brought to the attention of other editors, isn't the To-do list template on the talk page enough? ~ HAL333 15:33, 18 May 2021 (UTC)Reply

beyond being an eyesore for the reader Did a reader tell you this? Genuinely curious. I strongly suspect it's just the editors who think these are eyesore. I see no reasons not to highlight issues as these are helpful in telling readers that WP is a work in progress and inviting them to contribute. Many of these tags don't get addressed, but they also do no harm. – SD0001 (talk) 16:04, 18 May 2021 (UTC)Reply
I was a reader for quite a while before I took the jump and became an editor. That's first-hand but it's also anecdotal, I guess... ~ HAL333 22:40, 18 May 2021 (UTC)Reply
If you disagree that a tag is accurate then please remove it. People who add tags have no more power than those who remove them, and if there is a dispute about them it should be discussed on the article talk page. Phil Bridger (talk) 18:23, 18 May 2021 (UTC)Reply
To avoid some potential problems it would be preferable to leave an edit notice specifying why you think the tag is inappropriate when you delete it. Cheers, · · · Peter Southwood (talk): 15:04, 19 May 2021 (UTC)Reply
On "being an eyesore", please don't wander off into the direction of "reader mode" and "editor mode", etc and related arguments. EpicPupper (talk) 19:02, 19 May 2021 (UTC)Reply
@HAL333: to be honest, most clean-up tags should be moved to the talk page, with the exception of issues which are important but not obvious. And yeah, removing excessive clean-up tags should be considered just as acceptable as adding tags in the first place is. Elli (talk | contribs) 20:02, 19 May 2021 (UTC)Reply

Sometimes, when you're editing in an internet cafe or school, you get user-talks which aren't meant for you, since your IP address is shared. Maybe if each IP user had a special ID attached to them which is tied to device characteristics, the minor confusion of getting talks for someone else could be lessened. — Preceding unsigned comment added by 2600:1702:3520:3D50:D68:F0AD:9C8D:4EA2 (talk) 16:35, 18 May 2021 (UTC)Reply

This would be hard, to ensure that things like your session information remain private. The easiest work around to this problem currently is to create an account. Accounts are free. — xaosflux Talk 17:03, 18 May 2021 (UTC)Reply
This is why you can create an account. Sungodtemple (talk) 17:05, 18 May 2021 (UTC)Reply
If they edit from the same device and same IP, then there is no way to differentiate it. Create an account and then you will only get messages for you and not your classmates. RudolfRed (talk) 22:30, 18 May 2021 (UTC)Reply

I have been through Category:All articles with too few wikilinks and I think a bot to link every word with a article on it would be good. It would mean editors don't have to go and make links anymore. Give me your opinions. TigerScientist Chat > contribs 20:21, 18 May 2021 (UTC)Reply

