Wikipedia:Village pump (technical): Difference between revisions

Page for discussing Wikipedia technical issues

Right now, Template:Location map is used in many infoboxen to show the location of a city or building on a map of the surrounding area. It’s a great feature, except that when you click on the map, it takes you to the file page for the underlying image without the pushpin. I think this is quite unintuitive, and a big problem for accessibility since by default the map is quite small (at least on desktop). It’s possible to zoom in by simply enlarging the web content in-browser, but not everyone knows how to do this.

The most intuitive solution would be for clicking on it to enlarge the image with the pushpin still included, but I’m not sure how feasible this is (I don’t really know how modules work but it seems like you would have to implement a new GUI element). What seems probably easier, and arguably more useful, would be for clicking on it to open an interactive map, like you get when you cilck on a Template:Infobox mapframe. That would reuse an existing interface and allow the user to zoom and pan as much as they like on a full-sized map (at least in browser).

Does anyone here know anything about how to do that? And do people think that would be a good idea? Justin Kunimune (talk) 20:53, 30 December 2025 (UTC)[reply]

@Justinkunimune: Pushpin maps like that are not a single image, they are two. You don’t give specific examples, but in the case of Didcot it’s File:Oxfordshire UK location map.svg upon which is superimposed File:Red pog.svg. By using this method, we only need two images for every location in Oxfordshire – if we did it as a single image, we’d need several hundred, each taking up about 2 MB of storage. —Redrose64 🦌 (talk) 22:29, 30 December 2025 (UTC)[reply]

@Justinkunimune: totally agree on the unexpected empty file page map behaviour.

Some (all?) infoboxes will produce an interactive map as you describe by default if you leave out the pushpin parameters. Then you get a clickable map with marker on the location of the article’s subject. Santa Barbara Church example (took me two edits). However, in that example the pushpin has two zooms with a toggle. I couldn’t get the same behaviour using mapframe-switcher = auto on {{infobox church}} – it shows four maps unhdiden (none of which include the good zoom of the default) without a toggle – the feature is broken/unavailable? Commander Keane (talk) 07:18, 31 December 2025 (UTC)[reply]

It might be that the toggle only appears when the edit is finalized; I’ve noticed that for location maps it doesn’t show the toggle in the editor preview for some reason. I tend to like the static location maps because the zoom is set manually by a human instead of automatically, and it lets the editors choose between a relief map and a political map, so I think it would be ideal to combine the static location maps with the existing interactive mapframe functionality. Justin Kunimune (talk) 01:49, 1 January 2026 (UTC)[reply]

I made a click to enlarge feature when working on {{CineMol}}, potentially that approach could be copied for location map. Bawolff (talk) 19:52, 31 December 2025 (UTC)[reply]

Oh, that looks perfect! How difficult was that? Would it be simple to adapt it to {{Location map}}? Justin Kunimune (talk) 01:50, 1 January 2026 (UTC)[reply]

@Justinkunimune I made a proof of concept at User:Bawolff/location_map_lightbox. Its not perfect but demonstrates the idea. Bawolff (talk) 09:03, 6 January 2026 (UTC)[reply]

Oh sweet, thank you for putting that together! Definitely needs polishing, but that’s more or less just what I was thinking. I was able to find another approach with Template:Maplink that might work, so I’ll try messing with both to see what seems like it would work best. Justin Kunimune (talk) 16:12, 6 January 2026 (UTC)[reply]

Temporary account sticky header overlaps hamburger menu when scrolled all the way at the top. Occurs on Chrome on IOS, may occur on other mobile devices.

And the archive template for AN, ANI, and such noticeboards don’t conform properly to the mobile layout and cause the template to go off the screen. ~2025-31733-18 (talk) 10:56, 3 January 2026 (UTC)[reply]

Hello? ~2025-31733-18 (talk) 17:29, 4 January 2026 (UTC)[reply]

The archive template is a known issue and the fix takes some time (I have started before). You can leave a note on the template talk page for that one so we don’t forget if you want.

The other is an issue for upstream, though it’s not clear if your issue is with Vector 2022 or Minerva (the latter of which is the default skin for mobile). Izno (talk) 21:23, 4 January 2026 (UTC)[reply]

I think it’s Minerva or whatever the default mobile skin is. Not sure whether the issue is present on any other skin/device. ~2025-31733-18 (talk) 13:58, 6 January 2026 (UTC)[reply]

I filed a bug, I couldn’t see it already lodged. Tracking template added above.—Commander Keane (talk) 14:36, 6 January 2026 (UTC)[reply]

(I’ve copied the following message from a similar one of mine at the Teahouse)

Hey! I’m trying to add a module to an infobox template of mine – that is, I’d like my template to consist of a part which I made myself and one below it that already exists as a separate template. I am very new to creating templates. Does anyone know how to insert this module? I have absolutely no coding experience, so could this be done without Lua? I can show an example of my progress so far, so maybe someone can point out where I went wrong, as the module currently doesnt work:

{{Infobox
| child = {{Yesno|{{{embed|no}}}}}
| title      = {{{Title}}}
| label1     = Any label
|  data1     = {{{Any data}}}
| label2     = Another label
|  data2     = {{{Another data}}}
|  header3   = Separate module should go below:
| module4 = {{Infobox any example
| child = yes
| image = {{{Any image}}}
| caption = {{{Any caption}}}
}}
}}

Of course, this is just an example, but I’m wondering if my format is wrong. Any help is greatly appreciated! Rockfighterz M (talk) 23:54, 3 January 2026 (UTC)[reply]

