Wikipedia:Village pump (proposals)

Wait.. you are the one who started IPC? This is notable Wikipedia history, given how influential it has been (for better or worse!). If you don't mind, this should be in a "history" section at

WP:IPC -- GreenC 15:48, 12 August 2021 (UTC)[reply

]

Wikipedia:"In_popular_culture"_content#History -- GreenC 15:55, 12 August 2021 (UTC)[reply]

I thought I'd coined the phrase, but on doing some digging there were already a few IPC pages existing prior to that—the oldest I can find on a very quick dig is

Iridescent 16:07, 12 August 2021 (UTC)[reply

]

The oldest deleted page titled "...in popular culture" - where I can't find evidence that it was moved there later, like, say,

Cryptic 18:45, 12 August 2021 (UTC)[reply

]

IIRC these started showing up as sections in articles in early 2004, but I wouldn’t be surprised if someone dug up a diff from late 2003, see for example 1 2 3; even

Cultural influence of Plato's Republic, Venus in culture, Cultural references to Hamlet , Women warriors in literature and culture, Synesthesia in fiction etc. Doubtless someone with a bit more free time available will be able to find other formulations in current use. Regards, 81.177.3.8 (talk) 19:20, 13 August 2021 (UTC)[reply

]

Might I suggest Wikipedia:"In_popular_culture"_In_popular_culture pending this RFC? @GreenC Shushugah (he/him • talk) 02:20, 15 August 2021 (UTC)[reply]

What I was thinking was using "Cultural influence" or "Cultural allusions" or maybe "Cultural influence and allusions". in WP:VG, we don't have "in pop culture" sections but just "Legacy" sections instead.Blue Pumpkin Pie (talk) 16:02, 12 August 2021 (UTC)[reply]

• Oh but now here's a thought regarding the content rather than the name of these sections. There's plusses and minuses, but to my mind, the purpose of these sections is to indicate how well known the entity is. So, let's take "Do not go gentle into that good night"... is this an obscure poem like say "The Legend of Novgorode" or thousands of others, or is it more well known? That's a very important point for the reader to know! But we can't just assert it. If we have a source saying "This poem is super well known!" fine, but first of all we usually don't, and second even if we do it's just one guy asserting it.

But... remember Writing 101: show, not tell. Well, Do not go gentle into that good night#Use and references in other works shows very well that the poem's reasonably famous. Important info. Even minor examples can help with that. If somebody says a line in passing on some HBO show, that that's another demonstration that it's floating around in the culturespace, unlike say Trilce.

But then: as we all know, these sections can grow out of control, too. So here's what I've tried, and I think it maybe works OK: Make the "References in other works" short but dump the excess examples down into either the Notes or the References sections, where they are they are still there but don't bother anybody. At Anyone for tennis? I gave some examples (basically the bluelinked ones), then added "And so forth" with the ref for that containing all the extra non-bluelinked examples. You can also use a Note instead. Reasonable approach? Herostratus (talk) 16:16, 12 August 2021 (UTC)[reply]

I don’t think that works particularly well— feels like hiding cruft that shouldn’t really be in the article. But I have no problem with ‘in popular culture’ sections at all, when put together properly. This whole discussion seems, to me, a waste of energy that would be more productively expended in other endeavors, such as writing content. Eddie891 ^Talk _Work 16:26, 12 August 2021 (UTC)[reply]

Yeah how about not telling other people how to spend their volunteer hobby time maybe. I've got 500 articles created and many thousands of article edits, how about you. Herostratus (talk) 16:43, 12 August 2021 (UTC)[reply]

…I’m just suggesting that I don’t think this is the most productive thread Eddie891 ^Talk _Work 18:18, 12 August 2021 (UTC)[reply]

since you asked, in the past five years since I began editing, I’ve created over 300 articles and made over 40,000 edits and written several featured content (incl one fa with an in popular culture section) and had almost 100 dyks. In that time, you’ve created 156 articles and made about 11,000 edits. Eddie891 ^Talk _Work 18:25, 12 August 2021 (UTC)[reply]

I'm going to nicely ask to stay on topic and remain civil. And if you feel its a waste of time, you can contribute to the articles you want. it's ok to not find a topic productive. Others don't feel the same way.Blue Pumpkin Pie (talk) 18:29, 12 August 2021 (UTC)[reply]

? I don’t see the problem with my conduct. I said what I thought, herostratus replied to say ‘yeah, no’— which I understand— and I answered his explicitly posed question. Eddie891 ^Talk _Work 18:55, 12 August 2021 (UTC)[reply]

Since it has come up, I have over 1,800,000 edits, with over a million being article edits. I have created over 6,000 articles. To reiterate my earlier position, I think "in popular culture" sections are fine. If you want to police them for proper citation to sources, go ahead. If you want to structure them into prose, have at it. Trying to remove them altogether is counter to the information-sharing function of the encyclopedia, and is indeed a waste of volunteer time. BD2412 T 20:09, 12 August 2021 (UTC)[reply]

• Whatever such sections should be called, they should not encourage every casual reader to think, "Hey, this list doesn't include the fact that in the third episode of the fifth season of Family Man, it is revealed that the neighbor's cat is named
  Crookshanks." Calling them "In popular culture" seems to do exactly that. —valereee (talk) 17:07, 12 August 2021 (UTC)[reply
  ]
  • This also brings up that, it's also the list format that many of these take. Lists "look" easy to add onto so draw in every trivial mention. It is far better to try to encapsulate how a topic has entered popular culture by prose, if possible, as I've found new editors tend to be a bit more adverse to trying to alter that and making a mistake. --Masem (t) 17:21, 12 August 2021 (UTC)[reply]
    • It does draw in every trivial mention, but that also means it's drawing in first posts by readers, sometimes. That's key. We get them to add a reference to say something the saw on American Dad. That's the start. Then we sloooowly reel them into our dysfunctional little group here, and one day they wake up and they're writing articles about ninth-century Inner Mongolian minor poets. But by then it's too late for them to rejoin the world of normal people. BWAHAHAHAHA Herostratus (talk) 19:07, 12 August 2021 (UTC)[reply]

      OTOH, when that edit gets reverted five minutes later with an annoyed edit summary of "rem trivia"...the bait is doing its job, the reel not so much. :D —valereee (talk) 19:17, 12 August 2021 (UTC)[reply]

      While lists are a good way for first edits of new editors, 99 times out of 100, when added to a pop culture section, it is unsourced. Which if unchecked creates a self-replicating problem. --Masem (t) 20:54, 12 August 2021 (UTC)[reply]

      I think this gets well off-topic, but it's not specific to IPC: we have a huge problem because no-one's first 100 edits are good, but 100 bad edits in 2006 is alright because at least it creates something new, and 100 bad edits in 2021 is not fine because it makes existing alright content worse. So whatever you do is wrong and you'll get reverted and scared off the website. Or you don't get reverted initially, but someone finally notices you and tells you everything you've spent many hours on is wrong, and now you get scared off the website but you're also in tears. My best answer to this so far is "people should start off at Fandom and then come here when they're already used to the website design and some of the culture". — Bilorv (talk) 01:29, 13 August 2021 (UTC)[reply]

      Oh $deity no don't send people to Fandom for a good experience. It's as good as testwiki for learning Wikitext I suppose, but does absolutely nothing for learning to edit in a collaborative environment. AntiCompositeNumber (talk) 03:51, 15 August 2021 (UTC)[reply]

      With the levels of hostility expressed to both newcomers and old hands, there's not much collaborative environment in Wikipedia either. — Bilorv (talk) 12:15, 22 August 2021 (UTC)[reply]

    I think that is an excellent point. Lists absolutely beg people to add to them. Maybe such sections shouldn't usually be lists? —valereee (talk) 19:21, 12 August 2021 (UTC)[reply]

• I think that most of the posts are getting away from the main point, which is that the word "popular" shouldn't be used here. For example opera and ballet are not considered "popular", but mentions in such are just as valid or invalid as mentions in Hollywood films or pop music or computer games.
  Phil Bridger (talk) 19:37, 12 August 2021 (UTC)[reply
  ]
• This is a fairly interesting point, and I would add my voice to those that more or less hate these sections as they almost inevitably become long lists of unsourced cruft. I'm not sure changing the name will solve that core problem, but I'm nt opposed to the idea of trying it. My personal approach is to just remove anything without a source since that's pretty basic, and also remove mentions based on one throwaway line from a tv show, and maybe add things like hidden comments or edit notices if they persist. (Not that it always works, it's been over a decade of trying get people to stop posting that
  talk) 20:29, 12 August 2021 (UTC)[reply
  ]
• Where there are also things like opera & novels, I tend to use headings like "Cultural references", which might also cover a whole themed episode of say South Park, but not a passing reference. Johnbod (talk) 01:36, 13 August 2021 (UTC)[reply]
• I have no real opinion on the title discussion, "popular culture" is intuitive if not dictionary perfect. Like many above I too generally dislike these sections, principally because they are usually an unsourced dumping ground for passing mentions in an single episode of a sitcom or the like. I think a good start may be to make
  MOS:POPCULT more succinct and punchy, emphasising the 2015 RfC, possibly elevating the sourcing requirements from that RfC into a guideline itself. Cavalryman (talk) 04:00, 13 August 2021 (UTC).[reply
  ]
• I suggest "Society and culture" as a title (used in
  WP:TRIVIA and could be completely blanked.Tom (LT) (talk) 10:32, 13 August 2021 (UTC)[reply
  ]
• Strong oppose to a change (although hopefully this isn't a decision that is being taken here). On individual pages I'd guess that 'In popular culture' sections are themselves quite popular, have likely been a Wikipedia mainstay since 2001, and have a name which is both extremely recognizable and a more-than-adequate descriptor. On stand-alone article pages the idea of changing this ramps the "Huh?" factor up a level or six. Important pages such as World War I in popular culture, World War II in popular culture, Apollo 11 in popular culture and dozens if not hundreds more have educated readers about cultural events which have themselves educated the public since their inception. A good discussion, but let's leave it at that. Randy Kryn (talk) 11:12, 13 August 2021 (UTC)[reply]
• I don't know. Who cares. This seems like something that should be handled on an article-by-article basis.--WaltCip-(talk) 17:08, 13 August 2021 (UTC)[reply]
• I think that the underlying issue is this. The normal practice is that In the normal inclusion / exclusion decisions are made based on a multitude of factors including degree of relevance and degree of importance to the topic. Headings should be for material that inherently belongs in the article. Certain headings distort that process towards bringing in material that would not have otherwise been included. Headings like "in popular culture" are of that type. Particularly promotional or fandom items. North8000 (talk) 17:28, 13 August 2021 (UTC)[reply]
• Oppose. I'm not going to talk about whether or not we should have these sections in the first place, because that doesn't seem to be the point of this RfC. But I don't think most people intend a negative implication when they're talking about popular culture. I also agree with Randy Kryn's reasoning. Clovermoss (talk) 22:03, 13 August 2021 (UTC)[reply]
• The existence of the article Mermaids in popular culture is presumably what allows the article Mermaid to pursue a purely zoological examination of these alluring beasts. The former article includes such pop culture phenomena as Zemlinsky's Die Seejungfrau; which seems only right as Die Seejungfrau has been put out on at least ten commercial recordings, AFAIK five times as many as there have been of Tubular Bells. -- Hoary (talk) 22:34, 13 August 2021 (UTC)[reply]
• Oppose any across-the-board deprecation of "in popular culture" sections, which to me is where you place mentions of snippets of musical pieces heard in video games, TV shows, etc., instead of discussing modalities, orchestration, and influences (which probably fairly applies to the Dylan poem referenced above), etc. The lack of sourcing in such sections is notorious, but that's for editors of individual articles to decide how much original reporting to let in. Dhtwiki (talk) 09:03, 14 August 2021 (UTC)[reply]
  • The lack of sourcing in such sections is notorious, but that's for editors of individual articles to decide how much original reporting to let in. We already decided that; the answer is "none," and we wrote that up on a core content policy page called WP:No original research, which is global consensus that editors of individual articles cannot change. Levivich 14:39, 16 August 2021 (UTC)[reply]
    • I tend to let plausible, innocuous statements stand, not to violate policy but in the hope that other editors will come along and find sources for them. Leaving unsourced a statement that a subway's door-closing chime is inspired by a piece by Handel isn't at the same level as, say, leaving an unsourced statement alleging criminal wrongdoing. If left unsourced, even the innocuous statements may eventually be swept away by more deletionist editors; and I won't object. Dhtwiki (talk) 01:32, 21 August 2021 (UTC)[reply]
• Oppose deprecating the term 'in popular culture'. It's merely one of several acceptable titles for such sections/articles, and I see no need for any new rules.
  LEPRICAVARK (talk) 23:14, 14 August 2021 (UTC)[reply
  ]
• Deprecate such sections. Listing every time X appears in fiction (or popular culture, or whatever) is what
  WP:CARGO—fiction is not fact and collecting raw data does not produce analysis. TompaDompa (talk) 01:32, 15 August 2021 (UTC)[reply
  ]
  • You are iVoting in the wrong discussion, TompaDompa. This is just about the changing of the section title, "In popular culture", not about whether the sections should exist or not. (That is for a whole different discussion.) GenQuest ^"scribble" 07:02, 15 August 2021 (UTC)[reply]
    • I'm not at the wrong discussion, I'm just proposing a more radical solution – one that would
      T:CENT says Should articles continue to have sections titled "in popular culture"? My answer is no—not because the title is bad, but because having such sections is. TompaDompa (talk) 03:24, 16 August 2021 (UTC)[reply
      ]
• Oppose deprecating the term 'in popular culture'. Like it or not, this content is going to continue to be added to the articles, and we have to put this junk somewhere, and the un-cited and trivia can then be easily deleted from these sections. Now, someone start a discussion about whether these sections should even continue to exist and things will get quite interesting. GenQuest ^"scribble" 07:02, 15 August 2021 (UTC)[reply]

• Comment I don't understand your reasoning for the opposition. You just commented that this discussion isn't about removal of these sections, but you used that very reason for that. What do you mean by "we have to put this junk somewhere". Would it hurt the article if we don't?Blue Pumpkin Pie (talk) 16:13, 15 August 2021 (UTC)[reply]

So I'm seeing kind of three questions:

1) There's the question of whether sections should even exist' at the end of articles that have things "In Joyce Carol Oates's 2000 novel Blonde, the character of The Ex-Athlete is based on DiMaggio" and "In Seinfeld, Season 3, episode 1 "The Note, Kramer spots DiMaggio in a Dinky Donuts" and "DiMaggio appears in Harvey Comics' Babe Ruth Sports Comics (August 1949)"

2) There's the question of, if they do continue to exist, (and they will) how should they be named (the original and basic question of this thread).

3) And then there's also always the eternal the question of, if they do exist, how to curate? Should stuff like the first item above (important character in a serious novel) be in them and stuff like the second (passing mention in a vulgar and witless, but very popular, TV program) or the third (cover appearance in a long forgotten and lost children's comic) not? (And other questions of details.) And what practical procedures might be used to curate them?

So, for the first question, there are a lot of decent arguments either way, but as a practical matter, come on. We're always going to have these sections. It's entirely legit to express your opinion for/against them, but it's not going to change anything. So moving on.

For the third, well the current procedure is for editors -- driveby anon readers often enough -- to keep adding stuff (some good, some marginal, a whole lot silly; some well ref'd, some poorly ref'd, some unref'd) and for other editors to come across them, mutter "oh my God" under their breath and trim them (or even delete them), and for people to occasionally argue about it, since ultimately it's a matter of opinion how to curate. That's kind of kludging along (like a lot of the Wikipedia!), but it's OK, and I honesty don't think there's a better way. It's alright. It works OK. The project is not going to collapse over this. I can't imagine any rules that could be put in place ("No more than ten items" or "No refs to non-bluelinked sources" or whatever).

For the second, that's the question.

• Should these sections continue to be named "In popular culture" as is usual (altho far from universal)?
• If not, should they mostly be named one other single name, or let 1000 flowers bloom?

I just don't think there's any way to make a rule. There is the general practice of naming them "In popular culture", and rules are to codify common practice, so there could be a formal guideline made to that effect, such that someone could come along and rename your "In art and literature" to "In popular culture" and have the high ground. But I mean that's not going to happen. You're not going to get even 60% of a large group to agree to that. So just forget it. Trying to make a rule to have the sections be named some other thing is forget it squared.

It would be preferable to have a (generally common) name. As we do for "Early life" and "Personal life" "Discography" and "See also" etc. That's a good point. And they only possible generally common name is "In popular culture", barring a long-term sea change. So it's fine for individual editors to keep doing that.