I do not think this is a good idea... --Trialpears (talk) 20:23, 18 May 2021 (UTC)Reply
This would likely create articles that have a WP:OVERLINK problem as Trialpears has so admirably demonstrated. Better too few than too many. OTOH the too few category would be a good start for any wikignomes looking for a new task. MarnetteD|Talk 20:31, 18 May 2021 (UTC)Reply
Okay yeah I can see why no but maybe if it somehow could recognize normal words not worth linking. TigerScientist Chat > contribs 20:34, 18 May 2021 (UTC)Reply
I'm sorry if that was a bit snarky, I just thought it was funny. On the other hand the WMF recently released Special:Homepage as a beta which has an add a link task which suggest places to add links. It's not perfect and I would definitely not want that to run unsupervised but there has been work in this direction. --Trialpears (talk) 20:37, 18 May 2021 (UTC)Reply
No that was a great way to show why it would be bad. If it could identify links very well do you think it would be a good idea? TigerScientist Chat > contribs 20:40, 18 May 2021 (UTC)Reply
If it was incredibly good with a near human level judgement then yes, but I can't see that happening in the coming decade. The growth team efforts here are almost certainly the best feasible improvement. --Trialpears (talk) 20:50, 18 May 2021 (UTC)Reply
If a bot could do human-like constructive editing, running it would be a good idea. Actually, if it did its job correctly, running it would not even require approval. You'd just run it, and noone would notice that ToBeFree has been a bot all the time. Or TigerScientist. Or Trialpears. Whoever. So your "if" question is clearly answerable with "yes", but this answer does not actually contain information. ~ ToBeFree (talk) 21:06, 18 May 2021 (UTC)Reply
I don't think you realise just how complex a problem this is. If an article contains the word "Chord" how does the bot figure out whether it's supposed to be linked to the musical, geometric or aeronautical article? If an article mentions "John Smith" how does the bot figure out if we have an article on the specific person mentioned? They might be listed on John Smith or we might not have an article on them. How does a bot figure out which words are a phrase relating to a concept that should all be joined together as one giant wikilink - if an article mentions the Internet of things it should be a single link, rather than e.g. Internet of things. How does a bot deal with things that go by multiple names, e.g. an actor that has a stage name? This would be an extremely difficult bot to program - it's a lot more complex than just putting square brackets around "interesting" words, and if done poorly it could make an enormous mess of incorrect links. 192.76.8.73 (talk) 21:18, 18 May 2021 (UTC)Reply
My own experience (which is far from a proper statistical study) is that there are as many articles that have too many, or bad, internal links as there are with too few. This is just one more area where human judgement is still way ahead of any artificial, bot-like, intelligence. Phil Bridger (talk) 20:44, 18 May 2021 (UTC)Reply
TigerScientist, bots are useless for this. It could link Earth, Wind & Fire which isn't even remotely connected to Earth, Wind & Fire. It could give Tony Hawks skateboard to a British comedian. It could render the windows in my house useless and the only sound the Doors could make would be being slammed shut. The Red Hot Chili Peppers would become edible and the tomatoes would become inedible. (unless you're a cannibal) It would be completely and utterly chaotic. You'd be giving that bot a license to kill Wikipedia. — Alexis Jazz (talk or ping me) 08:23, 19 May 2021 (UTC)Reply
No, no. Avoid bots for this. They just give lots of errors. Redirection works fine. --Joujyuze (talk) 07:44, 19 May 2021 (UTC)Reply
Yeah, I don't see a reason to do this. What should be linked is an editorial decision, and this requires actual editors. Elli (talk | contribs) 01:49, 20 May 2021 (UTC)Reply

On the iOS and the Android Wikipedia apps, there is a "Reading List" feature. Would it ever be possible for this to sync with the website too, and not just other apps? A new "reading list" button on the website would be very helpful! Squid45 (talk) 14:44, 22 May 2021 (UTC)Reply

There are some browser extensions that allow you to save articles to the apps' reading list. I don't know how well these work, or if they work the other way round (i.e. view articles in the apps reading list from your desktop browser). the wub "?!" 14:29, 23 May 2021 (UTC)Reply

I think the ability to be able to select and place "citation needed" through the cite feature in the visual editor with a single click would be helpful. Not needed but it would save time. TigerScientist Chat > contribs 21:05, 18 May 2021 (UTC)Reply

Copy the following text and paste it in any article: {{cn}} This will simply work with the visual editor in the way you're requesting. ~ ToBeFree (talk) 21:09, 18 May 2021 (UTC)Reply
ehh I guess maybe. TigerScientist Chat > contribs 15:10, 19 May 2021 (UTC)Reply
You may want to post this on Phabricator if you really want this, however I think that just using the template is easy enough personally. EpicPupper (talk) 19:00, 19 May 2021 (UTC)Reply
@TigerScientist: If you'd like, I can see if I can put together a user script which could add another button to to this. Tol | Talk | Contribs 05:28, 25 May 2021 (UTC)Reply
Tol not needed but if it doesn’t take much time or effort go ahead. Thank you for even suggesting this. TigerScientist Chat > contribs 05:30, 25 May 2021 (UTC)Reply
@TigerScientist: No problem. I haven't worked with VE before, so I don't know how long it'll take, but I'll try to get something to you within about two weeks. Tol | Talk | Contribs 18:24, 26 May 2021 (UTC)Reply
  Done: The script's documentation is at User:Tol/VECN, including installation instructions. Let me know if you have any questions! Adding stuff to VE is remarkably simple. Tol | Talk | Contribs 20:36, 26 May 2021 (UTC)Reply
WOAH thank you Tol! TigerScientist Chat > contribs 16:01, 27 May 2021 (UTC)Reply
@TigerScientist: You're welcome! Tol | Talk | Contribs 20:19, 27 May 2021 (UTC)Reply

PDFs aren't really the best file format. They're hard to crop, hard to resize, they have certain privacy issues etc. I was wondering if there is support for a bot that extracts one-sided PDFs with only one image contained and saves that image in the filetype that it is. If the file inside the PDF is a JPEG, then save as JPEG, if it's TIF, then save as TIF etc. A few years ago User:718 Bot enhanced thousands of GIF files and saved them as PNGs since the latter is a better format, you can see the results here. I was thinking that this bot would do the same. Please elaborate further whether this is a good idea or not down below! I found this neat tool which extracts images from PDFs automatically.Jonteemil (talk) 10:47, 27 May 2021 (UTC)Reply