When reporting a problem, always link to an example page. – Jonesey95 (talk) 00:02, 4 January 2026 (UTC)[reply]

Sure: Template:Infobox esports duo Rockfighterz M (talk) 01:15, 4 January 2026 (UTC)[reply]

Perfect. I have cleaned up that template per the infobox guidelines and fixed a few things. You can see a basic test case at Template:Infobox esports duo/testcases. – Jonesey95 (talk) 02:50, 4 January 2026 (UTC)[reply]

Thank you! Is there any way for the two infoboxes at the bottom to cover the entire width of the original infobox? Now there’s an awkward margin Rockfighterz M (talk) 12:11, 4 January 2026 (UTC)[reply]

Yep. The template documentation helped me fix it. Keep on plugging away. It looks like your immediate questions have been answered. Feel free to ask more specific questions at Template talk:Infobox esports duo. – Jonesey95 (talk) 18:53, 4 January 2026 (UTC)[reply]

Appreciate the help @Jonesey95! Rockfighterz M (talk) 22:49, 4 January 2026 (UTC)[reply]

When an article is moved to draft space, a redirect is automatically created from main space to draft space, which is against rules/guidelines. Editors who move the article are then supposed to tag the redirect for speedy deletion.

If we do not want redirects from main space to draft space, then why is the redirect created automatically? It’s just adding an unnecessary extra step, both for the editor and the administrators.

I know only administrators can delete articles, but can’t the whole thing be avoided by not creating an automatic redirect? TurboSuperA+_[talk] 12:45, 4 January 2026 (UTC)[reply]

Admins, page movers, bots, and ‘crats can suppress the redirect creation when appropriate. Opening this up to anyone, even just for article to draft moves, would make it too easy for vandals. Anomie⚔ 14:12, 4 January 2026 (UTC)[reply]

… and by default we do want moves to generate redirects. Izno (talk) 21:25, 4 January 2026 (UTC)[reply]

@TurboSuperA+: The mover does not have to tag the redirect for deletion. We have an automatic procedure. MediaWiki automatically adds MediaWiki:Move-redirect-text to the redirect code. We made it include {{R from move}} which automatically detects a redirect from mainspace to certain other namespaces and adds a note and Category:Redirects from moves from mainspace to R2 namespaces. Here is an example which wasn’t deleted so it can still be seen by non-administrators. Enable “Show hidden categories” at Special:Preferences#mw-prefsection-rendering to see the category. An administrator can then decide whether to delete the redirect, change the target to a relevant article, revert the move, or something else. PrimeHunter (talk) 22:44, 4 January 2026 (UTC)[reply]

I made a public watchlist with all possible listings in a year in September using some magic words to update itself, however it seems to have broken after the new year. The links have updated on the page itself, but on the recentchangeslinked special page none of the 2026 listings are shown. Any idea on how to fix this? Tenshi! (Talk page) 17:27, 4 January 2026 (UTC)[reply]

Special:RecentChangesLinked depends on the data recorded in the pagelinks table, which still had the links from a 2025 parse. A null edit should have fixed it for you now. Anomie⚔ 18:00, 4 January 2026 (UTC)[reply]

That seems to have fixed it, thanks. Tenshi! (Talk page) 21:53, 4 January 2026 (UTC)[reply]

I’m able to login with my username and password, but I’m not receiving the 2fa email. Can someone tell me how I would be able have the email associated with my account updated? ~2026-40921 (talk) 19:00, 4 January 2026 (UTC)[reply]

IIRC, there should be a link on the page asking for the email code to a request form. Anomie⚔ 19:32, 4 January 2026 (UTC)[reply]

I’m not sure if the your 2FA involves setting up timed one-time passwords like in Authy or Aegis Authenticator or the email confirmation.

Because if it’s setting up one-time passwords that are sent to your phone app then it does not require an email (I just reset and reenabled my 2FA credentials to check; see generally WP:2FA)

If you didn’t get an email that was supposed to have a code you need to enter, like in this case, then I think you can request the WMF to resend it to you. Or maybe you messed up and made a typo in your email address, which is why you won’t get the code. If so, it’s maybe better to make a new account and later disclose the failed attempt to set up an old account. They may deny you sending the code to a different email address you provide for security reasons because they don’t exactly know if the author of that request is the same person who initially set up the account. Szmenderowiecki (talk · contribs) 19:52, 4 January 2026 (UTC)[reply]

email authentication is not the normal two-factor authentication. If you can’t log on and can’t get your email authentication your only option is to contact cawikimedia.org. — xaosflux ^Talk 00:55, 5 January 2026 (UTC)[reply]

Or, more specifically, fill out the form at Special:AccountRecovery. * Pppery * _{it has begun…} 02:07, 6 January 2026 (UTC)[reply]

Hope this is resolved soon. — Maile (talk) 22:23, 4 January 2026 (UTC)[reply]

@Maile66 Are you still seeing lag? AfDstats doesn’t do any caching, so the only thing that would cause it not to update would be database lag on the toolforge servers, of which there is is currently none. —Ahecht (TALK
PAGE) 15:49, 5 January 2026 (UTC)[reply]

Everything is working today. Thanks — Maile (talk) 17:38, 5 January 2026 (UTC)[reply]

Anyone know what would have caused that? FWIW, I bypassed and then tried clearing the cache to no avail. Browser is Mozilla Firefox 146.0.1 (64-bit). —Locke Cole • t • c • b 00:52, 5 January 2026 (UTC)[reply]