For my part, personally, in my personal opinion it just sticks in my

So... different names in different articles for sections that are pretty much the same content? Oh well. Least bad option IMO. Herostratus (talk) 18:48, 15 August 2021 (UTC)[reply]

• Oppose deprecating the Popular Culture sections. That is the name that Wikipedia has had for those sections, and usually they are useful additions to the encyclopedia. In cases where they are silly or useless, we can get into ugly edit-wars that go to
  WP:ANI a little less heavy. That's only half humorous. But the sections are useful and the name is as good as anything else. Robert McClenon (talk) 19:45, 15 August 2021 (UTC)[reply
  ]
• Deprecate the sections as
  WP:OR because it's including something in the Wikipedia article Foo that isn't mentioned in any of the RSes about Foo. Anything not mentioned in the Foo RSes doesn't belong in the Foo Wikipedia article at all. Levivich 14:34, 16 August 2021 (UTC)[reply
  ]
• I seriously doubt that changing the typical name for these sections would stop them from accumulating cruft.
  talk) 22:24, 16 August 2021 (UTC)[reply
  ]
• Meh. There are probably far more important things to be doing right now, this feels very navel-gazey. Stifle (talk) 09:54, 17 August 2021 (UTC)[reply]
• Oppose change This is a solution looking for a problem. There is nothing wrong with the phrase "in popular culture"; everyone knows what it means. I don't see any good reason to change it. Mlb96 (talk) 20:49, 17 August 2021 (UTC)[reply]
• Support change. I just had to wipe out the entire "In Popular Culture" section in
  WP:UNSOURCED issues. I suspect that renaming the sections will reduce the affinity for people adding random trivia. I would also lean towards User:Levivich's position of a general depreciation, but that goes beyond the scope of this discussion. BilledMammal (talk) 07:02, 26 August 2021 (UTC)[reply
  ]
• Support change This is not, unlike what some claim, a "solution in search of a problem". The problem is very real; and you don't need a particularly acute sense of what is encyclopedic and what is not to figure that out - even xkcd gets it. There are some decent examples of sections which appropriately deal with the cultural impact of something; for ex. this (which is half decent) or this (probably one of the better examples). However, far too often, it looks just like unrelated notices of appearance, almost sillier than the xkcd comic I link as a parody earlier. RandomCanadian (talk / contribs) 23:09, 3 September 2021 (UTC)[reply]
• Oppose Popular culture is a thing. We have an article on it but it wasn't invented on Wikipedia as there is scholarship going back long before. As for the archetypal IPC section or article, that's a thing too. The main thing that's not clear to me is who updates such sections and why. Is it done by regular editors, specialist gnomes or the general public? Anyway, Wikipedia is the encyclopedia that anyone can edit and this seems to be one of the consequences. So it goes. Andrew🐉(talk) 23:55, 3 September 2021 (UTC)[reply]
• Comment I think that In Popular Culture sections can be appropriate if (1.) the thing is actually significant for its role in popular culture and (2.) the section is a succinct summary that gives a general overview and names only the most noteworthy works. The Empire State Building article does a really good job of the latter. In practice however, most In Popular Culture sections are just indiscriminate lists of times that really famous or significant things appeared in works of fiction; many of these listings either lack sources or are sourced solely to the fictional work that the article topic appears in. Honestly, I think any In Popular Culture section that consists solely of a bullet pointed list can be safely removed without negatively impacting the article. Spirit of Eagle (talk) 02:19, 5 September 2021 (UTC)[reply]
• Comment. I oppose depreciation, but I'd support considering a new name. "In culture", shorter, will do fine. What's popular is debatable anyway.--_{Piotr Konieczny aka Prokonsul Piotrus| reply here} 12:31, 7 September 2021 (UTC)[reply]

Background

This is a proposal to expand the functionality of {{find sources}}, a template frequently used in {{Talk header}} (example) to help editors find sources to improve articles. Specifically, we seek to make it possible for different sets of links to appear for different types of articles, such as links to medical sources for medical articles.

Currently, there are three four related templates that generate find sources links:

• {{find sources}}, the basic set of sources starting with five Google links, plus JSTOR, NYT, and a couple more
• {{find video game sources}}, which contains the basic set above, plus three more links targeted to video gaming
• {{
  WP:MEDRS
  -compliant sourcing
• {{find biographical sources}}, a set of links aimed at supporting biographies.

We anticipate that additional templates in this family for other subject areas may be created in the future. Under this proposal, talk header will choose between the available templates, either automatically or on the basis of a manually set parameter, to display the one most appropriate for a given page.

Approaches for selection

We are considering several possible approaches for how to select which set of links to use:

1. Through a new parameter that can be manually set (i.e. to display medical links at Talk:Giardiasis, we'd change {{talk header}} to {{talk header|search-domain=medical}} at that page)
2. Through auto-detection of WikiProject banners (i.e. Talk:Giardiasis would switch to using medical links because it includes the {{WikiProject Medicine}} banner)
3. Through auto-detection of Wikidata properties (i.e. Talk:Giardiasis would switch to using medical links because an automated analysis of its Wikidata item identifies it as a medical topic)

It would be possible to combine these approaches, such as implementing a combined option 2 – 1 approach, which in the case of {{Talk header}} would default to option 2 (project auto-detect) but would allow any editor to set a parameter (e.g., {{talk header|search-domain=medical}}) which would then take precedence over the project auto-detect. Having the parameter available would also be extensible to use by other templates on article pages where project detection isn't applicable; see below.

Previous discussion is at Template talk:Talk header, and we have created functional prototypes of options 1 and 2 in the talk header sandbox, which can be seen at the associated test page. (Current sandbox revision uses method 2, so testcases for method 1 currently fail, but pass successfully with the correct sandbox revision in place.)

Which approach(es) would you prefer?

Design

We are considering implementation via a wrapper template (here) which does all the detection of search domain, and transcludes the correct flavor of template. By placing the wrapper template at the current title (Template:Find sources) and moving the old content to Template:Find general sources, this remains transparent at the top level; all current transclusions of Template:Find sources after the changeover will invoke the wrapper, which invokes {{Find general sources}} by default. Outside of the Talk header template, this means a seamless transition for all other transclusions which will do exactly the same thing after go-live as they did before; that is, they will continue to invoke the basic "Find sources" link set, albeit by one extra call where {{Find sources}} transcludes {{Find general sources}} which invokes the Module.

Currently Template:Talk header includes the basic set of source links for all articles where it is not suppressed by parameter. After go-live, the behavior of "find sources" in Template:Talk header may change, depending which solution is chosen. If the parameter method is chosen, then the links in the Talk header would remain the same, until someone added the parameter. If auto-detect by WikiProject is chosen, then the links in the Talk header of pages on medically-related topics will switch to the medical links, and on video-related topics will switch to the video links; all other Talk headers would remain as before.

Impact on other templates

Other templates use the {{

We look forward to your feedback on the considerations above and the proposal overall. Thanks! Mathglot (talk) and {{u|Sdkb}} ^talk 23:20, 12 August 2021 (UTC) updatedto add biographical sources; by Mathglot (talk) 22:32, 13 August 2021 (UTC)[reply]

• Listed at: Wikipedia talk:WikiProject Medicine. Mathglot (talk) 23:49, 12 August 2021 (UTC)[reply]
• Listed at: Wikipedia talk:WikiProject Video games. Mathglot (talk) 23:53, 12 August 2021 (UTC)[reply]
• Support as helpfully expanding the utility of this talk page banner. I would in particular be a fan of the wrapper template idea, as it is the least disruptive imo. I think the parameter approach via "search-domain=medical" is the best solution for how to label pages as needing MEDRS in the banner, though I think the auto-labelling is a good idea. My suggestion would be to try the auto-labelling first, and revert to manual labelling anything that is medical if it goes haywire and labels a lot of unnecessary things.--Shibbolethink ^{(♔ ♕)} 23:57, 12 August 2021 (UTC)[reply]
• As co-nominator, I obviously support this. Given how widely the find sources links appear, I think it's important that they are as useful as possible, and this will help with that by making them more appropriate to their context.
  Regarding selection approaches, I favor mainly option 2 (project banners), with the ability to override via option 1 (manual parameter setting). That's because I think it's important to have an element of automation to get this adopted on a wide scale. Option 3 (Wikidata) doesn't seem technically feasible, so that leaves project banners. I think "Is this tagged with {{WikiProject Medicine}}?" is an excellent metric, because the project tag belongs on all medicine-related articles, which is also basically the set we want to have the medicine links. However, it's nice to have the ability to override the default for the rare cases where it isn't working as expected, and having the parameters from option 1 will allow that. The only thing weighing against them is added code complexity, which isn't a huge downside.
  Regarding the wrapper template implementation, that's been more Mathglot's area of focus, so I'll defer to them and others who comment here. Also, I realize all this is more complex than some of the stuff that comes across VPR, so if anyone has questions or wants clarification, please don't be afraid to ask! Cheers, {{u|Sdkb}} ^talk 05:50, 13 August 2021 (UTC)[reply]
• I support the general proposal. I do have some questions about the implementation - I think if an automated detection system is implemented it will need to have a bit more subtlety than 'does this have a WikiProject Medicine banner?'. Articles about people in medicine, healthcare companies, medical books or journals, the history or social/political aspects of medicine, etc. are usually tagged with WPMed, but these articles do not necessarily require MEDRS sourcing for all claims. I wouldn't say this is a 'rare' case. If it's possible to take the intersection of WikiProject banners into account (e.g. don't default to 'find medical sources' if there is a WPBiography tag on the page in addition to WPMed) that might help, but I think this might need to be handled via a supervised AWB run or similar. Also, a hybrid of the 'find medical sources' and 'find general sources' template could be helpful for topics that might contain claims needing MEDRS sources but which are not strictly medical. Spicy (talk) 19:54, 13 August 2021 (UTC)[reply]

  I like the thought of the hybrid template, since for the examples you gave, most sources wouldn't need the MEDRS standard but several might. For instance, a medical researcher would need it for the part of the page discussing their discoveries, a company for discussing their products, a journal for discussing medical controversies about their work, a history page for discussing medical details about viruses involved in a pandemic, etc. {{u|Sdkb}} ^talk 20:40, 13 August 2021 (UTC)[reply]

  @Spicy:, regarding "don't default to 'find medical sources' if there is a WPBiography tag on the page in addition to WPMed" this could definitely be added; it poses no great difficulty and would be isolated in the wrapper. It just depends on whether that idea gains consensus.

  Another option to consider, is that there's no need to choose only one set of "find sources" links. If the article is in both projects as in your example, one could imagine various outcomes: we could let the Talk header template grab the medical links automatically, and then add a biographical links below it, the second one either added automatically as well, or manually by adding {{

  biographical sources box

  }}. Yet another possibility, would be to have the order of WikiProjects matter: if Med is first and Biog second, then you only get the "find medical links" automatically, anything else you'd have to add manually; if Biography WikiProject was first, then it would be the other way round. But maybe it's too subtle operationally, even if well documented ("You mean, I'm spozed to read the stinkin' documentation?") Not sure why AWB would be involved; I think it's all doable in the wrapper.

  Sometimes having too much choice gets to be a problem because you can get

  paralyzed with all the possibilities, so at the outset I'm inclined to go with the simplest useful proposal that gains consensus; not because that makes it easier for template writers, but because that will give editors some time to see and use the new functionality, which may subsequently percolate into a more specific and informed wish list of increased functionality based on their real-world experience with the early version. Mathglot (talk) 22:11, 13 August 2021 (UTC)[reply

  ]

  The reason why I suggested AWB is because I think human judgment could be required for some of the edge cases - I don't think we should be giving people the impression that a MEDRS-compliant source is required to cite a medical device company's yearly revenue, for example. I suppose this could be handled in other ways, like the system you've described or Sdkb's hybrid template idea (although both of those rely on the assumption that the WikiProject tagging is accurate, which isn't always the case)... Spicy (talk) 06:16, 31 August 2021 (UTC)[reply]

  Not too familiar with

  AWB but would there be any significant downside to first using Mathglot's logic-based approached based on WikiProject with a manual override? Centralizing the logic where possible seems preferable. In the case of a medical company, maybe just add priority logic to check for 'WikiProject Companies' and assign those topics to display the general link set. AWB could still be used to mass apply overrides if there's a scenario that can't be accounted for in the root logic. - Wikmoz (talk) 20:48, 31 August 2021 (UTC)[reply

  ]

  @Spicy:, Hopefully as people see it live and perhaps read the expanded /doc, they'll see some of the options and apply whatever makes sense to get the desired display. As far as not requiring MEDRS links for the medical devices company revenue case, two possibilities occur to me without changing anything said prior (although these examples are no doubt not obvious since it's not live, and there's no doc):

  • Add an additional set of links to the header outside the scope of the new header functionality, by separately adding {{
    General sources box}} after the Talk header, so that you have both the medrs links automatically added by the new template functionality per Med project presence, and the "regular" links. You can see how this would look in this mocked up Talk page
    for the "Cochlear Limited" article (note: the actual Cochlear article TP does *not* have WikiProject Medicine on the Talk page; I had to add it to the mockup for the purposes of this test)
  • turn off find-source links with the existing 'hide' param, and add {{
    General sources box
    }} (i.e., the Talk page will look the way it does now) (the third, "hybrid" solution has been mentioned already)

  I was trying to find a medical devices company with the Medicine WikiProject already present so it would be a good example of your concern, but I gave up and had to force project medicine into the mockup for the test. (I didn't look very hard.) The question, I suppose, is this: "Is the benefit of having medrs links produced automatically via project detection worth the disruption that might be caused by losing the general "find sources" links for cases like medical device companies tagged with project Medicine, or in other cases where the correct WikiProjects are not applied and so the general link set would be better?" Is that a fair statement of your concern? I'm not sure we know the answer to that until we try it, unless someone wants to do some clever analysis and see what's out there now. I'm not opposed to undoing the installation after the fact, if on balance it turns out that the new approach is not helping editors improve the articles concerned by proposing a better set of reliable source links for most articles most of the time, because that is the whole point, right? Mathglot (talk) 01:47, 1 September 2021 (UTC)[reply]

• I broadly support the idea, allowing for tweaks to get the usability issues right. It's worth steering people towards suitale sources somehow. Shooterwalker (talk) 22:18, 13 August 2021 (UTC)[reply]
• Support. I think this is a good idea. Clovermoss (talk) 22:23, 13 August 2021 (UTC)[reply]
• Listed at: Module talk:Find sources. Mathglot (talk) 22:28, 13 August 2021 (UTC)[reply]
• Listed at: WT:WikiProject Biography. Mathglot (talk) 23:23, 13 August 2021 (UTC)[reply]
• Support. Very cool idea and great work Mathglot and Sdkb! I think Option 2 (with the manual override function) sounds like the best approach. It's probably not worth getting too hung up on the edge cases. For those, I'd favor whichever approach is easiest to engineer and maintain with a slight preference to avoid hybrid or stacked link sets. - Wikmoz (talk) 01:35, 22 August 2021 (UTC)[reply]
• Mathglot, Sdkb: What would be the next steps? - Wikmoz (talk) 04:46, 31 August 2021 (UTC)[reply]

  ​Thanks for the follow-up, Wikmoz. This hasn't garnered too much participation, but since it's been in a visible location and reactions so far have been positive, I'd say it's safe to move forward with implementation using option 2 with manual override function. {{u|Sdkb}} ^talk 20:31, 31 August 2021 (UTC)[reply]

  Yes, that sounds reasonable to me. Once it's been moved live in template form and had time to gain some page views, we could take another look to see if more tweaks or other changes are needed based on any feedback from those encountering it for the first time. Perhaps subsequent to that, I imagine one would check to see about moving parts of it into the module. Adding Sdkb. Mathglot (talk) 20:48, 31 August 2021 (UTC)[reply]

The following discussion is an archived record of a request for comment. Please do not modify it. No further edits should be made to this discussion. A summary of the conclusions reached follows.

There is a (near-unanimous) consensus for the proposal; and with comments only trickling in very sporadically in the past week or so, it's clear that

WP:SNOW applies, too. RandomCanadian (talk / contribs) 22:42, 3 September 2021 (UTC)[reply

]

---

Should administrators be allowed to apply extended-confirmed protection to

high risk templates

? 03:59, 17 August 2021 (UTC)

Moved from Wikipedia:Administrators' noticeboard § TemplateProtector bot updated; Should we use ECP on templates?