I now learned that images inside PDFs aren't really stored as JPEGs, PNGs or etc. but rather it is the binary data for the pixels, the colorspace used for the image, information about the Image. The tool I found that extracts images from PDFs only seem to find JPEGs and PNGs but I don't know how it determines what is one or the other. For example I put File:Example.tif inside a PDF and but the tool extracted the image as a PNG, not a TIFF.Jonteemil (talk) 23:24, 27 May 2021 (UTC)Reply

Now, I'm a regular user of Special:RecentChanges, and I oftentimes use tags like #possible BLP issue or vandalism or #possible userspace spam to filter out messages. The issue is, there are tags like #possible vandalism and #Possible vandalism, and #Possible self promotion in user or draftspace and #Possible self promotion in userspace. I'm confused, why are the tags separate? And there are already tags like #possible vandalism that are tagged by multiple filters. Sungodtemple (talk) 01:27, 28 May 2021 (UTC)Reply

Can you create a button "Report this page to the administrators" like on other sites? It was very hard to find the page to call administrators.Tint Last (talk) 04:24, 28 May 2021 (UTC)Reply

Could you give us examples of those "other sites"? Nardog (talk) 04:27, 28 May 2021 (UTC)Reply
On https://www.reddit.com/r/pics/comments/nmy9sd/at_the_salt_lake_city_farmers_market_a_few_years/ there is a small flag and the word "report" to report the post. On Twitter I can use "Report Tweet". Tint Last (talk) 17:23, 28 May 2021 (UTC)Reply
If you see an error, just fix it. Sungodtemple (talk) 17:24, 28 May 2021 (UTC)Reply
More often than not, new users don't know what the role of administrators is on Wikipedia. The rule of thumb is that any normal process starts with editors who are not administrators, and only when a decision is made are administrators called to implement it. There are ways administrators find such pages and usually don't need to be specifically alerted for them. It's quite rare for administrators to be called to act straight away without some editor-driven process preceding it. – Finnusertop (talk ⋅ contribs) 08:23, 28 May 2021 (UTC)Reply
{{sofixit}} or send to CSD are "expected" responses for the most obvious "problem pages" - but these are not necessarily intuitive for brand new editors. (c.f. why edit requests get sent to OTRS by brand new editors) -- ideas for improvement could be useful. — xaosflux Talk 12:43, 28 May 2021 (UTC)Reply
I'm sorry. I thought I had found the page to report to administrators:Wikipedia:Administrators' noticeboard and the dispute on Bob Coronato was settled by the administrators. But the other administrator told me I should have reported List of songs about Alabama by email? With a button to report pages to the noticeboard it would be easier to make a report. Tint Last (talk) 17:35, 28 May 2021 (UTC)Reply

I’ve reported a lot of vandals to administrator intervention against vandalism, and while I think AIV is great for most cases, I think that it can be too slow in some serious cases. Last night was one of those— there was a group of trolls attacking Waiuku; I took them on with the help of two others, but the edits kept coming. LizardJr8 reported them, but I knew it would be a while before an admin came to help. The vandals were destroying the page quicker than we could fix it. Then I decided to cut the AIV line and go straight to Liz. She speedily dealt with it. What I think we should do is create a new template. It will serve as a kind of Bat-Signal to quickly get an admin’s attention for cases of vandalism that need to be dealt with immediately, not to be dealt with at the whim of anyone that’s just passing by. Of course, I recognise there would be an issue with new users thinking that their cases of vandalism trump all other cases in importance, and admins would be flooded with requests. I’m not sure if this is possible, but perhaps we can make it so that only more experienced editors have access to the template (maybe as a part of Twinkle?). That way, they have enough experience under their belt to know what cases are high priority and what cases aren’t. It would work like this— once a user finds an admin, they go the the admin’s talk page, open the Twinkle menu, and select the template. Maybe there’s one or two boxes they should fill out, such as the name/IP address of the vandal or amount of disruptive edits they made. In summary, this feature would help users in an active vandalism crisis quickly get hold of an admin. A Twinkle template would allow them to quickly get the information conveyed instead of creating a lengthy plea for help (like what I do, LOL). I would love to hear your thoughts on the matter. 🐍Helen🐍 20:27, 28 May 2021 (UTC)Reply