It points to this explanation. — xaosflux ^Talk 00:54, 5 January 2026 (UTC)[reply]

It worked at the top so I made tests to see where it broke and made a fix.[1] {{cot}} opens both a div and a table. {{cob}} closes both. PrimeHunter (talk) 01:20, 5 January 2026 (UTC)[reply]

Thanks for sorting that, I couldn’t find the offending bit. =) —Locke Cole • t • c • b 17:15, 5 January 2026 (UTC)[reply]

There’s an error in this section where the excerpt to the 2026 United States strikes in Venezuela article begins with a broken image(?) link, which is embedded as “[ [File:1 |thumb|]]” (no spaces). Another editor also saw this error, but was unable to fix it, instead instructing me to go here. Does anyone know how to fix this without removing the excerpt link? SirDore (talk) 01:23, 5 January 2026 (UTC)[reply]

@SirDore: Fixed with |files=0 [2] per Template:Excerpt#Details. The infobox in 2026 United States strikes in Venezuela uses {{multiple image}} in the image parameter and this apparently produces code {{Excerpt}} cannot handle when it tries to extract the infobox image. PrimeHunter (talk) 01:51, 5 January 2026 (UTC)[reply]

Thank you! SirDore (talk) 01:52, 5 January 2026 (UTC)[reply]

Thank you for the quick solution. Module talk:Excerpt is another good place to get help with similar problems. Certes (talk) 10:55, 5 January 2026 (UTC)[reply]

I have my username, password, and am currently logged in, but I have an old email address I can no longer access listed in my wikipedia account. Is there some way to change my email address without using a code sent to the (old) email address? Newystats (talk) 09:57, 5 January 2026 (UTC)[reply]

Change your email address to nothing, save, change it to your new address, save. — xaosflux ^Talk 10:37, 5 January 2026 (UTC)[reply]

Now if the problem is that you can’t log on along the way, because you are triggering email authentication verification – you need to email the address on the message for help. — xaosflux ^Talk 10:38, 5 January 2026 (UTC)[reply]

Thanks! Newystats (talk) 07:49, 6 January 2026 (UTC)[reply]

https://www.maringensoc .org/ Marin County Genealogical Society

REMOVE SPACES

QUOTE

The following link has triggered a protection filter: c.org

Either that exact link, or a portion of it (typically the root domain name) is currently blocked.

UNQUOTE Piñanana (talk) 17:29, 5 January 2026 (UTC)[reply]

@Beetstra ^ Izno (talk) 17:31, 5 January 2026 (UTC)[reply]

@Piñanana@Izno .. crap, you solve one issue, create an other. My apologies, https://www.maringensoc.org should now work. Dirk Beetstra ^{T C} 18:10, 5 January 2026 (UTC)[reply]

I do small article creations at the ms wiki and stumbled upon a big cluster of articles which were desynced from their respective Wikidata. There wouldn’t happen to be a automatic bot or system that correctly places them for the en wiki are there?

If there is I’d be glad to introduce the system to the ms wiki to help clear out the thousands of desynchronization articles. Most are fairly obscure species of bugs but little clean-ups like that feels satisfying to resolve

PeepeeDino (talk) 18:45, 5 January 2026 (UTC)[reply]

Usually I would advise Duplicity, but we’re talking about 12k pages here. Sjoerd de Bruin (talk) 22:05, 5 January 2026 (UTC)[reply]

This doesn’t seem like an issue related to the English Wikipedia. If you want to perform bot edits on ms.WP, you would have to locate the bot operators’ noticeboard on that site. I don’t see one linked from Wikidata, but that might be something for your list. If you want to perform bot edits on Wikidata, same basic answer. – Jonesey95 (talk) 22:26, 5 January 2026 (UTC)[reply]

Very well then. Thanks for the help!

PeepeeDino (talk) 06:03, 6 January 2026 (UTC)[reply]

Can someone explain why File:Chuck Norris May 2015.jpg and the caption from another thumbnailed image appears inside {{infobox martial artist}} in this section Chuck Norris#Martial arts knowledge when neither the image nor the caption are stated in the infobox? Nthep (talk) 22:10, 5 January 2026 (UTC)[reply]

They are coming from wikidata:Q2673#P18 but obviously the template is not pulling the correct caption — Martin (MSGJ · talk) 22:17, 5 January 2026 (UTC)[reply]

A similar question came up at Template talk:Infobox#Mystery image appearing at top of InfoBox – has there been a change somewhere? —Redrose64 🦌 (talk) 22:23, 5 January 2026 (UTC)[reply]

There needs to be a simple and consistent way to suppress {{module:InfoboxImage}}. This infobox doesn’t support |suppressfields=image as suggested at Template talk:Infobox#Mystery image appearing at top of InfoBox. Using |image=blank.jpg will suppress the image but not the caption. Nthep (talk) 22:36, 5 January 2026 (UTC)[reply]

I agree the whole system is a bit of a mess and harmonisation is needed. For now I have added a manual caption to Chuck, which fixes the immediate issue. I note that Template:Infobox martial artist is using Module:Wikidata which is marked as deprecated, so the template probably needs to stop using that module — Martin (MSGJ · talk) 09:20, 6 January 2026 (UTC)[reply]

@Nthep I switched the infobox to use Module:WikidataIB, so |suppressfields=image is now supported. —Ahecht (TALK
PAGE) 15:10, 6 January 2026 (UTC)[reply]