 – Wug·a·po·des​ 20:45, 17 August 2021 (UTC)[reply]

In response to the

August 16 incident (permalink), and subsequent pings I've received on several of platforms, I've gone ahead and changed the behaviour of User:MusikBot II/TemplateProtector to elevate protection levels based on the configuration. Prior to today, the bot only looked for templates that were completely unprotected. In the past this was mainly a necessity for performance, but I believe this is no longer a problem

. After today's events, I hope we recognize the necessity for this extra layer of automation.

That said, prior to running the bot, I tediously reviewed the full list of affected pages. The ones that have had their protection changed I'm fairly confident in. I either verified few or no previous editors would be unable to edit it, or that the content rarely if ever changes (i.e. redirects). For the few outliers, I used a little

extended-confirmed protection

so that prior editors can still contribute. A few others I added to the bot's exclusion list so they will forever be ignored.

Which brings me to my next question, should we allow preemptive use of ECP on templates? Currently the

policy states Extended confirmed protection should not be used as a preemptive measure against disruption that has not yet occurred. I agree with this sentiment wholeheartedly, but the situation is obviously different for templates and modules that are vulnerable to causing massive disruption. Take Template:Australian politics/party colours

as an example. It is regularly edited by experienced users. All are extended confirmed, but none are template editors, and probably wouldn't become one solely to edit a template like this. The transclusions meanwhile are political in nature and hence are higher risk for targeted vandalism. So extended confirmed seems like a very nice compromise, which is why I applied it in this case (the bot would have otherwise raised to template protection).

If we do formally permit use of ECP for preemptive purposes, I suggest modifying the bot configuration to say, >500 transclusions for autoconfirmed (status-quo), then >5000 for extended confirmed, and perhaps >8000-10000 for template editor (currently set to >5000). With ECP in the picture, that should in my opinion take out a lot of potential for vandalism. However it's worth mentioning that, to my knowledge, since February 2019 there have been a but a handful of cases where someone lowered from template protection after the bot applied it. So that suggests 5000 is reasonable for template editor, in which case you might put ECP at 3000 or something. But we have to agree on using ECP first :)

Thoughts? Previous discussion on this topic can be found in the original bot proposal. Warm regards, — MusikAnimal ^talk 02:38, 17 August 2021 (UTC)[reply]

• I agree that this would be a reasonable compromise, especially for templates that are more content-oriented than technical. I'd also be interested in the viability of, for some highly visible content-oriented templates, breaking them into a TPE-protected template that covers the logic and XCP-protected subtemplates that cover the content (maybe with a mediating Lua module to sanitize potential attempts at CSS vandalism). But that's just spitballing there. -- Tamzin^{[cetacean needed]} (she/they) 03:43, 17 August 2021 (UTC)[reply]
• I also agree that this makes sense. HighInBC ^{Need help? Just ask.} 03:59, 17 August 2021 (UTC)[reply]
• Thanks for all the work and I support preemptive ECP. Perhaps there should be an exclusion method to allow only semi-protection for templates/modules in a certain category, but the exclusion would be ignored if the translusion count is above some threshold. The numbers proposed for the thresholds are too high. We don't want 3000 articles easily damaged by one malcontent. Johnuniq (talk) 04:04, 17 August 2021 (UTC)[reply]
• Thanks for your work on this, MusikAnimal. I also support ECP, and I concur with Johnuniq that the thresholds are too high. The 2018 RfC found consensus for semi-protecting above 200–250, so I'd suggest that as a better cutoff. I think often, as humans, we're bad at grasping what big numbers mean, so here's a way to look at it: How would we feel if an unconfirmed or IP editor decided to rapidly make 500 similar edits to a set of pages? Even if they were good faith, I think the response would be (ideally a friendly version of) "woah, slow down a bit, take some time to get acquainted with things around here before doing something that substantial". That isn't quite analogous to editing a template with 500 transclusions since the latter is more easily reverted, but still, we're talking about the same impact, and it should come with a similar level of restriction to prevent disruption. {{u|Sdkb}} ^talk 04:41, 17 August 2021 (UTC)[reply]
• I support these measures and also feel that the 3000-threshold should be lowered. Schwede66 04:56, 17 August 2021 (UTC)[reply]
• Yes, absolutely. I'm not seeing a significant downside here. If we have pages we're reluctant to use TP on, this is a good compromise. As with Schwede66 above, I would support lowering the threshold. Vanamonde (Talk) 06:26, 17 August 2021 (UTC)[reply]
• Support pre-emptive ECP at whatever threshold is deemed appropriate within the 4-digit range. ECP is widely available, so this should not be a significant block on updates. If such a change would allow for an increase in the Template protected threshold, so much the better as that remains quite an exclusive right. CMD (talk) 09:33, 17 August 2021 (UTC)[reply]
• Yes; I would even put the threshold for ECP a little lower, tbh, to around >1500. Thanks for the hard work, MusikAnimal. Lectonar (talk) 10:02, 17 August 2021 (UTC)[reply]
• No, there was an RfC prohibiting the use of EC here, for good reasons (like EC doesn’t correlate well with template writing ability). Wikipedia:Requests for comment/Extended confirmed protection policy 2. At minimum there should be another RfC on this, rather than an AN discussion. The thresholds of EC are ridiculous anyway, and it’s overly prohibitive in the topic areas on which it’s blanket applied ProcrastinatingReader (talk) 10:11, 17 August 2021 (UTC)[reply]
  • Also, would non-EC TEs now also need EC to edit ‘less high visibility’ templates?
    I’d be more happy with this proposal if EC could be granted at WP:PERM, so editors with less than 500 edits can request the perm if they have a need to edit one of these templates. ProcrastinatingReader (talk) 10:27, 17 August 2021 (UTC)[reply]

    "EC doesn’t correlate well with template writing ability" - It's not the correlation to template writing which is the issue. Some vandals, as yesterday's episode proved, are quite adept. The key is that EC has some correlation to a common investment in the Wiki. Cabayi (talk) 10:33, 17 August 2021 (UTC)[reply]

    It goes over the top. It may well exclude all vandals, but it excludes legitimate editors too. One editor adding Nazi symbols shouldn’t cause us to knee-jerk stop the thousands who edit them productively. Let them request it at WP:PERM. ProcrastinatingReader (talk) 10:35, 17 August 2021 (UTC)[reply]

    I'd like to see some actual examples of editors who would be shut out and

    bitten by this. As a general rule of them, somebody with less than 30 days' tenure and 500 edits is only interested in writing articles or improving content. The back end work of templates is only of interest to experienced editors who've been around the block a bit. Ritchie333 ^{(talk) (cont)} 10:42, 17 August 2021 (UTC)[reply

    ]

    There's not really a roadmap like that. Some editors only do anti-vandalism, some editors only write templates, some editors only write content, some do a mix of them all. I think we can all think of editors in each of these groups. I'd have been shut out by this, if I didn't happen to have my account registered earlier than I actually started editing. If I registered slightly later, would I have been any more of a vandal?

    Also it only just occurs to me... The two things that fix this Nazi issue are: 1) the protection bot not skipping already-protected pages, which is critical; 2) the change to the filter MusikAnimal applied. Those two things have already been done. It's not clear what this third change is preventing. The current config TE-protects >5000. This is proposing changing that to 8000-10000, and doing ECP for >5000. So... we're proposing letting more editors edit through the protection? ProcrastinatingReader (talk) 11:05, 17 August 2021 (UTC)[reply]

    Agreed with Ritchie. Additionally, anyone who cannot edit a template directly can always make an edit request on its talk page. Doing that from the view source page is not hard (and if someone finds it hard...maybe they aren't ready to be editing templates yet). {{u|Sdkb}} ^talk 11:06, 17 August 2021 (UTC)[reply]

    I remember before I had the TE perm it was pretty frustrating having to make edit requests. Anything more complex takes ages to get reviewed, or just doesn't get reviewed, because someone else (likely someone unfamiliar with the template) has to take responsibility for the code being functioning and fairly representing consensus, which sucks up so much time even reviewing them. I suspect that'll be less-so for ECP, since a lot more editors have the perm, but at the same time only a subset of ECP editors will review requests to edit templates (likely the same subset as those who are TEs, so similar numbers). There's also various changes one wouldn't want to approve but also wouldn't revert, and those are all incremental improvements. I don't think being able to edit a sandbox and request an edit is a replacement for being able to edit directly. ProcrastinatingReader (talk) 11:21, 17 August 2021 (UTC)[reply]

    ProcrastinatingReader, [a]lso, would non-EC TEs now also need EC to edit ‘less high visibility’ templates. I honestly cannot see a case where someone gets TPE before becoming extended-confirmed. firefly ( t · c ) 11:24, 17 August 2021 (UTC)[reply]

    Firefly, Well I suppose a sock of a banned editor might want to give it a go, but hopefully they'd be spotted. :-) Ritchie333 ^{(talk) (cont)} 11:52, 17 August 2021 (UTC)[reply]

    Ritchie333, indeed, I can understand that people might request it but as you say I doubt it'd actually be granted absent exceptional circumstances! firefly ( t · c ) 13:18, 17 August 2021 (UTC)[reply]

    @

    WP:PERM/EC. Also remember this isn't about template writing ability, it's about guarding highly visible templates from disruption. Many of these templates require no template editing skills whatsoever, but any intentional vandalism will have far-reaching effects. Right now we go from semi to template protection, which is quite a big jump. It'd be nice to have an in between in there. The 2016 RfC also predates the many waves of template vandalism we've had. I do not believe the points raised there have merit today, which is why many admins, myself included, have continued to use ECP on templates when it made sense to do so (high number of transclusions, but the regular editors are not TEs). I don't have the time or energy for another RfC, so if we feel that is needed, someone else will have to spearhead that effort. — MusikAnimal ^talk 15:37, 17 August 2021 (UTC)[reply

    ]

    People can request it for alt accounts. All other requests are denied. So I don’t think it really counts. As I understand, the change you made to your bot would’ve prevented this Nazi issue, as it has 55k transclusions, so I don’t see the need for even more layers of pre-emptive protections when the problem has already been patched. The only other reasonable idea is to TE protect, in addition to the automatic protections, any simple template that realistically does not need to be changed. (eg a hypothetical 1k transclusions variant of Template:Wbr.) But EC protecting basically anything meaningful in the template space seems overkill. ProcrastinatingReader (talk) 15:46, 17 August 2021 (UTC)[reply]

    WP:IAR does. If there were a situation where it would clearly benefit the encyclopedia for a non-EC editor to start editing ECP templates, it would be within standard administrative discretion to grant them the perm. I do know of at least one newer user who was offered EC for merit by several admins and declined. (An eon ago in 2012, I was given confirmed at the 3-day mark and rollback at the 13-day mark; that probably doesn't happen anymore.) -- Tamzin^{[cetacean needed]} (she/they) 18:00, 17 August 2021 (UTC)[reply

    ]

    @Tamzin: Can you link me a successful EC request by a non-EC editor (with no EC alts, and not a case of ECP being granted after it was initially pulled for gaming the system)? Ideally in the past year or two. ProcrastinatingReader (talk) 18:15, 17 August 2021 (UTC)[reply]

    Looking back a few months I don't see any, but that's not my point. My point is that it's allowed by policy, and in a scenario where a good-faith new user starts flooding the system with edit requests for ECP'd templates, any admin has the authority to grant them the bit. It's just that that's a rather unlikely scenario to occur. Under the current use cases for ECP, on the other hand, it's hard to think of a need-based reason to grant the bit, especially since, IIRC, granting ECP to a not-usually-eligible editor doesn't exempt them from DS/GS 30/500 sanctions. -- Tamzin^{[cetacean needed]} (she/they) 18:34, 17 August 2021 (UTC)[reply]

    @ProcrastinatingReader: I'm not going to go dig through my logs, but I have rarely given out ECP early, think it was related to a well established editor on another project wanting to do content translation here. It came with a strong reminder that it does not override the 500/30 arbcom remedies that they should now take extra care to avoid. — xaosflux ^Talk 19:00, 17 August 2021 (UTC)[reply]

• The more important decision is to let the bot re-assess templates which have already been protected. Semi protection on a template which has crept up to 10 times the threshold for TE protection is a major red flag (without any embellishment). Cabayi (talk) 10:24, 17 August 2021 (UTC)[reply]
• Oppose for all the reasons ProcrastinatingReader raised. I think Cabayi's idea is a good one, though. Jo-Jo Eumerus (talk) 10:31, 17 August 2021 (UTC)[reply]
• Support increasing the preemptive usage for security reasons, and also encouraging editors to get more involved with Template editing practices in general. Not discussed here, but of inverse concern are lightly transcluded templates, that are added to several popular pages, but I don't have a policy change for that. Shushugah (he/him • talk) 10:30, 17 August 2021 (UTC)[reply]
• Comment: does anyone know how long the disruption yesterday actually went on for? It seems to me that a large part of the problem is that vandalism can be undone on the template within 5 minutes but may not quickly propagate to all of the pages it's transcluded on for caching reasons. Anyone can manually purge a page but is there a tool to automatically purge all pages in a particular category or list (here, the list of transclusions)? If not, can one be created or does that open us up to some
  denial of service attack?

  If vandalism like this can be undone in 5 minutes then I will likely oppose an increase in stringency, but if there's no other way to stop a repeat of yesterday then I'll likely support it. Notice that all we needed to stop this attack was the MusikBox elevation fix (which I completely support). — Bilorv (talk) 15:39, 17 August 2021 (UTC)[reply
  ]

  Yes,

  certain types that take longer (those types are not applicable here). ProcrastinatingReader (talk) 15:48, 17 August 2021 (UTC)[reply

  ]

  Okay, so was it really just a few minutes or so (5? 10? 20?) that the vandalism yesterday was up for? Also, just spitballing, but has anyone considered some method of bot protection that gets to the root cause of the aim of this vandalism—to damage highly-viewed pages? A vandal wouldn't do this on 50,000 less-than-a-view-per-day non-mainspace pages, I would guess. We could have a cascading protection like we use for the main page but on a lower scale, applied by bot to anything with above X views in the last month. Just interested in the feasibility. — Bilorv (talk) 16:06, 17 August 2021 (UTC)[reply]

  Fucking magnets, how do they work? El_C 22:49, 17 August 2021 (UTC)[reply

  ]

  @Bilorv: I suspect, after the change was reverted, that propagated to the pages within a minute. (It would take my bot 1 minute to purge 55k pages; job queue is currently not backlogged, and rarely is, if the statistics are accurate.) As for your idea, it sounds feasible. One possible approach: WMF provides this data dump that can be used. Using that, one can bulk fetch the templates used by each page above X pageviews and then sum it all up. ProcrastinatingReader (talk) 14:07, 18 August 2021 (UTC)[reply]

  Yes

  WP:VRT and at least one press enquiry (just counting the ones I saw & handled myself), 2+ press reports, and the reputational damage to Wikipedia. The loss of 5 minutes effort was the least of the damage. Cabayi (talk) 09:51, 18 August 2021 (UTC)[reply

  ]

• Support. Good stuff, MusikAnimal. Make sure to give the bot extra treats. El_C 17:06, 17 August 2021 (UTC)[reply]
• An update to our
  WP:AN is not that location. --Izno (talk) 19:19, 17 August 2021 (UTC)[reply
  ]
• Support and I'm frankly surprised this wasn't already done. Arguments about how this might bite new template editors aren't compelling: you can be patient and get your 500 edits! It's not hard! Further, we should do what we can to minimize harm done to readers and our subjects (especially BLP pages); "accidentally" suggesting that someone is a Nazi in a BLP seems like a bright line we shouldn't be crossing - or even provide the road that could be used to cross it. Jorm (talk) 19:33, 17 August 2021 (UTC)[reply]
• Support. If they'd done this on some template that is used on fewer, it probably wouldn't have been caught for much longer. I get that this means there are some people who could make positive direct contributions and instead for a short time will have to make edit requests, but that's true for many articles, and those are single articles. —valereee (talk) 21:11, 17 August 2021 (UTC)[reply]
• Support reasonable actions to make it harder to vandalize thousands of our most highly viewed pages in a few mouseclicks. Unprotected templates transcluded into huge numbers of highly salient pages, especially BLPs, make mass-scale vandalism ridiculously too easy and common. - Astrophobe (talk) 21:31, 17 August 2021 (UTC)[reply]
• Support this makes sense. Also support lowering the threshold for ECP to 3000 pages (or even lower). Jake Wartenberg (talk) 21:38, 17 August 2021 (UTC)[reply]
• ’’’Support’’’ templates are highly visible so the impact of vandalism / test edits / not understanding template workings/code or formatting is high. I support lower thresholds for this reason: eg > 100 transclusions for autoconfirmed > 250 for EC and > 1000 for TE. Tom (LT) (talk) 22:00, 17 August 2021 (UTC)[reply]
• Support. Our long-standing policy on High-risk templates (and it is a policy despite its guideline tag) pre-dates the EC policy by a long way. I'm never a fan of specifying exact thresholds for these things, but should we use EC for high-risk templates where appropriate? Sure. -- zzuuzz ^(talk) 22:18, 17 August 2021 (UTC)[reply]
• Support protection, regardless of whether we classify it as ECP. Regarding "high-risk", I'd go so far as to say that the template namespace should be protected by default, with a whitelist to allow editing e.g. DYK nominations. Has anyone ever seen a new user make meaningful positive contributions to any template such that it wouldn't be better handled through a protected edit request? — Rhododendrites ^talk \\ 22:35, 17 August 2021 (UTC)[reply]
  • Oh, lots. Some templates are maintained almost entirely by anons - notably sports templates but also many other types. To get some idea of the massive backlogs and content deficit that would create, take a look at recent changes. -- zzuuzz ^(talk) 22:46, 17 August 2021 (UTC)[reply]