HelenDegenerate, See Wikipedia:Requests for comment/Responder role for a proposal (which doesn't appear to be making any progress) from ProcrastinatingReader for one way to deal with this. -- RoySmith (talk) 21:46, 28 May 2021 (UTC)Reply
I think we had a few more things to figure out, but to some degree I don't think it would pass if it went live now. It's probably better to try it if/when a greater portion of the community feels there is a problem here. ProcrastinatingReader (talk) 21:57, 28 May 2021 (UTC)Reply
Oh yeah, this problem. I think we should have a chat room (IRC, Matrix, Discord, whatever) where some admins hang out, and there could be a command that'll ping one of them, rotating among the admins in the room. There should be either a Twinkle button or a separate user script that'll send the command with a link to a specific AIV case. We can figure out access control later. I've been meaning to slap together a prototype for a while now - we don't even need to ping anyone for starts, just post a message - and thanks for the reminder. Enterprisey (talk!) 03:22, 29 May 2021 (UTC)Reply
Did you know that... believe it or not, we need more administrators? The solution is actually things like: a) making RfA more active. b) making RfA less toxic. c) making RfA more encouraging. If we could do that, then this thread never would have started. 🐔 Chicdat  Bawk to me! 12:16, 29 May 2021 (UTC)Reply
From what I have heard, d) be more circumspect before reporting things to AIV might also help. Anecdotally, there are a lot of questionable reports. Jo-Jo Eumerus (talk) 13:06, 29 May 2021 (UTC)Reply
What about having the bot move to the top reports where the user has fresh reverted edits after the report tinestamp? –xenotalk 13:15, 29 May 2021 (UTC)Reply

Most article talk pages are ghost towns, filled only with assessment templates. A lot of article related discussion happens on user talk pages, wikiproject talk pages or in edit summaries instead of on the article talk page. In the rare case where people come and try to discuss actual article content on article talk pages, they tend to get ignored for years, as nobody notices their contribution to the talk page. Should we encourage people to try to find out who the actual page authors are and ping them when they try to write a new talk page post? The suggestion to do so would need to be in {{talk header}} and/or the editnotice to have any chance of being seen. Or should we try to encourage people to go to a more widely watched forum when they want to say something about an article? —Kusma (talk) 10:12, 29 May 2021 (UTC)Reply

@Kusma: If the users never go to the talk page, then how, if I may ask, would a notice on the talk page help? 🐔 Chicdat  Bawk to me! 10:42, 29 May 2021 (UTC)Reply
@Chicdat: I am trying to solve the problem that people who have managed to use the talk page are being ignored. The alternative is to change the talk page header to discourage people who have found it from using the talk page. —Kusma (talk) 11:12, 29 May 2021 (UTC)Reply
@Kusma: Can you give me a few talk page examples with years-old, unnoticed posts? 🐔 Chicdat  Bawk to me! 11:15, 29 May 2021 (UTC)Reply
@Chicdat: From my own contributions: Talk:Leningrad Cowboys Go America, Talk:Ursula von der Leyen (although this is an article with 1.2 million views per year, none of the talk page posts get actual discussions). Additionally, I have noticed that I don't go to article talk pages to discuss the article. I think I should have posted this at Talk:Lynching of John Carter, with a ping to Drmies and Uncle G. I posted at a user talk instead because I'm used to ignoring article talk pages. I think that is a bad habit, but it is a common habit. I am trying to propose we all use article talk pages more, and use pings to make sure the right people get notified of the discussion. —Kusma (talk) 12:05, 29 May 2021 (UTC)Reply
@Kusma: Wow... you're right! I never noticed any of this before. The Talk: namespace really is quite inactive. Thank you for noticing this problem. 🐔 Chicdat  Bawk to me! 12:08, 29 May 2021 (UTC)Reply

Ok, so TLDR: Article talk pages are dead. Should we bury them or revive them, and if we want to revive them, should we try to use pings to do so? —Kusma (talk) 13:07, 29 May 2021 (UTC)Reply

First, yes - I agree it's a problem. Often if something I wrote sits more than a day or two then I'll check history and go to one of the recent editors to discuss. I like the idea of notifying the original author, but I'd include the editor with the most edits as well. Also, should it be a "ping" on user talk, or just an email that someone has posted to the article talk and it hasn't been responded to in x hours, or x days? — Ched (talk) 13:25, 29 May 2021 (UTC) edited: revived implied by my response. — Ched (talk) 13:27, 29 May 2021 (UTC)Reply
To expand a bit on the why notify the most prolific editor as well: There are a TON of articles where the original author simply isn't editing wiki anymore. Even the most prolific editor of that article may be long gone. At that point, I guess you just look at page history and try to find someone still active. (which doesn't really solve the problem, I know.) — Ched (talk) 13:33, 29 May 2021 (UTC)Reply