Thanks. Nthep (talk) 11:29, 7 January 2026 (UTC)[reply]

User:LaundryPizza03/CSD log is not being archived by ClueBot III (talk · contribs). I don’t know why it stopped working. –LaundryPizza03 (dc̄) 23:53, 5 January 2026 (UTC)[reply]

There’s only one thread eligible for archiving, and it’s only been eligible for four days. Maybe you just need to wait. —Redrose64 🦌 (talk) 00:27, 6 January 2026 (UTC)[reply]

@LaundryPizza03: User:LaundryPizza03/CSD log says archiveprefix=User:LaundryPizza03/XfD log/, but it should be a subpage per User:ClueBot III/ArchiveThis#archiveprefix. My question isn’t why it stopped archiving but why it archived previously like here when it had the same bad archiveprefix. Bots usually don’t archive at all when the prefix is invalid but maybe the bot code changed. Try setting archiveprefix=User:LaundryPizza03/CSD log/Archives/. PrimeHunter (talk) 19:26, 6 January 2026 (UTC)[reply]

ClueBot III doesn’t require the archive target to be a subpage of the archived page (unless that changed recently and I’m not aware), though it can cause very strange behaviour if not set up properly. Lowercase sigmabot III will outright refuse to archive a page if it is not a subpage of the current page. I’m not sure why cluebot stopped archiving in this case. Aidan9382 _(talk) 21:20, 6 January 2026 (UTC)[reply]

I don’t know whether it’s enforced but User:ClueBot III#key says: “The value of this parameter must match an internally generated key in order for the archives to be stored anywhere other than as subpages of the page being archived”. User:LaundryPizza03/CSD log has no key. The previous archiving was to subpages and not to the non-subpages specified by archiveprefix. PrimeHunter (talk) 23:42, 6 January 2026 (UTC)[reply]

Yes, but when the archive target is set incorrectly, ClueBot III tries to guess what it should be and archives accordingly, sometimes with odd results, as noted above. Graham87 (talk) 04:29, 7 January 2026 (UTC)[reply]

That is fixed now, and archiving should begin soon. –LaundryPizza03 (dc̄) 23:45, 6 January 2026 (UTC)[reply]

It is still not archiving. –LaundryPizza03 (dc̄) 08:08, 9 January 2026 (UTC)[reply]

The configuration looks OK to me. The page was archived most recently in early December. The configuration says that it should leave 12 threads, and you have 13, so there might be an off-by-one error somewhere, or one of the threads may have some goofy formatting in it. I would wait until February to see if one or two of the threads gets archived. – Jonesey95 (talk) 13:24, 9 January 2026 (UTC)[reply]

Hi everyone,

We’re sharing an early look at an exploration into improving how people search on Wikipedia in order to gather your input. The goal is to help readers find the information they’re looking for more easily on Wikipedia itself, without needing to rely on external search engines.

One focus of this work is semantic search, a type of search that looks at the meaning of a query, not just the exact words typed, to help people find information resources. Today, Wikipedia search relies almost entirely on keyword matching, which works well when readers know the exact article they want, but less well when they have a question or are exploring a topic and the answer is inside an article without a keyword title match.

This post outlines why we’re exploring this, what our early research shows, and what kind of small, early experiment we’re considering.

Why are we working on this?

Many readers do not start their searches on Wikipedia. Instead, they often use external search engines or AI-powered tools, which then direct them to Wikipedia – or sometimes provide answers based on Wikipedia content without sending readers to the site at all.

If Wikipedia search does not meet modern expectations, especially for question-based or exploratory searches, readers are less likely to begin or continue their journeys here and instead rely on platforms where information isn’t made by humans, and is less reliable, neutral, and complete.

In short, improving search is one way to help Wikipedia readers find and enjoy what they read on our platform.

What has our preliminary research shown?

Our early research checked whether this problem is real and whether improving search could meaningfully help readers. Our findings suggest that it could.

1. About 98% of Wikipedia reading sessions originate outside Wikipedia search.

• The small group who use internal search are much more likely to be editors than casual readers. Most readers move between articles by returning to external search engines, even when links exist within Wikipedia itself.

2. Roughly 80–95% of on-wiki search sessions use autocomplete suggestions.

• The preference for autocomplete suggestions – those that appear as someone types – shows that small improvements to speed can have a large impact on success.

3. Between 4–7% of Wikipedia search queries are phrased as questions, but these queries are less likely to succeed.

• While this is a minority of searches, it shows that some readers attempt it and that many others likely avoid it because they’ve learned it doesn’t work.

What stage is this project in?

This work is currently in Phase 0, sharing early ideas, learning from research, and gathering community input.

What idea are we testing?

We’re exploring whether a hybrid search experience, one that combines keyword search with semantic search, could help readers find information more easily. The hybrid search would use machine learning, similar to how search engines rank and surface results today, to better match readers’ queries with relevant existing articles and sections.

Semantic search performs better for questions and exploratory searches, while keyword search works better for very specific or name-based queries. In early prototypes, combining the two approaches produced more useful results than either one alone.

Importantly, this exploration does not involve generating new answers or rewriting Wikipedia content. The goal is to better match readers’ queries to existing, editor-created articles and sections. Any experiment in this area would be small, limited, and designed to test whether this approach provides real value to readers.

What is the timeline?

Right now, we’re focused on discussing the problem space and sharing the findings of the report with you all. We especially want to understand if this problem space is worth learning more about. We are also trying to better understand if a simple Minimal Viable Product could be technically feasible.