    I agree that protecting the entire template space would be a significant change and that it's not unusual for templates such as sports season standings to be updated productively by new or unregistered editors. Plus since transclusion from any namespace is permitted, editors would just start putting more templates into project space. isaacl (talk) 23:14, 17 August 2021 (UTC)[reply]

    It might be worth having an edit filter that watches for new images being added to templates by non-EC editors. If we can't restrict access to the whole namespace we can at least make oversight easier. I'll run it over to

    WP:EFN. — Wug·a·po·des​ 00:30, 18 August 2021 (UTC)[reply

    ]

    Thanks for the link. I guess they're just not in any of the topic areas I watch. — Rhododendrites ^talk \\ 19:02, 18 August 2021 (UTC)[reply]

• Support, especially for content templates like {{Color topics}} where template-protection may unduly interfere with productive revisions. That template has 233 transclusions in mainspace, just short of the proposed EC cutoff proposed by Tom (LT). –LaundryPizza03 (dc̄) 01:16, 18 August 2021 (UTC)[reply]
• Support for so many reasons it would take all day to list them. Worthwhile, though, to verify that the template editor permission will allow editing of ECP-protected templates. There may be occasional rare exceptions where this is warranted. Risker (talk) 02:42, 18 August 2021 (UTC)[reply]
• Support This is the only viable alternative to more extensive use of full protection. I think however that admins will need to be reminded they can grant EC. I had forgotten until I saw this here. DGG ( talk ) 03:24, 18 August 2021 (UTC)[reply]
• Support as there is a clear need for templates with more than auto-confirmed and less than template-editor protection. The language Extended confirmed protection should not be used as a preemptive measure against disruption that has not yet occurred should be removed from the ECP policy page. User:力 (power~enwiki, π, ν) 03:54, 18 August 2021 (UTC)[reply]
• Support - as a very reasonable compromise between semi-protection, and TPE protection (which it seems would impose barriers to some content templates being updated). I am also very glad to see that MusikBot II will now raise protection levels as transclusion counts increase, given the performance issues have been elided. Nice work MusikAnimal :) firefly ( t · c ) 06:24, 18 August 2021 (UTC)[reply]
• Support I have long supported this and believe it could have a significant impact on reducing template vandalism while minimizing the harm for legitimate users. --Trialpears (talk) 08:39, 18 August 2021 (UTC)[reply]
• Support the introduction of ECP as an extra level in protecting templates. Oppose the increase in threshold for TE protection. Support the bot re-assessing previously assessed templates. Noting the need for the bot to respect manual adjustments to the protection, preferably it would flag up somewhere (pinging the adjusting admin) that the protection level is out of sync with the norm. Also noting that many sensitive templates are substed & that the transclusion numbers are not the whole story. Cabayi (talk) 09:57, 18 August 2021 (UTC)[reply]
• Support per above. ― Qwerfjkltalk 10:06, 18 August 2021 (UTC)[reply]
• Pile on support per above.--
  John Cline (talk) 13:21, 18 August 2021 (UTC)[reply
  ]
• Oppose - I realize I'm coming to this late and it's already
  snowing, but hear me out. ECP is a game-able protection level - it takes a more dedicated vandal to game it but trust from my experience at SPI that it's very little deterrent over semi, and with templates the payoff is having your vandalism immediately appear in maybe tens of thousands of articles all at once. Plus, the disruption you cause still produces ANI complaints three days later and serious reactions like this proposal here with widespread ramifications. In exchange for 500 nonsense edits (or one edit plus a script that clicks undo 499 times over a few days) and a 30 day hold (set a reminder in your calendar), this would've been a pretty sweet deal for our recent Nazi vandal. On the other hand, template protection secures high-risk templates to only editors who have been vetted by other members of the community, which is a process that's far more difficult to game. If template vandalism is becoming a widespread problem (I would argue it has been for quite some time already) then we should consider some form of pending changes protection for templates that appear in article space. Ivanvector's squirrel (^trees/_nuts) 13:55, 18 August 2021 (UTC)[reply
  ]
  • If I understand correctly (in no way a given!), this proposal is whether to allow ECP on templates, with the specific levels to be determined later. On re-reading the proposal it seems that the example given would indeed lower the protection of some templates - which I would disagree with were it proposed formally. I'd probably suggest keeping the current thresholds and then inserting another at 2500 transclusions for ECP. MusikAnimal - thoughts? :) firefly ( t · c ) 14:45, 18 August 2021 (UTC)[reply]

    Yes, we will decide on adjustments to the thresholds in a separate discussion (I don't think that requires an RfC). Regardless, even if the bot doesn't use ECP, I would still like to see use of it in preemptive fashion written in our policy. As noted above, Template:Australian politics/party colours is a prime example, given the usual template editor protection would have shut out all editors. — MusikAnimal ^talk 16:13, 18 August 2021 (UTC)[reply]

    • Agreed, and I assumed as much. Thanks! firefly ( t · c ) 16:46, 18 August 2021 (UTC)[reply]

• Support. It seems like it would be sensible to have another level of protection available here, since the two current options are basically "everyone can edit" and "only administrators + a very small number of users can edit". Yes, of course it is possible to game the EC permission, but I don't see how restricting ourselves to a permission that is even easier to game is the right solution. 192.76.8.74 (talk) 21:28, 18 August 2021 (UTC)[reply]
• Support. This seems reasonable, especially if it helps to prevent the mass vandalism we saw yesterday. Clovermoss (talk) 22:29, 18 August 2021 (UTC)[reply]
• Support entirely reasonable proposal.-- Jezebel's Ponyo^{bons mots} 16:38, 19 August 2021 (UTC)[reply]
• Support. This proposal could enable us to reduce template vandalism, even if it doesn't eliminate it. And anything that reduces vandalism is still worth doing, per the nirvana fallacy.— Shibbolethink ^{(♔ ♕)} 15:31, 22 August 2021 (UTC)[reply]
• Support adding ECP as an intermediary step, but Oppose raising the 5000 transclusion threshold for TEP. I think something like the 250/2500/5000 proposed above would make more sense than lowering the protection levels on highly-used templates. --Ahecht (TALK
  PAGE) 21:30, 23 August 2021 (UTC)[reply]
• Support Making ten edits and waiting four days? Easy. Making five hundred edits and waiting thirty days? Not so much. 🐔 Chicdat  ^{Bawk to me!} 10:31, 24 August 2021 (UTC)[reply]
• Support — As for Ivanvector's oppose, I think that instead of lowering the threshold for template protection, this should mainly be used for templates that are currently semi-protected but are high-risk enough to warrant extended-confirmed protection. Tol (talk | contribs) @ 20:25, 1 September 2021 (UTC)[reply]
• Support with a certain transclusion count or otherwise high risk such as the template appears on controversial or popular topics. Crouch, Swale (talk) 20:51, 3 September 2021 (UTC)[reply]

  Support: Could be useful as another option for those templates in a dead zone of high use; too highly used to use semi, but not high enough to necessarily use template protection. Regards,

  Contact me | Contributions). 21:44, 3 September 2021 (UTC)[reply

  ]

Discussion (ECP & templates)

• I was just about to ask someone close this, because I don't have the stomach for a proper RfC, but it looks like we're doing it anyway! For those of you who are just now seeing this thread: This started as an update about the bot I maintain, in response to a recent vandalism incident. I was gauging interest on using ECP, as many of the articles that would have been raised to template protection by the bot seemed to call for it. So anyway, for the purposes of this RfC, !voters can ignore all the proposals regarding the bot such as what thresholds we will use. That can be ironed out later once consensus for the use of ECP is established. Thanks, — MusikAnimal ^talk 21:20, 17 August 2021 (UTC)[reply]

  Thanks for the clarification. I was about to say essentially what you said about dealing separately with any changes to the bot. I have a personal bias that appreciates having a step between semi-protected and template-editor protected, so this may be hindering me from evaluating the downside of using extended-confirmed protection. (There is a module I wrote, before the template editor role existed, that later got template-editor protected; an admin kindly lowered it and the corresponding template to semi-protected so I could continue to support it as needed. I know it could get raised back to template-editor protected any time and then I'd have to rely on edit requests. It's not a huge deal and changes are infrequent, but it's a potential future impediment.) isaacl (talk) 21:37, 17 August 2021 (UTC)[reply]

• Comment - although ECP is certainly a lot safer than semi-protection, it's still something a dedicated vandal could overcome without anyone else needing to cast eyes on them. Signing up a new account, doing 500 edits, and waiting 30 days, is not beyond the realm of imagination. Wouldn't it be better for us to use template protection for these high-visibility templates? Obviously this will cause irritation for other legitimate users wishing to make tweaks to those, but I think that's a better solution than risking the 16 August fiasco again.  — Amakuru (talk) 08:14, 18 August 2021 (UTC)[reply]

  Some vandals might go to that extent - we always see it elsewhere - although it is definitely a significant barrier. Many will be caught on their way to EC, but some will get there. At least we might get a few typos fixed in the process. We've previously had administrators who are mass sockpuppeteers and some who have deleted the main page, so nothing will ever be perfect. It's just a matter of getting the cost/benefit balance right, and this proposal gives us an extra option to do that. -- zzuuzz ^(talk) 09:02, 18 August 2021 (UTC)[reply]

• Please excuse me if I've missed some discussion, I haven't read everything related to this incident. I know MusikBOT doesn't escalate protections to avoid the bot overriding admins delibretly setting a lower protection level. Wouldn't it make more sense to control these cases using the {{bots}} template to prevent the bot from changing it in those cases? --Trialpears (talk) 08:34, 18 August 2021 (UTC)[reply]

  By "using the template", I'm assuming you mean {{bots}}? I don't think this task should be exclusion compliant in this way because a vandal could also do this to prevent it from being protected, say right before they transclude it into a much more visible template (which should also be automatically protected, but for the sake of the argument let's pretend that's not the case). The bot only ignores pages with recent page protection changes in the past 7 days, according to the ignore_offset option in the User:MusikBot II/TemplateProtector/config. We don't want this to last for eternity because humans won't bother to go back and check that a template with just 500 transclusions now has several million. Humans forget, the bot will not :) However admins can also explicitly exclude specific pages via the config, so there's still a means to make the bot ignore a template. — MusikAnimal ^talk 15:53, 18 August 2021 (UTC)[reply]

  That's indeed the template, just forgot the {{

  t}}. You're probably right about the bot being exclusion complaint isn't a great idea, but your config solution seems sufficient. The ignore offset of 7 days also seems find, but then why didn't it escalate protection at {{wbr}} and others mentioned in related discussions. --Trialpears (talk) 21:31, 18 August 2021 (UTC)[reply

  ]

  Up until yesterday, the bot only acted on templates that were completely unprotected. Now it will escalate from semi to template editor in accordance with the configuration. The technical means to do this was only made possible some months ago. So while the bot came too late to the rescue in this case, we should at least be glad our new database replicas are fast enough to make this functionality possible :) — MusikAnimal ^talk 03:37, 19 August 2021 (UTC)[reply]

  I appreciate the desire to review the transclusion count periodically, but am not sure if a seven-day minimum period is long enough. For bot-set protection, it doesn't matter, but if an admin set the protection level, surely they considered the future potential usage of the template for more than the next week? isaacl (talk) 20:17, 19 August 2021 (UTC)[reply]

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

If you visit https://en.m.wikipedia.org/wiki/Mona_Lisa while logged in, you'll see the "Article Talk" tab bar. ("minerva__tab-container" element) But if you're logged out, this tab bar goes missing and there's no way to access the talk page unless you manually alter the URL. This turns out to be by design and now consensus is required for a configuration change to ensure that anyone who can edit can also reach talk pages. — Alexis Jazz (talk or ping me) 11:29, 22 August 2021 (UTC)[reply]

• Support as proposer. — Alexis Jazz (talk or ping me) 11:29, 22 August 2021 (UTC)[reply]
• Support, Obviously. Sometimes I really do question if the WMF understands the site that they run, this is supposed to be a collaborative encyclopaedia building site - how on earth is collaboration supposed to occur when you hide the main methods of communicating from the vast majority of users? The requirement for consensus seems to be backward here - removing the ability for anonns to communicate on mobile should have required community engagement, restoring longstanding functionality should be the default. The rationale for removing talk page access for all mobile annons seems to be a combination of "mobile IP users are obviously too stupid to understand the concept of a talk page and would just fill them up with nonsense and spam" (I'd like to see the data that supports this, according to the phabricator task this is based on them putting a talk page link in a rarely visited settings page and being surprised that people didn't understand what it did???) and that the mobile talk pages are too buggy to be publicly available (In which case they need fixing, not hiding.). I just can't see the sense in having a huge body of users who are allowed to edit the site but are barred from discussing their editing - it just seems dumb. 192.76.8.74 (talk) 12:42, 22 August 2021 (UTC)[reply]

  For background, talk pages were never removed, rather they were added. The mobile site was built from the ground up. In fact, we didn't have editing for a long time as it was communicated to WMF by editors that notifications were mandatory before we could do that. We added notifications, then editing. Talk page access was only added in current form circa 2019 because communities spoke up and helped drive that priority.

  Building the mobile site has always been based on priorities, so nobody in WMF has ever removed this functionality as you are alleging here. Please assume good faith. Our software is extremely complex with very few maintainers. Jdlrobson (talk) 19:16, 23 August 2021 (UTC)[reply]

• ? @Jdlrobson: - any reason someone is going to try to block setting $wgMinervaTalkAtTop = [ "base" => true ]; if the enwiki community asks for it? — xaosflux ^Talk 15:49, 22 August 2021 (UTC)[reply]

  Replied with suggestions: https://phabricator.wikimedia.org/T54165#7301936 Jdlrobson (talk) 19:17, 23 August 2021 (UTC)[reply]

• I'm a bit torn here, in that I only partially agree that this is supposed to be a collaborative encyclopaedia building site. The primary purpose of this site is to be an encyclopedia serving our readers. Obviously, to have an encyclopedia, you need to write an encyclopedia, but the writing is not the primary purpose. It's just a means to the end. Screen real estate on a mobile device is precious, and wasting pixels on a feature that's mainly of benefit to writers degrades the experience of our readers. -- RoySmith (talk) 16:41, 22 August 2021 (UTC)[reply]

  RoySmith, that's a valid argument, but you can't have it both ways. Either mobile anon users can edit (which they currently can) and they need to be able to reach talk pages. Or you remove their edit button (that saves even more pixels) so they won't have a pressing need for talk pages anymore, but that's a difficult topic. — Alexis Jazz (talk or ping me) 17:09, 22 August 2021 (UTC)[reply]

  I'd be happy to see all anonymous editing go away. Want to edit? Make an account and the editing buttons and links become visible. IMHO, the amount of effort we put into making anonymous editing work isn't justified by the value it adds. For example, the whole "IP Masking" effort that's currently underway. -- RoySmith (talk) 17:41, 22 August 2021 (UTC)[reply]

  RoySmith, I don't oppose, but unless/until we manage to establish a consensus to end IP-editing (this recent effort went nowhere) we have to deal with it. So that's what I'm trying to do. As long as they have an edit button, they should have a talk page link. — Alexis Jazz (talk or ping me) 18:04, 22 August 2021 (UTC)[reply]

  @RoySmith and Alexis Jazz: I don't think we should make the fork for showing editing features logged in vs. not. We want users to log in even if they never plan to edit so that if they do decide later on to fix a typo it'll be a smoother journey from there down the editing rabbit hole. But if their first reaction when they create an account is "ack, this just introduced all these ugly buttons I don't want", they'll never stay logged in. {{u|Sdkb}} ^talk 07:23, 24 August 2021 (UTC)[reply]