Should there be alignment around further exploration, a possible next step would be a tightly constrained, time-bound A/B test with a limited group of readers, potentially beginning in February, to help answer open questions surfaced together.

What input are we looking for from you?

We’d especially appreciate your thoughts on the following:

1. What are your overall reactions to this exploration and the research behind it?
2. Are there risks or concerns you think we should be paying closer attention to?
3. What signals or outcomes would matter most in deciding whether a hybrid search approach is worth pursuing further?

For more details, including links to our research and early mockups, please see the project page.

Thank you! EBlackorby-WMF (talk) 16:09, 6 January 2026 (UTC)[reply]

Google is not a good role model here, given the amount of negative media coverage they have received in the last few years about how they are currently destroying their flagship product, in part by messing with people’s search queries. Bringing this up as something Google “excels at” is actually baffling given how constant and how inescapable the negative commentary has been. (But I guess these are just “Internet pundits” and the feature isn’t for them.)

Also, based on the project page, this seems to go well beyond “improving search.” The mockups that claim to be “semantic search” appear to actually illustrate a “Because you liked…” recommendation feature, and a “suggested questions” widget, which surfaces AI-generated questions for no apparent reason. (For Q&A use cases, AI-generated questions can potentially introduce factual or representational bias. — you don’t say!)

(Neither here nor there: I’m pretty sure that the project page was at least edited, if not fully generated, with AI.) Gnomingstuff (talk) 04:41, 7 January 2026 (UTC)[reply]

@Gnomingstuff Google is good at semantic interpretation (and many other things as well). It’s not as good at giving the right answer as it USED to, but that has more to do with the overall pool of garbage that they have to sort through, not keeping up with development of their own platform’s core functionality, as well as actively worsening their quality because they want to keep you coming back to their website.

On a scale of 1 through 10, they are however still at 7. If people want to focus on Google messing up the 7-10 part, then they are ignoring that WE are at 0 and not even close to their 7.

this seems to go well beyond “improving search.”

I think that depends on if your definition of search is that of someone used to old style search (a specific word matching technology), or the actual meaning for most people in the world (finding what they are looking for). —TheDJ (talk • contribs) 14:50, 7 January 2026 (UTC)[reply]

The post here seems to only discuss “old style” search: We’re exploring whether a hybrid search experience, one that combines keyword search with semantic search, could help readers find information more easily.. What the Phase 0 proposal page seems to actually be proposing are “since you liked…” and AI-generated questions widgets, with proof-of-concepts already built out.

I also guess I don’t see the problem with people getting to Wikipedia pages via Google rather than internal search or blue links. They still get to the page either way, the only difference is squeezing out 2 extra pageviews.” Gnomingstuff (talk) 20:21, 7 January 2026 (UTC)[reply]

I feel WMF developers should be informed somehow that any LLM-generated text placed within the Wikipedia UI is unlikely to receive a positive community reaction (either due to anti-AI sentiment or cautious acceptance but feeling Wikipedia’s human-written nature is what gives it a purpose distinct from ChatGPT etc.). Otherwise they will continue to spend effort on stuff that is unlikely to be accepted by the editing community.

On the core idea of semantic search, I somewhat disagree. Yesterday I was searching for some New Zealand-related topics. I typed NZ {phrase}, but it didn’t come up with the correct article; although it works often enough for me to instinctively do it, stuff like this still fails for me a reasonable portion of the time. Having a search system that handles this sort of thing would be a better way of handling this rather than creating redirects for every conceivable topic.  novov talk edits 09:28, 8 January 2026 (UTC)[reply]

“with proof-of-concepts already built out” proof of concepts get built all the time for a variety of reasons. to spark discussion, to visualize ideas etc etc. Stop telling people what to do. —TheDJ (talk • contribs) 12:00, 8 January 2026 (UTC)[reply]

…I didn’t tell anyone what to do? Pointing out the contents of a page is not telling anyone what to do and I have no idea how you are twisting “proof of concepts are built out” into “I order you to do XYZ.”

…also, the entire point of this topic was to ask for feedback, in part to help decide what to do Gnomingstuff (talk) 14:31, 8 January 2026 (UTC)[reply]

Gnomingstuff, I’m pretty sure that the project page was at least edited, if not fully generated, with AI. – No, I’m pretty sure it isn’t. WP:AISIGNS is not a good metric for finding AI-gen pages outside of encyclopedic articles. A lot developers write with bolded phrases as signposts and bulleted lists since it makes it easier to skim documents. with proof-of-concepts already built out I can personally tell you that those prototypes could be built at speed, over a evening if required. (in fact, if I were to speculate, [3] (which is their other design) took more time to create than the AI question screenshot).

That being said,

EBlackorby-WMF, wrt What signals or outcomes would matter most in deciding whether a hybrid search approach is worth pursuing further? I do want to amplify one specific criticism by @Gnomingstuff and @Mir Novov, I do think The robot icon and AI label failed to clarify what was machine- vs human-generated and instead increased confusion. is a alarming finding that should have been a guard-rail, we cannot have Although the questions were AI-generated, participants often assumed questions were crowdsourced or editor-curated be a thing. The (enwiki-) community cares deeply about making sure folks understand that the encyclopedia is human-generated and organic and having some mixing of AI-gen content/attribution of human-gen content to AI will likely be poorly recieved by the community.

Moving to my personal feedback, I really really like “Concept 1” (the “because you read” feature). It’s something I’ve been missing a fair bit. I really dislike the Q/A interfaces (mostly due to the reasons that folks in the thread have outlined). I’m personally ambivalent to the ask.toolforge.org prototype, though I wonder if instead to prompting the question, the interface could silently engage the reader to page snippets? (For example, iff a user types in “Company that owns biggest browser”, y’all could silently switch from using CirriusSearch to surfacing Google#Google Chrome or similar). Sohom (talk) 14:19, 8 January 2026 (UTC)[reply]

I specifically had in mind stuff like “underscoring the need for transparent provenance labeling” and “align with Wikimedia’s infrastructure and privacy standards.” As I said it’s neither here nor there, this kind of document at any workplace is more likely than not to be AI-assisted and there’s no rule against it. Gnomingstuff (talk) 14:33, 8 January 2026 (UTC)[reply]

I wonder if the AIs learned to write that way from the kind of techno-corporate bureaucratese that certain types of people have been writing for a long time? Anomie⚔ 00:06, 9 January 2026 (UTC)[reply]

(Probably, which is why I suspect no AI usage, that language is a bit corporate-speaky, but I’ve seen it used in earnest were folks were not using GPT) Sohom (talk) 10:51, 9 January 2026 (UTC)[reply]

@Sohom Datta and @Mir Novov, I appreciate your notes and acknowledge your concern around perceived mixing of AI- and human-generated content, as it’s one we take very seriously as well. User testing showed that the bot labeling was confusing and would have to be completely rethought if Q&A explorations are found to be worth continuing. Right now it’s clear that the current iteration of Q&A is not in production- or even an experiment-ready state.

However, there does seem to be interest in improvements to the search to support semantic-style queries. If this continues to be the case, we would return with a proposal for an experiment around search improvements and establish guardrails and goals together. Similarly, if there is another concept for improving wayfinding (that wouldn’t lead to confusing AI- with human-generated content) we’d return again to discuss before proceeding.

Interesting idea on silently engaging the reader to page snippets! Would this look like directly taking the user to the relevant heading instead of the top of the article? Do you have any other suggestions for areas for us to look into for information-finding? EBlackorby-WMF (talk) 20:47, 9 January 2026 (UTC)[reply]

@Gnomingstuff, thanks for your feedback. It helped us realize we uploaded a still of the design concept instead of a GIF, fixed here, which may have led to confusion about this work focusing on “Because You Read”-style suggestions.

While Google is definitely not a role model for this work, it is the most popular global search engine for people to discover content from Wikipedia, thus making it an important tool for understanding modern reader expectations. We’re trying to see how we can support the kind of queries users are used to making.

As for people coming to Wikipedia from Google, readers are now often simply reading the previews provided by Google Knowledge Panels and AI summaries (or their LLM of choice), rather than visiting the articles themselves. This tendency could harm their understanding of both the article content and the Wikipedia project, as well as prevent them from potentially becoming editors.

Do you have any other thoughts on ways to improve findability of article information for readers? EBlackorby-WMF (talk) 20:39, 9 January 2026 (UTC)[reply]

Asking for help to remove Talk page. In my wanderings around Wikipedia, I happened to click on a wikilink that took me to the “ha.wikipedia.org” encyclopedia (Hausa language). First I did blank the Talk page & that did not work. Next, I added {{db-authorreq}} template & that showed up as a Redlink. So I’m hoping an expert here can help, as I do not need or want that talk page. Regards, JoeNMLC (talk) 23:52, 6 January 2026 (UTC)[reply]

It is normal for many Wikipedias to create talkpages using bot welcomes. Generally however even on en.wiki it is common for other editors to create a user’s talk page. It would be best to ignore this. CMD (talk) 00:09, 7 January 2026 (UTC)[reply]

@JoeNMLC: Each Wikipedia language has its own templates, rules and procedures. Template:Speedy delete (Q4847311) shows our general {{db}} should correspond to {{gogewa}} in the Hausa Wikipedia. I don’t know their reaction to using it for this but at least the template exists and adds the page to ha:Category:Candidates for speedy deletion. Some of the pages have been there for months so a review may not be so speedy. Your Hausa account was created automatically 6 January, probably because you viewed a page there. Later a bot posted a welcome message to your talk page. If the page is deleted then I don’t know whether welcome bots will stay away from it in the future. meta:Requests for comment/Welcoming policy discusses a proposed policy to only allow welcome messages to users if their account was originally created at the wiki, or the user has at least one non-imported edit there. You now have edits to your talk page so you would no longer be covered by the policy if it’s adopted. Some of us have received a lot of these welcome messages in languages we don’t know. PrimeHunter (talk) 00:31, 7 January 2026 (UTC)[reply]

@JoeNMLC: Our page Wikipedia:Speedy deletion isn’t linked to any page on ha:, so maybe they don’t have a speedy deletion policy. Even if they did, the criteria (and the codes for them) would certainly differ from ours; ha:Samfuri:Db-authorreq certainly doesn’t exist. Now, at English Wikipedia {{db-authorreq}} is strict – the criterion that this is for, WP:CSD#G7, applies If requested in good faith and provided that the only substantial content of the page was added by its author. – the sticking point here is that the only substantial content was added by ha:User:AmmarBot, not yourself, so we would decline the request on those grounds. Also at English Wikipedia we do not delete user talk pages except in very exceptional circumstances – such as newly-created pages where the only edits so far are clearly covered by WP:CSD#G10 and WP:CSD#G12. It’s probable that other Wikipedias have similar restrictions. —Redrose64 🌹 (talk) 08:50, 7 January 2026 (UTC)[reply]