• We should block all edits from all platforms that do not have full access. —Kusma (talk) 16:50, 22 August 2021 (UTC)[reply]
• Support per nom with my thanks for stewarding this issue. Not giving mobile IP editors access to talk pages is just stupid, I have no other words for it. WP:Communication is required. Levivich 06:13, 23 August 2021 (UTC)[reply]
• Support Paid employees have missed the boat here. An IP user edits, is constantly reverted, but doesn't see explanations on their talk page. Effectively a campaign to frustrate and drive away IP editors, while further stigmatizing IP users to the community.—Bagumba (talk) 11:13, 23 August 2021 (UTC)[reply]
• This is now a forked discussion, can I suggest redirecting this back to the
  WP:VPP section, as that is also diverging into something separate from what the section author originally wrote. ProcrastinatingReader (talk) 12:34, 23 August 2021 (UTC)[reply
  ]

  This is a specific request for the mobile platform software, the other is a high-level policy proposal.—Bagumba (talk) 14:23, 23 August 2021 (UTC)[reply]

  I agree with Bagumba: this is a specific proposal for a skin and should be allowed to reach its own conclusion. The other discussion is a broader one covering all clients. Regardless of its outcome, the skin can be changed as per community consensus. isaacl (talk) 15:20, 23 August 2021 (UTC)[reply]

  @RoySmith: oppose early closure. This is a specific proposal to do a specific actionable thing. The thing is boolean as well so it is very easy to determine the result. That VPP discussion is extremely broad and vague. — xaosflux ^Talk 15:49, 23 August 2021 (UTC)[reply]

  No problem, I've backed out my close. -- RoySmith (talk) 16:54, 23 August 2021 (UTC)[reply]

• Support If a given setup supports editing, it should support talk pages.
  WP:ENGAGE. MarioGom (talk) 16:59, 23 August 2021 (UTC)[reply
  ]

• 

  Question I agree that giving everyone (including logged-out folks) access to the talk page is a good idea. However screen space is indeed limited on mobile, as RoySmith mentioned, and we know from our data that most logged-out people are not visiting Wikipedia with the intention of editing. They want to read articles, and the Article Talk tabs push the article down the page a bit, so does that design align with what they want? Additionally I wonder if the "Talk" tab, as it is currently presented to logged-in folks, might be confusing logged-out folks and newcomers in general (i.e. what does "Talk" mean? what will happen if I tap on it?)? Which brings me to these mockups made a few years ago by User:Npangarkar_(WMF), of an expanded article footer that has additional tools and information (including a link to the talk page). I feel like the inclusion of an icon, and the more descriptive title ("Discuss" vs. "Talk") make it easier for newcomers to understand what they might find if they tap on it. It could be even more helpful if the button/link included the number of discussions (something like "2 active discussions"), which was an idea explored in Winter. AHollender (WMF) (talk) 18:40, 23 August 2021 (UTC)[reply]

  @AHollender (WMF): The note about screen real estate makes sense, but nobody really scrolls to the bottom either. There is a pencil icon on each section to edit it. Editing and discussion workflows should ideally go hand in hand, so if it's prominent and easy to edit there needs to be a complimentary method of discussion. Perhaps a button on the warning message that shows up when you click "Edit" is one way forward. For logged in users, the 'advanced mobile' mode (which shows the talk button) should perhaps be default, since most logged-in people do probably edit (?) ProcrastinatingReader (talk) 19:31, 23 August 2021 (UTC)[reply]

  That's a good point about wanting the workflows to go hand in hand, ProcReader. Maybe we want some more mockups of what that could look like.

  AHollender, regarding "discuss" vs. "talk", I believe the desktop tab used to say "discuss" instead. I'm not sure why it was changed, but that's probably a discussion to seek out. The biggest issue I tend to see is

  WP:NOTFORUM—it's not always intuitive that the discussion page is for discussing improvements to the article, not a general comment thread for discussion about the topic. We designed {{Talk header}} to help explain this, but the mobile developers got fed up with the banner bloat on talk pages and just shoved them into an "about this page" submenu no newcomer will ever click on, so ¯\_(ツ)_/¯. {{u|Sdkb}} ^talk 20:22, 23 August 2021 (UTC)[reply

  ]

  A common trap is for people to do mobile mockups in English, then be surprised when the German version is a mess because "Diskussion", "Quelltext bearbeiten" and "Versionsgeschichte" don't fit into the space allotted for "Talk", Edit", and "History". -- RoySmith (talk) 21:20, 23 August 2021 (UTC)[reply]

  It traditionally was Talk, then globally it was changed to "discussion" and then en.wp freaked out because "discussion" was too long (taking up more space in their monobook skin) and different from "talk" and they feared that "discussion" would invite discussions on the topic instead of the article, so en.wp changed it back to talk. —TheDJ (talk • contribs) 10:17, 24 August 2021 (UTC)[reply]

  People still use Monobook? ProcrastinatingReader (talk) 10:24, 24 August 2021 (UTC)[reply]

  Back then definitely. And the bar is all filled up with additional portlet actions for administrative actions as well, because dropdown menus are BAD :D —TheDJ (talk • contribs) 10:26, 24 August 2021 (UTC)[reply]

  I do. I try Vector once per year, but never enjoy it. Too much whitespace and all my buttons are hidden somewhere. —Kusma (talk) 10:36, 24 August 2021 (UTC)[reply]

  I absolutely do. Tried Vector for a while, but never could get used to it. Too much clutter. Monobook is clean and simple, and that's exactly what I want in an interface. Seraphimblade ^{Talk to me} 05:56, 25 August 2021 (UTC)[reply]

• Qualified support. This is a classic example of systemic bias toward editor desires over
  WP:READER desires. I agree with everyone that it's essential to have some way to access talk pages from every editable page, but I agree with RoySmith and AHollender (WMF) that we don't want to give it inappropriate emphasis. The mockup of what it could look like from the bottom of the page looks quite nice, although I'd want to see some more discussion around what we'd want to go in the edit overview section or whether we should just keep the current "last edited by JimboWales" bar. {{u|Sdkb}} ^talk 20:14, 23 August 2021 (UTC)[reply
  ]
  • Readers also use the talk page. An IP's angry post at Talk:F-777 prompted me to fix a redirect we didn't know was broken. — Wug·a·po·des​ 22:15, 27 August 2021 (UTC)[reply]
• Support, and (probably deserves its own section) switch from text labels on tabs, to icons with tooltips (ha-ha, the mobile folks won't see the tips). RoySmith's comment about the length of terms like Quelltext bearbeiten raises a good point, but the solution is to follow the example of European road signs and come up with good icons for 'Read', 'Edit', 'History', and so on, and relegate text equivalents to second place. I can visualize one for 'Talk' involving two little talking heads facing each other. It's well established^{[citation needed]} that people process images faster than text, and it would also be a god-send for those of us who occasionally wander off to other language Wikipedias in languages we can't read, and just want to check history, or find a related discussion or compose a diff or something. Mathglot (talk) 22:16, 23 August 2021 (UTC)[reply]

  Icons works well in addition to text, but on their own are known to sometimes increase confusion, esp the more there are and the more esoteric they become to try to convey the many meanings that have to be conveyed. Especially on mobile, where you have no hover labels etc this actually reduces discovery. Anyways, mobile already uses a lot more icons than desktop does. The Advanced mode for editors on mobile, currently has language, a watch star, a clock winding back, a pencil and a dropdown menu.... and I think most people on first glance have no idea what any of them do...... —TheDJ (talk • contribs) 10:25, 24 August 2021 (UTC)[reply]

• Support. However, as other editors have suggested, it might be better to find an alternative to the [Article | Talk] tabs to save some screen real estate. There's a lot of blank space between the language icon and the edit icon. Maybe slip in a Talk page icon in that row? Alternatively, the proposed Tools section at the end of the page looks great. - Wikmoz (talk) 04:56, 24 August 2021 (UTC)[reply]

Actually, it looks like the problem was already solved in the Wikipedia app. At the bottom of each article, there is an ABOUT THIS ARTICLE section with links to View talk page and View edit history. - Wikmoz (talk) 05:16, 25 August 2021 (UTC)[reply]

• Support as at least a start. Mobile or not, we should always make the channels of communication clear and easy to find, and treat every reader as a potential editor. Seraphimblade ^{Talk to me} 06:56, 24 August 2021 (UTC)[reply]
• Support. It doesn't make sense that a group that is allowed to edit cannot even view talkpages and edit-histories. Why not make pressing the existing "edit" button on the mobile site open up a sub-menu with the options "Edit article", "Go to talk page", and "View history"? I think it's unlikely that anyone technically inclined enough to edit a page would be confused by this additional step. Rabbitflyer (talk) 22:19, 31 August 2021 (UTC)[reply]
• Strong support - Talk pages are the backbone of Wikipedia. It would help me tremendously with my work on Wikipedia in mobile. If we did it in the userspace, we should do it to others as well. Interstellarity (talk) 16:34, 1 September 2021 (UTC)[reply]

Request from Editing Team

Comment: Hi y'all – I work as the product manager for the Editing Team. We're actively working on a series of improvements to talk pages on desktop, and within the next few months, we'll be shifting our focus to improving talk pages on mobile.

With this in mind, can you please review – what I currently understand to be – the issues you all have raised here and let us know what issues you think are missing from this list and/or what about this list needs to be edited?

For added context: I'm asking the above because it's important to me that our team accurately and exhaustively understand the issues y'all are raising here, so that we can make sure the issues we prioritize working on in the next few months are the issues that will be the most impactful to address.

Issues

1. Meta
   1. Talking is a core part of editing and there is a large segment of people who can edit and not talk. As 192.76.8.74^[1], @Alexis Jazz^[2], @RoySmith^[3], @MarioGom^[4], @ProcrastinatingReader^[5] , and others in previous conversations have articulated^{[6] [7] [8] [9] [10]}: collaborating is an important part of writing an encyclopedia. In order to collaborate, you need to be able to talk to people. Currently, there is a large segment of people (anons editing on mobile), who are: A) able to edit and B) not able to collaborate, by way of them not having access to talk pages.
      1. The above, "...wastes time and energy for editors to post explanations/guidance/warnings for contributors who cannot respond" @Johnuniq^[11]
      2. It also helps contextualize why anons editing on mobile not having access to talk pages is problematic.
   2. It's disconcerting to learn that we – volunteers and WMF staff – do not seem to share a core assumption/understanding that people who can edit, must also need to be able to talk.
2. Talk page design
   1. People do not intuitively recognize discussion pages as places to talk about improvements to articles.^[12]
   2. People accessing talk pages on mobile lack access to important context about the conventions that guide how these pages can be used constructively. E.g. Talk page banners/templates are difficult to discover and edit notices are absent.^[13].

Additional context

Considering there are three teams within the Foundation working on improvements to talk pages, I thought it would be worth making sure you all were aware of the work that is being planned and done to improve volunteers' ability to communicate with one another.

• Editing Team | Talk pages project
  • *Reply Tool: a way to reply to talk page comments in one click
  • *New Discussion Tool: an inline form for adding new topics with keyboard shortcuts for pinging and inserting links
  • **Notifications: subscribe to receive notifications about comments posted in specific topics/sections
  • Usability Improvements: a series of improvements to help people instinctively recognize and use talk pages as spaces to communicate with other people
  • Mobile: we'll be introducing all of the above on mobile as well.
• Android Team | Communication Improvements
  • Implementing talk pages and the Watchlist natively within the Android app
  • Improving notifications so people can be aware when others are contacting them
• iOS | Notification improvements
  • A series of improvements intended to help iOS participants to know when they have a new notification without them having to check Wikipedia’s website or check pages in the app, so that they can take timely action on their notifications.

*You can experiment with the Reply and New Discussion Tools right on desktop, by enabling the DiscussionTools beta feature in Special:Preferences.

**You can experiment with Topic Subscriptions on desktop by appending ?dtenable=1 to any talk page URL like this. PPelberg (WMF) (talk) 00:03, 28 August 2021 (UTC)[reply]

I inserted a subheading to make this more easy to respond to. A quick additional point is that there needs to be a way of alerting users of mobile devices that is different from the normal semi-spam that occurs with nearly all apps competing for user attention. I have seen phones where a dozen apps have badges showing notifications—the owner ignores them as background noise. For Wikipedia use, when a user opens their browser or app, and if there are outstanding talk-page messages, there must be something like the orange-bar-of-death that more or less compels the user to respond. The suggestion that real-estate on a phone is important completely misses the point that the first thing the user should do is at least view their talk. That raises another tricky point. By convention, we bottom-post, but on a phone that might require a bunch of scrolling. I'm not sure what to do about that but ideally the current sections would be shown. Johnuniq (talk) 00:18, 28 August 2021 (UTC)[reply]

I inserted a subheading to make this more easy to respond to.

Oh, wonderful! Thank you for doing that @Johnuniq.

As for the additional issue you are raising, I'm glad you drew out this nuance. You can expect a response from me about this point next week. PPelberg (WMF) (talk) 01:01, 28 August 2021 (UTC)[reply]

there needs to be a way of alerting users of mobile devices that is different from the normal semi-spam that occurs with nearly all apps competing for user attention.

@Johnuniq: would it be accurate for me to understand the issue that's prompted you to share the above as: "Anonymous editors do not respond to the messages others leave for them on their talk pages."

Note: I appreciate the language I proposed above does not include the solution/requirement you proposed. I've done this intentionally so as to ensure I'm accurately understanding the underlying issue any solution(s) would need to address.

By convention, we bottom-post, but on a phone that might require a bunch of scrolling.

This is a great callout and something we will need to consider as part of the work we have planned to help people identify new talk page activity. I've documented – what I understand to be – the question inside of what you are saying on Phabricator: "How might bottom-posting impact the likelihood that people accessing talk pages, particularly on mobile, will see the new messages others have left for them?". PPelberg (WMF) (talk) 01:47, 2 September 2021 (UTC)[reply]

@PPelberg (WMF): It's not quite right to sum up the issue that's prompted me in the terms above. First, some IP editors are experienced and do respond to talk page messages (so "do not respond" is over generalized). Second, that wording makes it sound as if the anonymous editor might have made a choice to not respond whereas my original concern was for the many edits made by mobile IPs and accounts where the UI does not show them that they have a talk message, and/or does not rub it in their face that the message might be something they "must" see as opposed to the normal spam from many social media platforms, and/or does not provide a simple way to respond. Finally, I am one of many editors who occasionally write detailed explanations for new users and it is destructive for editors like me to later learn that the recipient probably never even knew about my explanation. My point is that the UI is damaging the collaborative community because people like me now think that all IPs/accounts using inoperative software should be banned. Regarding "simple way to respond": I think there should be an orange bar with suitable wording that is hard to avoid. Clicking that bar would take them to the first section on their talk with the most recent date. I understand enough about code to know that would be difficult, but something much better is needed. Perhaps force all new contributors (with a cookie?) to follow a quick tutorial. That might be mandatory if any of their edits have been reverted. Thanks for taking the time to engage here. Johnuniq (talk) 05:11, 2 September 2021 (UTC)[reply]

"Anonymous editors do not respond to the messages others leave for them on their talk pages." I'd say "Anonymous users do not realize that there are messages/that others are trying to tell them something, and if by chance they do realize, they have a hard time responding to those messages/notifications" —TheDJ (talk • contribs) 20:43, 3 September 2021 (UTC)[reply]

I assume we're talking about talk pages only here? If so, I think this question is broken down into two parts. The first is deficiencies compared to the desktop editing experience. However, deficiencies compared to desktop is not the full list of problems with mobile talk pages or things that should be done differently on mobile, but I would have to think harder on that part to produce a list. One example: if you see a protected page and want to edit it, the workflow on enwiki is via a talk page edit request. It's a pretty poor UI esp on mobile (you can explore it by going to Donald Trump in incognito on your phone and trying to edit). Some of this falls on the community but there's not really a much better workflow possible without software changes. IIRC(?) not too long ago the "View source" button wasn't shown at all so it wasn't clear how to request a change, so I guess the situation is improving...