Deleting a page on that project is something any of their administrators can do. Though really, my suggestion for a better solution is to replace all the text with [[#REDIRECT:w:en:User talk:JoeNMLC]]; which would make it clear to anyone there that if they want to talk to you they should come do it here. — xaosflux ^Talk 10:54, 7 January 2026 (UTC)[reply]

Thanks for all the above info. I did successfully add the redirect, but did change the wikicode to #REDIRECT:[[w:en:User talk:JoeNMLC]] with the left brackets location changed. Cheers! JoeNMLC (talk) 13:43, 7 January 2026 (UTC)[reply]

oops, yes, that was a typo above! You should put that line as the first line of the page (no “:” needed between), you can leave notes below it if you want, which I think should also turn off the discussion tools prompt for others (else you can use __NOTALK__ under it. — xaosflux ^Talk 14:29, 7 January 2026 (UTC)[reply]

It seems interwiki redirects are disabled on hawiki. Children Will Listen (🐄 talk, 🫘 contribs) 14:33, 7 January 2026 (UTC)[reply]

@ChildrenWillListen: It’s not that interwiki redirects are disabled on hawiki – they’re disabled pretty much everywhere (see the blue box at mw:Help:Redirects#Types of redirects); the thing to do is set up a soft redirect. For hawiki, the template is ha:Samfuri:Softredirect so JoeNMLC should add the code

{{Softredirect|w:en:User talk:JoeNMLC}}

to ha:Tattaunawar user:JoeNMLC. —Redrose64 🌹 (talk) 18:31, 7 January 2026 (UTC)[reply]

Today, I added that soft-redirect above in addition to the hard-redirect. Interesting that both work correctly. Cheers, JoeNMLC (talk) 12:51, 8 January 2026 (UTC)[reply]

your also lucky that hawiki let’s you blank the page. Several others don’t let “new” users do this. Nthep (talk) 14:37, 7 January 2026 (UTC)[reply]

You can just use {{subst:void}} in that case, assuming the wiki in question has such a template. Children Will Listen (🐄 talk, 🫘 contribs) 14:39, 7 January 2026 (UTC)[reply]

That’s pointless. When {{subst:void}} is used, nothing is saved. JoeNMLC effectively did that already, by blanking the page. —Redrose64 🌹 (talk) 18:31, 7 January 2026 (UTC)[reply]

Yeah I know, but it gets around some annoying abuse filters when you’re trying to blank xwiki vandalism. Children Will Listen (🐄 talk, 🫘 contribs) 17:44, 8 January 2026 (UTC)[reply]

 Done – Thank you everyone for the help. Cheers! JoeNMLC (talk) 15:06, 9 January 2026 (UTC)[reply]

In the past few days I have encountered what seems to me to be radical attack on one of the foundations of Wikipedia, individual editorial contribution. In trying to track the history of a heavily debated Wikipedia article, on Heat, I have found my edits to be overwritten by what seems to be an AI algorithm. Perhaps I could work around this. But Wikipedia has worked with vandalism for decades. This isn’t a matter of dealing with vandalism. It’s a matter of a crippling “improvement”.Chjoaygame (talk) 03:29, 7 January 2026 (UTC)[reply]

Almost all the edits to that talkpage for the last month or so are by you (plus one reply by another user and one bot archiving some old sections). I don’t see any overwriting – could you link to sone examples of the problem you are seeing? Andrew Gray (talk) 04:04, 7 January 2026 (UTC)[reply]

Thank you for your response. I don’t see how to give you such a link.

The article on Heat has problems. I am trying to tackle some of them by examining, on the Talk page, the edit history of the article, which is long and loaded with conflict.

This involves quoting snips of historical versions of the article. Apparently, the AI sees that I am doing so, and instead of allowing me to post what I write, it imposes its own “improved” version of what I write. That practically ‘over-writes’ or obliterates what I have written.

In the past, there were a fair number of editors who took a constructive and well informed interest in the article, but now there are very few. The latest substantial edit to the article practically erased the work of the earlier well informed editors. I don’t like to try to deal with that by directly editing the article, for fear of edit conflict. So I try to deal with it by writing to the Talk page. As you observe, my efforts on the Talk page get practically no response. I see two ways of dealing with that. One is that I should abandon my efforts. The other is to quietly continue to make sober edits to the Talk page.

The article is on a fundamental concept of thermodynamics, a topic that has its conceptual subtleties. The article also attracts edits from editors on other topics and viewpoints, such as engineering. They may be expert in their respective topics, but thermodynamics has its own history and logic. I think it important that Wikipedia offer a sound article on heat as it appears in thermodynamics, which is a fundamental topic for physics. There is in Wikipedia a sort of companion article with a specifically engineering viewpoint, with the title Heat transfer. At present, the article on Heat starts with the words “In thermodynamics,…”

It is laborious to examine such an edit history. I can try to get past the deleterious actions of the AI, but it worries me that I seem to have to struggle with an AI that seems not to understand the task in hand.

Perhaps you can tell me how to give you a link to the inner workings of my efforts to write on the Talk page?Chjoaygame (talk) 09:30, 7 January 2026 (UTC)[reply]

I think there might be a misunderstanding here. Wikipedia doesn’t really have any kind of “AI” that changes your talkpage comments, so I can’t see what this might be. Can you give an example of a talkpage edit you think was “overwritten” by the AI? A link to the diff would help.