Although I suspect 'solutions' is separate to 'issues' I would like to take the opportunity to note, in regards to 2.2 (talk page banners), that I have some feelings on the usefulness of the banners on many talk pages. They're a mess, like look at this. It wouldn't be reasonable to show it all to mobile users, and very often there is nothing useful in them IMO (but I suspect I may be a minority view there). Conversely, see here for a useful non-generic banner and toggle "read as wiki page" on and off to see it. It's a hack used to make the message visible on mobile but apparently still only displays when you "read as wiki page"; I don't know why that feature even exists? If a solution to the talk banner issue can't be figured out, showing editnotices (somehow) would be a feasible solution. A talk page editnotice should only be placed when there's something substantive to say about that particular article. ProcrastinatingReader (talk) 01:02, 28 August 2021 (UTC)[reply]

The problem with not showing banners to mobile users is that we expect all users who use the talk page to have read them. It is not reasonable to expect Desktop users to be aware which banners are visible to other editors and which are not. —Kusma (talk) 19:09, 28 August 2021 (UTC)[reply]

But I agree that we have far too many banners (and many of them are useless). The whole Wikiproject and assessment stuff really should be in a "meta" area, not taking up space on a discussion page. —Kusma (talk) 19:16, 28 August 2021 (UTC)[reply]

@ProcrastinatingReader – I appreciate you sharing these thoughts. Comments and questions in response to the points you raised are below...

I assume we're talking about talk pages only here?

Yep, exactly.

if you see a protected page and want to edit it, the workflow on enwiki is via a talk page edit request. It's a pretty poor UI esp on mobile...

To confirm: are you referring to how people wanting to edit a protected page on mobile need to submit an edit request by way of starting a conversation on said page's talk page and that workflow not being straightforward? [i]

...the usefulness of the banners on many talk pages. They're a mess, like look at this. It wouldn't be reasonable to show it all to mobile users, and very often there is nothing useful in them IMO...Conversely, see here for a useful non-generic banner and toggle "read as wiki page" on and off to see it.

I've tried to put what you described into my "own words" to ensure I'm understanding this as you intended it. Can you please let me know if there is anything you would change in order for it to better reflect what you are communicating?

Peter's "own words": "Volunteers need to be able to display information to people, across devices, in ways that will enhance their understanding of: A) The subject page they are likely reading about and/or B) What to consider before participating on the talk page."

Also, these examples are great. Particularly the Talk:COVID-19 lab leak theory example. Thank you for sharing them; it's clarifying to be able to see what you are imagining in your mind.

---

i.The workflow for submitting an edit request as an anon volunteer on mobile, by my count, requires ~7 less-than-intuitive steps: 1) Click edit pencil, 2) Click the View source link that temporarily appears at the bottom of the page, 3) Scroll the page, 4) Locate and click the Submit an edit request button, 5) Scroll to the bottom of the talk page's source, 6) Draft a message, and 7) Post said message. PPelberg (WMF) (talk) 02:20, 2 September 2021 (UTC)[reply]

• @PPelberg (WMF): I find the "new section" interface is pretty intuitive for new users. If that link could be more easily accessible, not just the talk page link, I think it would be more useful. I tend to use the &preloadtitle=Foo so that the subject line is filled with some default text, and I've found this useful for action links like leaving me messages about user scripts or questions about things from other projects. — Wug·a·po·des​ 17:53, 28 August 2021 (UTC)[reply]

  @Wugapodes: you sharing the experience you've had with, what we've been calling, the New Discussion Tool is helpful to know, especially as we approach offering as an on-by-default feature at the first set of Wikipedias in the coming weeks (see: phab:T271964).

  If that link could be more easily accessible, not just the talk page link, I think it would be more useful.

  Agreed...can you say a bit more here? What about its current location, how it appears, etc. do you think detracts from it's usefulness? Also: have you seen gadgets here, or on other wikis, that you think are effective at addressing the issue(s) that prompted you to share this feedback? For context: you sharing this is timely as well because we will soon be thinking about ways to make the affordance for starting new discussions easier for people to identify and access, regardless of where they are on a given page.

  I tend to use the &preloadtitle=Foo so that the subject line is filled with some default text, and I've found this useful for action links like leaving me messages about user scripts or questions about things from other projects.

  Interesting! Are you able to share a link to a diff and/or page where I can see what you're describing "in action"? PPelberg (WMF) (talk) 03:40, 2 September 2021 (UTC)[reply]

  @PPelberg (WMF): You can see an example at User:Wugapodes/Capricorn#Feedback and bugs. The feedback link is created using the fullurl parser function which also allows you to specify the url query. So for that link, I have it add action=edit&section=new to the url which opens the new section interface as well as preloadtitle=... which fills the subject line with the currentuser parser function so that the title is unique and I know who sent it regardless of whether they remember to sign. The full code for that link is {{fullurl:User_talk:Wugapodes|action=edit&section=new&preloadtitle=Message%20regarding%20Capricorn%20from%20%7B%7Bsubst%3Acurrentuser%7D%7D}} At Wikipedia:20th anniversary we did something very similar for the "Say happy birthday" button. I set that button up so that it opened a specific section on the talk page that we made for the purpose, and it used &summary=Wishing Wikipedia a happy birthday to prefill the edit summary for readers.

  W/r/t the link placement, full disclosure, I use responsive monobook on my mobile so take my feedback with a grain of salt. At least as I've seen, the links on articles tend to just be to the talk page which for lots of readers doesn't mean much. Getting from article to talk post is a couple extra clicks in my experience. If the link were "ask a question" or "suggest an improvement" instead of (or in addition to) a talk page link, I think it would encourage more use of talk and depending on the phrasing make the posts more useful. — Wug·a·po·des​ 07:08, 3 September 2021 (UTC)[reply]

• Thanks for that info, PPelberg. At a glance, the issues summary you put together looks good and seems to capture the main points. {{u|Sdkb}} ^talk 03:29, 2 September 2021 (UTC)[reply]

  At a glance, the issues summary you put together looks good and seems to capture the main points.

  Oh good. This is helpful to know...I appreciate you stopping back by to say as much, @Sdkb. PPelberg (WMF) (talk) 03:43, 2 September 2021 (UTC)[reply]

Requirement of 2 votes for template deletion. I understand most experienced editors avoid deletion talks like a plague ....but...Think we need a basic number of votes for template deletion. As of now we have many templates deleted by 1 vote...in many cases that 1 vote is by an editor here editing for only a year or so.--Moxy- 20:03, 30 August 2021 (UTC)[reply]

Bad idea. A ton of TfDs are rubber stamping following cleanups like migrations of rail templates to

WP:SOFTDELETE). Is there actually any evidence of a problem here, such as links to DRVs showing TfDs closed in this manner reached incorrect decisions and thus were overturned? ProcrastinatingReader (talk) 20:09, 30 August 2021 (UTC)[reply

]

Correct as per

WP:NPASR...just to bad that area of Wikipedia does not have better oversite....one day one can hope for better.Moxy- 00:58, 31 August 2021 (UTC)[reply

]

Agree with proc. A lot of TfDs don't get any votes, and like AfD and RfD, those can be deleted after a week. Of course, I think most admins would be willing to undelete upon request if the consensus is very weak. Elli (talk | contribs) 20:12, 30 August 2021 (UTC)[reply]

Proc said it well. A ton of TfDs are completely uncontroversial with many similar deletions occurring in the past with more discussion. These are often unused and not used on hundreds or thousands of pages like most of the templates that editors know about. If you have concrete TfDs you have issues with I'm sure the regulars are willing to discuss or change as appropriate. While there are a lot of problems with having a quite small pool of closers it does make changing norms a lot easier. --Trialpears (talk) 20:28, 30 August 2021 (UTC)[reply]

• Oppose. Agree with Proc's analysis of the situation. Also, strictly speaking, there are no votes in deletion discussions. Editors make recommendations, and the closing party decides based on quality of arguments as well as quantity. —C.Fred (talk) 20:32, 30 August 2021 (UTC)[reply]
• Oppose per Proc. {{u|Sdkb}} ^talk 00:48, 31 August 2021 (UTC)[reply]
• Oppose I guess per above. It the closers are admins, they're probably pretty good at figuring it out. Just treat it like a PROD I guess... somebody nominated it, nobody objected after a fair amount of time, the admin can deleted BUT use her discretion not to, and anyway yea undelete upon request in a case like that. I guess. Herostratus (talk) 00:47, 3 September 2021 (UTC)[reply]
• Oh no, all these editors editing for only a year or so, what could we possibly do about them? Oppose:
  WP:SOFTDELETE linked above can handle the job well enough in my experience. ~~~~
  User:1234qwer1234qwer4 (talk) 17:44, 8 September 2021 (UTC)[reply
  ]

It's very obvious that these two options would make the mobile version of Wikipedia more user-friendly. Interstellarity (talk) 16:31, 1 September 2021 (UTC)[reply]

• Support as nom. Interstellarity (talk) 16:31, 1 September 2021 (UTC)[reply]
• Would the proposed “contact us” button link to the article’s talk page, or are you proposing something different? Blueboar (talk) 16:47, 1 September 2021 (UTC)[reply]
• Wikipedia:Growth_Team_features#Help_panel is already being developed. If you're talking about the mobile left sidebar, that could probably benefit from some broader discussion. {{u|Sdkb}} ^talk 16:57, 1 September 2021 (UTC)[reply]
  • I've found that newbies make good use of the growth team features for essentially this purpose. I agree we should look into expanding its use if possible. — Wug·a·po·des​ 20:28, 1 September 2021 (UTC)[reply]
• ? @Interstellarity: which mobile view (please be specific)? What do you propose the target of these be? Is this something that we, the English Wikipedia editors, can technically implement? — xaosflux ^Talk 17:56, 1 September 2021 (UTC)[reply]
• Hi all,

The proposed contact us link would link to Wikipedia:Contact us and the help link would like to Help:Contents. In the mobile sidebar (the image to the right), I think both links are best placed next to the About Wikipedia and Disclaimers link. I'm not sure if it can be technically implemented by English Wikipedia editors, but I hope I provided enough information on this in detail. Interstellarity (talk) 18:19, 1 September 2021 (UTC)[reply]

There does not appear to be anywhere for us to do this locally, you could start by inquiring with the mobilefrontend developers to see if: (a) project-specific sidebar links could be supported here such as they are on the webui; (b) possibly if project specific parameters would be supported here; (c) if this is something they would want to incorporate in to the master extension. — xaosflux ^Talk 18:53, 1 September 2021 (UTC)[reply]

@Xaosflux: Could you provide the contact information for the people responsible for developing the mobile interface? That would help a lot. Thanks, Interstellarity (talk) 18:54, 6 September 2021 (UTC)[reply]

@Interstellarity: like most things, it is supported by the developer community. The workboard for this is here. @Fuzheado: is listed on that project and may have more information. — xaosflux ^Talk 19:17, 6 September 2021 (UTC)[reply]

Hmm.. not sure what you're referring to in terms of me being listed on a project? Fuzheado | Talk 00:25, 7 September 2021 (UTC)[reply]

@Fuzheado: click here you are actually listed as the only member of the "Mobile" project. I'm assuming phab project management is a sloppy mess now though! — xaosflux ^Talk 02:44, 7 September 2021 (UTC)[reply]

Also here is the MobileFrontend project listing. — xaosflux ^Talk 02:45, 7 September 2021 (UTC)[reply]

Would editors have any way of replying to such a contact? The most valuable part of this development could be a way around

WP:THEYCANTHEARYOU. Certes (talk) 19:06, 6 September 2021 (UTC)[reply

]

Hi everyone – I'm Marshall Miller, the product manager for the Growth team at WMF. Over the past three years, the Growth team has been developing a set of features meant to improve the experience for new editors. The goal of our work is to help newcomers make successful first edits, instead of them leaving from frustration or confusion. Thank you to the many community members who participated on this page and helped put together this proposal!

Following a VPR discussion in April 2021, the features have been applied to 2% of new accounts as a trial. The community members following along have agreed that the initial trial has had positive outcomes, and so we now propose increasing the share of new accounts receiving the features from 2% to 25% (with some caveats).

The Growth features aim to guide newcomers as explained on the project page. The most important elements are:

• Newcomer homepage: a special page with useful actions that newcomers are encouraged to visit immediately after account creation.
• Suggested edits: a feed of articles that have maintenance templates such as {{Advert}} and {{Underlinked}}. Newcomers can choose topics of interest to filter the feed, like "Fashion", "History", or "Physics".
• Mentorship: newcomers are assigned a mentor from a list of experienced volunteers. A pop-up form helps them submit a question to their mentor's talk page using a simple interface. Whereas Adopt-a-user is meant to be an enduring connection for a newcomer, "mentorship" in the Growth features is meant for quick questions.
• Help panel: when the newcomer is editing an article, the help panel is like a collapsible mini-homepage, providing relevant help links and the ability to ask mentor questions.

The Growth features are now on 50 Wikipedias, including large ones like French, Spanish, Russian, and Portuguese -- and with German and Dutch Wikipedias having recently started trials. Evidence shows that the features improve newcomer engagement without burdening communities.

In April 2021 we discussed testing the features on the English Wikipedia at this page. We started giving the features to 2% of new accounts in June (about 2,000 new accounts per month), and posted data and results after a month. There were two main outcomes:

• The "suggested edits" feature seems to be leading to many good edits with low revert rates. During the test period in June, the revert rate was 9% (with the baseline revert rate for newcomer edits that month being 27%). To date, 2,325 article edits have been made. Many users do dozens of these edits. You can see these diffs by filtering to the "newcomer task" edit tag in Recent Changes.
• The "mentorship" feature yields some good questions and some low-quality or nonsense questions. To date, there have been 329 questions asked to the 22 mentors who have signed up. We're still discussing how best to scale up the mentor capacity and how to encourage high-quality questions. You can see these diffs by filtering to the "mentorship module question" and "mentorship panel question" edit tags in Recent Changes.

We think the next step is to give the Growth features to 25% of new accounts for one month (about 25,000 new accounts). The exception would be the mentorship features, for which we have open questions mostly discussed here. We want to increase the share of newcomers getting mentorship to only 5%, so as not to overwhelm the mentors. We would run this phase of the test for one month, and then bring the results back here to decide whether to further increase the share of newcomers with the features. In looking at the data, we will think about these questions:

• Revert rate of suggested edits. Are high-quality edits being generated without an undue burden on patrollers?
• Volume of mentor questions. This would allow us to extrapolate how many mentors would be needed to handle 100% of new accounts, or how the mentorship features might need to be improved to handle increased volume.
• Quality of mentor questions. Are there improvements we could make to the features to help newcomers ask useful questions?

We're interested to hear any questions, ideas, or concerns -- and we're hoping there's general support for moving to this next phase of testing the Growth features. We also are looking for about 20 more people to sign up as mentors to handle the increased volume of questions that will come in with the next phase. Mentors should expect to get 3-4 questions per week on their talk page. You can learn more about that and sign up on this page. Thank you! -- MMiller (WMF) (talk) 23:33, 2 September 2021 (UTC)[reply]

I like the "Your impact" section. It can be really discouraging when you're starting out with something new to not get any feedback on whether what you're doing is making a difference. I'd be interested to learn more about what goes into picking the suggested edits, but certainly the idea of presenting a newbie with specific things they can do right now is great. And I like the tinder-esque "swipe left, swipe right" U/I in the mockup. -- RoySmith (talk) 00:00, 3 September 2021 (UTC)[reply]

Thanks for checking things out, @RoySmith! Seeing your questions made me realize I should have posted instructions for how anyone can turn on the Growth features and try them out. You can do it with these instructions.

About your comments and questions: I'm glad to hear that you think the Impact module is on the right track. The goal is to help newcomers have that realization that editing Wikipedia is valuable and exciting. It's definitely really rudimentary right now, and we're planning to make it more powerful in the coming year (we'll eventually be posting ideas on this currently barebones page).

In terms of suggested edits, we want newcomers to find interesting pages that need easy edits, so they can quickly have success on the wiki. We're currently sourcing them through maintenance templates, like {{Advert}}, which is one of the templates we draw on to find articles for the "copyediting" task. You can actually see (and edit!) exactly which templates are used for which type of task through a new Special page we created to enable community members to configure the Growth features: Special:EditGrowthConfig. With this page, much of the control over how newcomers experience the Growth features is in the hands of community admins.

Beyond the user-applied maintenance templates, we've been experimenting with creating suggested edits algorithmically. One way that's currently being tested on several wikis is called "add a link", which uses an algorithm to suggest words and phrases that could be made into wikilinks. It's going well so far, and the details are here on this project page. And in that vein, we're also building a first attempt at a task that suggests images that could be added to unillustrated articles. That one has many more open questions and risks, and so we hope to hear from more community members on its talk page.

And in terms of allowing users to choose topics of interest (e.g. "Music", "Sports", "Europe"), those come from a machine-learning model that tries to assess which topics an article is about based on the words used in it. It is trained using WikiProject tags from English Wikipedia, and tends to be pretty accurate. The information about how that model works is here.

I'm happy to answer any other questions or hear any other ideas! MMiller (WMF) (talk) 00:30, 3 September 2021 (UTC)[reply]

• Support the increase to 25%. MMiller and the rest of the Growth team have done an exemplary job collaborating with us in the community on the development of these features, and as summarized above, the results have been very promising so far. This rollout is suitably cautious and will allow more newcomers to benefit from them while giving us more information to inform the final launch. {{u|Sdkb}} ^talk 00:20, 3 September 2021 (UTC)[reply]
• Support - MMiller and his team have both an excellent project and have been doing a fantastic job at working with the community, adapting to our specific requests (one example is they added the functionality to decouple the % getting the Growth panel and the % getting mentors to adapt to our trialling wishes). I think this is a great next step in rolling it out, and will give enough trial data to better predict benefits and risks. Nosebagbear (talk) 00:53, 3 September 2021 (UTC)[reply]
• Support Seems like a reasonable next step. * Pppery * _{it has begun...} 00:56, 3 September 2021 (UTC)[reply]
• Support. Just in case it's not clear from my comments above, making this explicit. -- RoySmith (talk) 01:06, 3 September 2021 (UTC)[reply]
• This seems very worthwhile and I support moving forward as suggested. --Malcolmxl5 (talk) 01:25, 3 September 2021 (UTC)[reply]
• Support I think this is a very worthwhile project. The WMF team in charge of this has been very responsive to the community. Calliopejen1 (talk) 18:41, 3 September 2021 (UTC)[reply]
• Support per above. ― Qwerfjkltalk 19:42, 3 September 2021 (UTC)[reply]
• Support as someone who has offered to help as a mentor. I think helping new users getting the hang of how to edit on Wikipedia is one of the most important tasks we can do to make sure we have a healthy, growing community. Isabelle ^🔔 22:52, 3 September 2021 (UTC)[reply]
• Strong support, as a mentor and someone who's been giving a bit of feedback for this on the Growth features talk. As far as I've seen, it's going great so far and upping to 25% is the right next step. — Bilorv (talk) 23:04, 3 September 2021 (UTC)[reply]
• Support. Well done; this seems to be a very welcome initiative that should benefit both projects and editors. Certes (talk) 15:31, 4 September 2021 (UTC)[reply]
• Support Absolutely! Promising results so far. This seems a valuable tool for editor retention and recruitment, my number one priority. CaptainEek ^{Edits Ho Cap'n!}⚓ 22:38, 6 September 2021 (UTC)[reply]
• Support New editors need guidance. Thanks to MMiller and the team for their collaborative work. Johnuniq (talk) 23:43, 6 September 2021 (UTC)[reply]
• Support with thanks to MMiller for his diligent work with the community on this initiative. ProcrastinatingReader (talk) 00:15, 7 September 2021 (UTC)[reply]
• Support Great project that is ready to proceed forward. I still have have concerns about whether the mentorship program will scale up to 100%, but the other parts are. --Trialpears (talk) 09:12, 7 September 2021 (UTC)[reply]
• Support I turned this on for my sock account and I really like the idea, especially the "impact" summary on the homepage. I may even have finally changed my mind on the utility of cleanup tags, which I've always thought were useless clutter. Agree that the mentorship component seems least scalable and maybe should be a separate opt-in. I'd also like to see this available on mobile eventually; my first thought was that the "snack-sized task" model would be a good match for mobile editing, but I didn't see a way to enable it without switching to desktop mode. Opabinia regalis (talk) 20:51, 7 September 2021 (UTC)[reply]

  Hi @Opabinia regalis -- thanks for trying out the features! I should have said in my original post that these features are available on both mobile and desktop, and yes we do think and hope that they are a great fit for mobile editors. Maybe you're identifying some challenges with finding the features on mobile. If you tap on the "hamburger button" in the upper left on mobile, the navigation menu opens. Then you can tap your username in that menu to go the homepage. Does that work for you? Do you see a way to make it more clearly accessible?

  Speaking of mobile edits, the Growth team has been building some new types of tasks that are specifically geared toward ease of use on mobile. The first one that is currently being tested on about 10 Wikipedias (German, Arabic, French, Polish, and several others) is called "add a link", in which an algorithm suggests words or phrases that could be made into wikilinks and the user taps "yes" or "no" to add the links. It's a very simple kind of editing, but it's designed to be the sort of thing that can be done with just one hand on a cellphone, perhaps while hanging on to the railing on a bus commute. In the tests so far, it's going well. I hope you can check the project out and add any of your own thoughts here or on its talk page. MMiller (WMF) (talk) 17:06, 8 September 2021 (UTC)[reply]

  @MMiller (WMF): Thanks for looking into it! I think the issue I noticed are probably not relevant to newcomers with the features already enabled, so maybe not so important. Glad to hear about the mobile tasks!

  I should've been clearer, I noticed two things: first, I couldn't enable the features using the mobile site. Now that I look again, I think this is my misunderstanding in expecting "Settings" in mobile to correspond to Special:Preferences. I can get to preferences using a direct link, but there doesn't seem to be an interface link to it that I can find.

  Second, I hadn't realized I was using "Advanced" mobile view. In that view I don't see a link to my username in the hamburger menu. When I turned that view off, I was able to get to the new homepage as you suggest, by clicking my username. It is a little unintuitive, though, there is both a "Home" link and a "Homepage", and they're not the same thing - Home goes to the main page, so if you want your homepage you click on your username, not on "Home". I guess what I meant was, I'd like to see this in Advanced view too (which I think is generally better for editing.) Opabinia regalis (talk) 19:23, 8 September 2021 (UTC)[reply]

  I like the idea of small tasks just for mobile. Another one I could see working is spell checking/correction, but the "add a link" task seem particularly good because it's a wiki-specific skill. -- RoySmith (talk) 22:00, 8 September 2021 (UTC)[reply]

• Support Given the initial trial results, it makes sense to expand at this time. GenQuest ^"scribble" 12:31, 8 September 2021 (UTC)[reply]
• Support I've enjoyed participating in the mentorship program, and I'm glad to see the features as a whole have a positive impact. I can say from experience that the questions are of varying quality, but depending on how you can respond they can still be valuable. For example, I had one or two editors who simply said "hi". I took it as an opportunity to leave them a welcome template on their user talk page. Another posted multiple paragraphs related to their belief that they were deposed royalty. While unfortunate, I simply ignored the posts until the editor got bored and left, then I cleaned up my talk page. Those low-quality questions are outweighed by the positive interactions. Personally, I don't get too many posts, but going from 2% to 25% would probably change that. Perhaps advertising in newsletters like Tech News, Signpost, or Admin Newsletter would be useful to get more sign-ups? — Wug·a·po·des​ 21:50, 8 September 2021 (UTC)[reply]

According to the current tally, the number of articles with link rot issues slowly but steadily approaches the 300K mark. Even though that's less than 5% of the English Wikipedia and mostly we deal with articles that have most of the citations being all right, I believe that becomes quite a problem when we have articles referenced to a page which can't be found on any archive, and in particular favts referenced to that source. The info, in that case, might require update, deletion (based on

Since this is my first post at Village Pump, please assure me this is the proper venue for that proposal (or else redirect me to another forum), and, of course, I'd like to hear your opinions on it. Szmenderowiecki (talk) 14:40, 3 September 2021 (UTC)[reply]

Thanks for the idea, Szmenderowiecki! This village pump is used for proposals like this all the time, so you're at the proper venue. Resolving link rot is certainly a plus, so anyone who helps resolve it is doing good. I wouldn't say it's our most urgent backlog, though, as it's less severe than {{Citation needed}} (dead references at least likely supported the material, even if they're no longer verifiable, whereas citation needed references mark totally unsupported material) and we're never going to get through the citation needed backlog without radically increasing the number of editors.

One thing I'm wondering is where the more recently tagged instances of link rot are coming from, as Internet Archive is theoretically supposed to be archiving every outgoing link from Wikipedia. Are they old links from before IA was doing that, or editors mistaking links as permanently dead when actually they're available through IA, or is IA failing to capture everything? {{u|Sdkb}} ^talk 18:55, 3 September 2021 (UTC)[reply]

Probably all those things. Systematic saving of links at Wayback and Archive.today wasn't done prior to about 2012. Current save methods are not 100% for a couple reasons. Some percentage of {{dead link}} tags are resolvable but a bot or person has not done so yet. IMO a bigger problem is false negatives ie. links that are dead but not marked or being saved by the bots due to soft-404s and other issues. The methods we are using are problematic. The solution is every link is born archived ie. the archive URL is part of the rendered page automatically, no bots. -- GreenC 19:37, 3 September 2021 (UTC)[reply]

It seems to be the former case (as a lot of these are from 2007-2012, which is why it is often hard to find them; some are apparently cited in 2013 and dead (for example here), while this article describes a 2015 Spanish film, and yet the official website is dead and the link rot tag is there.

as Internet Archive is theoretically supposed to be archiving every outgoing link from Wikipedia but it isn't doing so automatically. At least the references in this article stay unarchived for two months or so (though to be honest, IA bot has gone once or twice through the page when the heat wave was in the news).

IMO a bigger problem is false negatives ie. links that are dead but not marked or being saved by the bots due to soft-404s and other issues. I haven't really encountered that problem at the time I was doing that on a sample of some 20-30 articles; and anyway we can't recognise it until we actually get to the page and check both the URL and archive, which can only be done by a person.

As for "citation needed" backlog, this one is going to be way more difficult than dead links, because the latter will just involve checking the page against its archived versions (or copying the title), while in the former, we would actually have to find the information, which often is difficult and involves several languages. I mean, the anti-link rot drive is a quicker fix (though admittedly not so fundamental).

The solution is every link is born archived ie. the archive URL is part of the rendered page automatically, no bots. You mean that we should automatically send queries of whatever links we submit to Wikipedia to IA, and then added to the citations? Sounds pretty cool, but that's hell of a load on IA's servers. Szmenderowiecki (talk) 01:30, 4 September 2021 (UTC)[reply]

"Sending every link.. we submit to Wikipedia to IA" is what Internet Archive does see Wikipedia:Link_rot#Automatic_archiving. To give a sense of scale, Wikipedia has about 100-200 million unique URLs in about 500 sites; Wayback machine has archived over 107 billion web pages. The false negative problem is not anecdotal, it can be extrapolated based on a decent size data set. It's not a wild guess to suggest there are many more than 300k articles containing silent dead links. For an example of links born archived, see frwiki - not recommending that solution, but is an example of how things can be done differently. -- GreenC 18:58, 4 September 2021 (UTC)[reply]

That's all the more reasons to initiate such drive in this case. Szmenderowiecki (talk) 06:47, 5 September 2021 (UTC)[reply]

There are also some problems with internal link rot that are worth considering as well. We have redirects (some categorized by {{r to section}} and {{r to anchor}}) which point to particular sections of articles that no longer exist or changed their name. These take you to the article, but sometimes the proper section or information can be hard to find. — Wug·a·po·des​ 20:59, 7 September 2021 (UTC)[reply]

So... I been seeing a lot of massively-viewed articles without a protection, due to it being expired. Should we start discussing the idea of an auto-protect bot for articles that get over 250K views a day, and depending on size like is it a B class or GA class, should it be protected. It will help most of the CVU/AVU jobs. MoonlightVector 16:21, 7 September 2021 (UTC)[reply]

You are seeing unprotected pages with lots of views because

requests for page protection where a human will review the best tools to stop the disruption. — Wug·a·po·des​ 20:04, 7 September 2021 (UTC)[reply

]

This has very little chance of passing, even with a very conservative threshold, but I wish we wouldn't dismiss it out of hand so quickly. We recently had a very acute lesson in what can happen when we wait to see if vandalism occurs rather than acting preemptively. There's not a huge difference in pageviews between a template that appears on 500 smallish articles and a single extremely high-visibility article. I would be interested to see a (hopefully non-public; feel free to email me) list of the unprotected pages with the most pageviews. There's good reasoning behind our founding-era principles, but at this point they're 20 years old and they're not gospel, so we shouldn't be afraid to re-examine them as needed. {{u|Sdkb}} ^talk 21:10, 7 September 2021 (UTC)[reply]

@Sdkb, while there might not be a huge difference in pageviews between a template that appears on 500 smallish articles and a single extremely high-visibility article, it is much more common for templates to use more complicated wikisyntax than for articles, so the protection likely prevents widespread damage from unsuccessful good-faith "fixes" by unexperienced editors and not just vandalism. Sure, this can (and does) of course as well happen to articles, but again, these usually have more simple syntax (and afaik the VisualEditor is now the standard option?) ~~~~
User:1234qwer1234qwer4 (talk) 15:43, 8 September 2021 (UTC)[reply]

Indeed, the reason for protecting templates isn't page views, it's the technical aspects of how templates work. The harm of good-faith-but-incorrect edits is a major reason, but the damage caused by bad faith edits to templates can persist even after a revert because of how pages are parsed and served to readers. Rather than processing each page every time a reader requests it from the server, we cache them. An edit always invalidates the cache, forcing a re-render of the content. Most vandalism is reverted quickly if it even gets past filters, and since the edit purges the cache the vandalism immediately disappears. This is not true for templates. Because updates to templates can invalidate millions of caches, any update to a template is slowly rolled out across pages that use it. So no matter how quickly vandalism to a template gets reverted, it can stay up on articles for potentially hours after the revert, until the cache gets updated. These are vastly different levels of damage, and the idea that preemptive template protection is the same as preemptive article protection only makes sense if you don't dig into the specifics of how these attacks work. Vandalism to a template used on many small articles can easily have much greater consequences than vandalism to a single highly visible article.

@

WP:HRT yesterday. If we are going to amend a foundational principle of the website and major clause of the protection policy, per the parable of Chesterton's fence I'd want a proposal that shows some understanding of the protection policy, or at least mentions it, before I support spending substantial volunteer time entertaining a perennial proposal. — Wug·a·po·des​ 21:27, 8 September 2021 (UTC)[reply

]

@

WP:PERENNIAL list yet, which might be a good idea. When recently-discussed topics are raised, it's good to have easy access to a concise history with wikilinks. {{u|Sdkb}} ^talk 21:35, 8 September 2021 (UTC)[reply

]

@

pending changes protections, which, while allowing everyone to edit an article, would make edits subject to a review before being displayed. I don't have any strong opinion on the issue, but I wanted to point out that the "protection" mentioned in the opening comment does not necessarily refer to a protection from editing. ~~~~
User:1234qwer1234qwer4 (talk) 15:55, 8 September 2021 (UTC)[reply

]

RFC: New PDF icon

Please consider joining the feedback request service. An editor has requested comments from other editors for this discussion. This page has been added to the following lists: Wikipedia technical issues and templates Wikipedia proposals When discussion has ended, remove this tag and it will be removed from the lists. If this page is on additional lists, they will be noted below.

Should we replace the current PDF icon? –MJL ‐Talk‐^☖ 05:33, 8 September 2021 (UTC)[reply]

Background

Our current PDF icon is

Berrely mentioned this in WP:Discord, so I set about coming up with a modern SVG version of the file. The result was File:Icons-mini-file pdf old.svg  File:Icons-mini-file pdf.svg 

• Option 1 to support using File:Icons-mini-file pdf old.svg  File:Icons-mini-file pdf.svg 
• Option 2 where you agree that we should change File:Icons-mini-file acrobat.gif  but don't like File:Icons-mini-file pdf.svg 
• Option 3 to oppose any changes to File:Icons-mini-file acrobat.gif 
• Option 4 for removing the icon entirely.

Consensus for Option 2 should be followed up in a separate discussion. [Updated 15:35, 9 September 2021 (UTC)]

Discussion (new PDF icon)

• Option 1. As proposer. –MJL ‐Talk‐^☖ 05:33, 8 September 2021 (UTC)[reply]

  Comment. I have uploaded a new file that does not have the GPL copyright concerns attached to it. Please see File:Icons-mini-file pdf.svg  for the old file. Currently, the old one is at File:Icons-mini-file pdf old.svg  and the new one is at File:Icons-mini-file pdf.svg , but they should be swapped in like 30 minutes or so. –MJL ‐Talk‐^☖ 21:38, 8 September 2021 (UTC)[reply]

  @Xaosflux, Anomie, WOSlinker, and Wugapodes: Sorry to ping you both again about this I have now replaced the SVG with a CC0 one per your concerns. –MJL ‐Talk‐^☖ 05:00, 9 September 2021 (UTC)[reply]

  The link in the summary above doesn't seem to point to that? Perhaps strike and insert to the proposal above? — xaosflux ^Talk 10:46, 9 September 2021 (UTC)[reply]

  Fixed. –MJL ‐Talk‐^☖ 15:35, 9 September 2021 (UTC)[reply]

• Option 1. Clean look and close enough to the original. I would also be open to moving away from the Adobe Acrobat logo if someone comes up with a different icon, since the company no longer holds a monopoly over the format. Yeeno (talk) 🍁 06:01, 8 September 2021 (UTC)[reply]
• Option 2. I would prefer something that isn't tied to Adobe, like File:Icon pdf file.svg: . There are many more options in commons:Category:PDF icons. – Joe (talk) 06:15, 8 September 2021 (UTC)[reply]

  Since this option is proving popular, but some have correctly pointed out that the "PDF" text is hard to read, I've created a version which is better optimised for small sizes: File:PDF icon.svg. Please feel free to tweak it further. – Joe (talk) 12:37, 8 September 2021 (UTC)[reply]

  I tweaked it further, but I think there's a limit to how well tiny SVGs can render text: . --Ahecht (TALK
  PAGE) 20:13, 8 September 2021 (UTC)[reply]

• Option 2, and concur with Joe that the Adobe logo is mega fail. The icon he posted in the comment above this one seems good, and it's an SVG, which is better than the tiny GIF being used currently as it can be rendered at any size without looking terrible. jp×g 06:34, 8 September 2021 (UTC)[reply]

  Man, there's some pretty great icons in that category. jp×g 06:37, 8 September 2021 (UTC)[reply]

  Do I look like I know what a JPEG is? ~~~~
  User:1234qwer1234qwer4 (talk) 09:41, 8 September 2021 (UTC)[reply]

• Option 2 The icon must change. The old thing is a relic of the dark ages. Initially I thought ooh the adobe svg looks great. But Adobe are no longer the pdf overlords, and I don't rather like Adobe, evil empire of tech that it is. Joe's suggestion of the generic PDF SVG is the perfect solution. CaptainEek ^{Edits Ho Cap'n!}⚓ 06:42, 8 September 2021 (UTC)[reply]
• Option 3 as a reasonable specific replacement. Robert McClenon (talk) 07:03, 8 September 2021 (UTC)[reply]

   Eeeehhhhhhh, @Robert McClenon, are you sure you typed in the right number for your preferred option? ~~~~
  User:1234qwer1234qwer4 (talk) 09:45, 8 September 2021 (UTC)[reply]

• Option 1 as a reasonable specific replacement. Robert McClenon (talk) 14:58, 8 September 2021 (UTC)[reply]
• Comment is licensed as copyright free but is licensed as GPL which could make a diffence as to how it can be used without any linking back to the file. -- WOSlinker (talk) 07:36, 8 September 2021 (UTC)[reply]
• Option 2 I agree with Joe.--
  here! 07:55, 8 September 2021 (UTC)[reply
  ]
• Option 2; the new SVG looks wrong to me without a gradient, and I think we should move away from Adobe promotion. ~~~~
  User:1234qwer1234qwer4 (talk) 09:39, 8 September 2021 (UTC)[reply]

  Option 4 below does not seem too far fetched either. We don't seem to have this for other file formats, why display an image specifically for PDFs? ~~~~
  User:1234qwer1234qwer4 (talk) 20:42, 8 September 2021 (UTC)[reply]

• Option 2: PDF is no longer specific to Adobe, so let's remove their logo. Option 1 (File:Icon pdf file.svg) looks ideal when enlarged, but the letters are hard to distinguish at icon size. I suggest a new copyright-free SVG with even larger letters (They won't occupy many pixels.) Certes (talk) 11:40, 8 September 2021 (UTC)[reply]

  @Certes, note that there are no "letters" in option 1. ~~~~
  User:1234qwer1234qwer4 (talk) 15:15, 8 September 2021 (UTC)[reply]

  Oops, I was referring to Icon pdf file.svg. Certes (talk) 17:08, 8 September 2021 (UTC)[reply]

• 1 or 2 is fine for me. --Izno (talk) 12:01, 8 September 2021 (UTC)[reply]
• Option 2: Although "oppose any changes" seems pretty strong, I was leaning toward Option 3 since the proposal seemed to be based on the argument It's a .gif made over 16 years ago, with no explanation of why that's bad. Personally, I'd rather use a 291-byte file than one 6 KB in size, ceteris paribus. But then, despite the weird threading, I noticed the replacement suggested by Joe Roe. It doesn't have the Adobe logo or (*gag*) name on it, it's clearly for PDFs, and it's only 2 KB in size. So if we're going to change, let's change to something like that. This one (, 707 bytes) works for me, too. — JohnFromPinckney (talk / edits) 12:04, 8 September 2021 (UTC)[reply]

  I don't think the file size matters here, because what is actually downloaded by your browser is a server-rendered bitmap of the appropriate size, not the original SVG. That's approximately the same for all the files,[1][2] and less than 1 KB. Also, "weird threading"? – Joe (talk) 12:18, 8 September 2021 (UTC)[reply]

  Did not know about the different download file, thanks. And it wasn't actually so much weird threading as it was confusion on my part from JPxG's contributions. My too-quick reading saw the big file image at the upper right, which wasn't either the current icon nor the proposed replacement. When they wrote concur with Joe that the Adobe logo is mega fail I thought they were agreeing with the proposer (which you're not) that the icon was mega fail (although that wasn't clear why). Then JPxG also said there's some pretty great icons in that category [sic], which would have been more appropriate, IMO, as a reply to your 06:15 post, not as a reply to themself. I got a bit lost. Confusion on my part from scanning too quickly, and thinking too slowly. — JohnFromPinckney (talk / edits) 14:52, 8 September 2021 (UTC)[reply]

  I suppose the problem isn't directly that "[i]t's a .gif" but rather that it has a fixed resolution looking pixelated on modern screens, while a vector version would be rendered in an appropriate resolution on any device, as Joe pointed out. ~~~~
  User:1234qwer1234qwer4 (talk) 15:22, 8 September 2021 (UTC)[reply]

  Striking my !vote until I have time to study the revised options. — JohnFromPinckney (talk / edits) 18:09, 9 September 2021 (UTC)[reply]

• Option 1. That logo is still the most widely associated with the PDF format, and anything else is just making things less clear for our readers. There doesn't seem to be a clear alternative being posited (the letters in the version proposed by Joe are illegible at this resolution), and most of the reasons above seem to hinge on people's person feelings about Adobe, which shouldn't enter into this debate. Either leave well alone, or adopt the new clearer version proposed by the OP.  — Amakuru (talk) 12:14, 8 September 2021 (UTC)[reply]

  "Personal feelings" is a bit dismissive. We're an free knowledge project and have a long history of supporting free software, free licenses, and free formats wherever possible. I highly doubt that the generic 90s software-looking acrobat squiggle is more widely recognised than the letters "PDF", but I agree the legibility of my first suggestion could be improved. – Joe (talk) 12:22, 8 September 2021 (UTC)[reply]

  If someone comes up with an alternative that actually works, then I might support it. But I'm not going to give a blank cheque to swap an easily recognisable logo for one which might not immediately convey its meaning to our readers. "Option 2" involves dispensing with the current logo without any consensus as to what we're swapping it for, and I can't support that.  — Amakuru (talk) 12:31, 8 September 2021 (UTC)[reply]

• Option 2 - some generic PDF logo (i.e. not the Adobe 'squiggly triangle') to be determined later. SVG > GIF, of course, but I think we should take this opportunity to swap it for a more generic logo. firefly ( t · c ) 12:25, 8 September 2021 (UTC)[reply]
• Option 2 or 3 Option 1 is a non-starter due to license. We need something that's public domain or CC0 to avoid a requirement to link back to the file description page for attribution and/or notice of license. I wouldn't oppose an identical image with a proper license; while it's annoying to have the Adobe software's logo in there, it's also recognizable. As for that several people above seem to like, all I see at that size is a document icon with a red bar over it. The text "PDF" is not visible. Again, I wouldn't oppose an alternative that's more legible. Anomie⚔ 12:26, 8 September 2021 (UTC)[reply]
  • Striking as the concerns I raised seem to have been addressed, the new image for option 1 has a usable license and people have suggested as a better choice for option 2. I don't have enough of an opinion on the current options to !vote. Anomie⚔ 16:57, 9 September 2021 (UTC)[reply]
• Option 2 Let’s move away from Adobe. --Malcolmxl5 (talk) 12:32, 8 September 2021 (UTC)[reply]
• Option 3 This is change for the sake of change and doesn't actually accomplish anything. * Pppery * _{it has begun...} 13:21, 8 September 2021 (UTC)[reply]

  @Pppery, the current file is 512 pixels, which is too small to be rendered properly on modern screens and appears conspicuously pixelated. ~~~~
  User:1234qwer1234qwer4 (talk) 15:29, 8 September 2021 (UTC)[reply]

• Option 3 or 2: the current icon is serviceable as is, but if we were to change, I'd rather something without the Adobe logo. Isabelle ^🔔 14:35, 8 September 2021 (UTC)[reply]
• Option 2 per Firefly and others. There is nothing wrong with using a 16-year-old icon per se, and the proposed replacement's only advantage is in file format and that's not enough reason imo to justify a change. However what does justify a change is using a generic icon that doesn't require someone knowing what the logo of a private company represents. Whatever icon we end up choosing, we should probably consider including it as an example in the PDF article. Thryduulf (talk) 15:07, 8 September 2021 (UTC)[reply]
• Option 2, PDF files are no longer the sole domain of Adobe and we shouldn't be using their logo, but none of the suggested icons have been readable. I modified one of the existing PDF SVG icons on commons to make it more readable (), but if the intent is to use this at 16px, pixel art is always going to render better than an SVG, e.g. . --Ahecht (TALK
  PAGE) 15:37, 8 September 2021 (UTC)[reply]

  I have to admit that the 16px PNG rendering does look like a usable option and also looks way less pixelated than the current icon (where even the border displays very blurry), at least on my end. ~~~~
  User:1234qwer1234qwer4 (talk) 15:48, 8 September 2021 (UTC)[reply]

• Ehhhhhhh ---- GPL are we opening a can of worms by changing from a free image to one that has to drag GPL around with it everywhere? — xaosflux ^Talk 16:58, 8 September 2021 (UTC)[reply]

  Leaning more towards Option 2 and using File:Icon pdf file.png or something similar, provided it is CC0 or other very-free license. — xaosflux ^Talk 18:50, 8 September 2021 (UTC)[reply]

  @Xaosflux and Anomie: Wouldn't a comment in the CSS be sufficient for linking to the license and authorship? –MJL ‐Talk‐^☖ 20:56, 8 September 2021 (UTC)[reply]

  Nope. The GPL seems particularly weird when it comes to images, and even more weird when it comes to SVG images. The bottom line is that we need to clearly distribute the image along with the author's copyright notice and the notice that it's under the GPL, which we satisfy by linking the image to the file description page that has all that information. Hiding a comment in a CSS file, where it'll be hard to find and may be minified out, won't cut it. Anomie⚔ 21:55, 8 September 2021 (UTC)[reply]

  Okay, I have managed to remake the SVG using stuff that was in the Public Domain. –MJL ‐Talk‐^☖ 05:06, 9 September 2021 (UTC)[reply]

• Option 2 I find the PDF text version quite promising. The one I think is the most legible is (show) (File:Icon_pdf_file.png) which is easily readable on both mobile (both vector and mobile version). There are of course scenarios where it wouldn't be legible, but I feel the current icon would be non-distinctive under the same circumstances and I could see many readers not knowing what the acrobat icon means now a days. --Trialpears (talk) 17:03, 8 September 2021 (UTC)[reply]
• Option 2 Go for File:Icon_pdf_file.png . It's more readable than the similar svg versions. -- WOSlinker (talk) 18:23, 8 September 2021 (UTC)[reply]
• Option generic PDF SVG
  b} 18:34, 8 September 2021 (UTC)[reply
  ]
• Option 4 get rid of it altogether. No GPL license, no trademark (US #3652386 and #3652388, I have no idea how to link these from https://tmsearch.uspto.gov/ so search yourself, it's the squiggly triangle), no BS. The document icon with "PDF", even Joe Roe's improved version, is hard to read and ultimately provides no additional information over just the text "(PDF)". This is currently defined in MediaWiki:Common.css#L-510 btw, it's not template specific or MediaWiki default. — Alexis Jazz (talk or ping me) 19:55, 8 September 2021 (UTC)[reply]
• Opposed to any GPL-licensed image or image restricted by trademark. Would prefer CC-0 license. No strong opinions on the design itself, I'm open to a new one but don't see a serious problem with the existing image. — Wug·a·po·des​ 20:29, 8 September 2021 (UTC)[reply]

  @Wugapodes: Wouldn't the trademark issue be a problem with File:Icons-mini-file acrobat.gif  as well? I'm a little confused there. –MJL ‐Talk‐^☖ 21:00, 8 September 2021 (UTC)[reply]

  I'm not an expert on trademark, but I presume so. My understanding is that having a trademark isn't a problem per se as long as we aren't using it to mislead readers about brand identity or disparage the trademark holder. The issue isn't a legal one, but a philosophical one: I'd prefer we use free equivalents that do not have copyright or trademark restrictions whenever possible. But unless we have consensus for an option that is free of copyright and trademark, I'd rather we have some graphical representation of the PDF over nothing. So by no "serious problem" I mean that it's not enough for me to say get rid of it immediately, but I do think there is room to improve. If we are going to improve, I want us to also move in the direction of copyright- and trademark-free images, but if the option is do nothing or remove the icon without replacement, I'd rather do nothing. (NB I do like Ahect's idea of just using stylized text instead of an icon.) — Wug·a·po·des​ 21:40, 8 September 2021 (UTC)[reply]

   Question: Why is specifying the file format necessary for PDF files? ~~~~
  User:1234qwer1234qwer4 (talk) 22:13, 8 September 2021 (UTC)[reply]

  Adobe Acrobat Reader. — Alexis Jazz (talk or ping me) 22:33, 8 September 2021 (UTC)[reply

  ]

  @1234qwer1234qwer4: PDFs are a bit of a weird file format. Sometimes when you click a link it will automatically download a file on your device, but other times it can just open up in a new tab. The biggest concern, however, is that not all mobile devices support PDFs across the board. My phone can just barely do it (and requires a download everytime I view one.. which can be a problem for larger files), so I have always found these icons helpful. –MJL ‐Talk‐^☖ 05:06, 9 September 2021 (UTC)[reply]

• Option Alexis Jazz -
  WP:ACCIM recommends that information should not be conveyed using only images, and while the revised icon with plain text is a slight accessibility improvement over the corporate logo version, it's still a long way off from meeting that standard. Alt text would help, but a simple (PDF) alongside the link is frankly much better and more useful than a small icon with tiny, barely-legible text. Ivanvector's squirrel (^trees/_nuts) 20:40, 8 September 2021 (UTC)[reply
  ]

  Or something like PDF, which is reminiscent of what Google uses. --Ahecht (TALK
  PAGE) 21:14, 8 September 2021 (UTC)[reply]

• With apologies to those who already knew, the choice of icon is implemented in
  .png may be the best format for this use case. Certes (talk) 22:52, 8 September 2021 (UTC)[reply
  ]

  Certes, I had already mentioned Common.css above, but you inspired me to add an anchor to the line number. — Alexis Jazz (talk or ping me) 23:02, 8 September 2021 (UTC)[reply]

  @Certes: Really, SVG is the best format because it is the most scalable (imo). –MJL ‐Talk‐^☖ 05:06, 9 September 2021 (UTC)[reply]

  Normally, yes, but the software doesn't seem to accept SVGs here. Also, if it's shown at a fixed 16px, we should probably optimise for that, e.g. align the paper edges and the orthogonal lines of the letters mid-pixel. If the day comes when MediaWiki renders an SVG at 128px on our 16k holodisplays, we can replace the icon again. Certes (talk) 10:19, 9 September 2021 (UTC)[reply]

  When it says "cannot be SVG format", I suspect that refers to the URL used. So https://upload.wikimedia.org/wikipedia/commons/c/cb/Icons-mini-file_pdf.svg would fail, but https://upload.wikimedia.org/wikipedia/commons/thumb/c/cb/Icons-mini-file_pdf.svg/16px-Icons-mini-file_pdf.svg.png would be ok. For that matter, the restriction on SVG may be outdated (it was written in 2011), or may have been because someone's browser in 2011 didn't support SVGs there, or may be to avoid explaining that the SVG's intrinsic size needs to be 16px; at any rate, I tested it and it seemed to work fine. But I do agree that optimizing the icon for the size would be a good idea. Anomie⚔ 11:51, 9 September 2021 (UTC)[reply]

• Option 2 I think File:Icon_pdf_file.png is a much better replacement, is very readable, and is simple. This is the obvious choice for me. DiamondIIIXX (talk) 03:47, 9 September 2021 (UTC)[reply]