Our bots will show up in the history as edits, and they aren’t involved here. We do have some edit filters which (for example) try to prevent links to spam sites, but they wouldn’t change the text you write – they would either warn you or prevent the edit saving. Is it possible that there is some “helpful” AI tool in your own web browser that is trying to rewrite the text you are copy/pasting? Andrew Gray (talk) 10:36, 7 January 2026 (UTC)[reply]

Thank you for your kind help. I am glad to read you writing that “Wikipedia doesn’t really have any kind of “AI” that changes your talkpage comments”.

I am a little worried about the possibility that you mention, of “some ‘helpful’ AI tool in your own web browser that is trying to rewrite the text you are copy/pasting”. I think it isn’t restricted to copy-and-pasting, but I will need to check that. Does Microsoft Copilot poke so far into the works? I have the ‘Personal’ version of Copilot that, so far as I know, doesn’t intrude on anything that I do outside the app. The problem happens also on my non-Microsoft Android machine. I am glad to read you using the word ‘helpful’ in much the same way as I use the word ‘improvement’ !!

When I write an edit, and ask for ‘Preview’, my effort is shown as the agent’s ‘improved’ version; my version doesn’t appear. So I don’t try to post my edit. I can’t show you an actually posted output of the algorithm, without posting something that I think is faulty. When it happens, I either give up, or try a work around.

It won’t be too easy for me to show you in explicit detail what happens. I will try to work out how to show you what happens. I think this will take me some time, so that I can’t do it right away.Chjoaygame (talk) 11:00, 7 January 2026 (UTC)[reply]

• I suspect this isn’t the talkpage but the article Heat itself, where the OPs attempts to rewrite the lead have been reverted by multiple other editors? So there’s not really anything to do here; this is a content dispute, not anything suspicious. Black Kite (talk) 10:41, 7 January 2026 (UTC)[reply]

  Thank you, Black Kite, for your suggestion. I am talking about attempts to edit the Talk page, not the article itself. As I noted, I have in this situation not attempted to edit the article itself.

  My concern is not about content disputes or reverts, but about the editing interface itself — specifically, the preview behavior and the apparent overwriting of draft text before submission. That’s not something other editors can revert. It’s something that happens before the edit is even posted.

  I am naming a problem that lives in the infrastructure, not in the article.Chjoaygame (talk) 11:10, 7 January 2026 (UTC)[reply]

  Mm, if that’s the case, I would look at your own setup. Do you have any browser extensions (for example, grammar checkers or language translators) set up which may be rewriting your input? Can you give us an example of what changes the problem causes? Black Kite (talk) 11:40, 7 January 2026 (UTC)[reply]

  Or a screen recording? — Qwerfjkltalk 12:42, 7 January 2026 (UTC)[reply]

  Thank you for your attention. It seems that perhaps the problem was that I wrote an unclosed ‘ref’, as indicated below by Editor PrimeHunter. He has kindly removed my error. Next time I edit, I will find out whether that was the problem. For the present, I will just wait and see. Chjoaygame (talk) 14:54, 7 January 2026 (UTC)[reply]

@Chjoaygame: It’s hard to guess what you have in mind without any examples but maybe you don’t understand the difference between source text and rendering. Source text (also called wikitext) is what you write and see in the text area where you write during an edit (unless you use VisualEditor). Some of it looks like computer code. Rendering is how it displays in preview and after saving. Our MediaWiki software processes the “computer code part” of the source text and uses it to make the rendering, e.g. formatting references. Several errors in source text can cause some of the text to not be rendered. One of them is tags which are opened but not closed. I removed an open ref tag here. It should have been closed with </ref> if there was an actual reference. Another thing: I can see you have copied some text from the rendered version of old revisions and pasted it into the source text in the edit window. This causes differences. You can click the “Edit” tab on the old revision and copy the source text if you want your copy to look more like the original. Sections don’t have edit links in old revisions so you have to edit the whole page and find the right place to copy from. If “Edit” opens VisualEditor for you (it depends on your preferences and the circumstances) then switch to the source editor on a pencil icon at the top right before copying the text. PrimeHunter (talk) 14:24, 7 January 2026 (UTC)[reply]

Thank you, Editor PrimeHunter, for your consideration of my problem. Perhaps, as you suggest, my writing an unclosed ‘ref’ is/was the problem. You have kindly removed that unclosed ‘ref’. Perhaps that has solved the problem. I will see if the problem is still there next time I edit. As I read you, I wrote source text (also called wikitext) and when I asked to see the preview, the machine didn’t work as I expected it to. It rendered something else. In other words, I wrote a syntactic error. Chjoaygame (talk) 14:48, 7 January 2026 (UTC)[reply]

Why is the last item (“December”) in ths following template detached from the other items?
Wikipedia:Graphics Lab/Illustration workshop/Archive/2025

In Wikipedia:Graphics_Lab/Illustration_workshop/Archive/2025, there doesn’t seem anything different about it, other than being the last item. Adding or removing items retains this behaviour.

Thanks, cmɢʟee τaʟκ (please add {{ping|cmglee}} to your reply) 12:51, 8 January 2026 (UTC)[reply]

@Cmglee I don’t know how the Mbox template and/or MW’s html rendering works, but the last element was being rendered inside its own paragraph (<p>...</p>) tag. Wrapping them all in a single paragraph tag seems to have fixed it. — DVRTed (Talk) 13:42, 8 January 2026 (UTC)[reply]

@Cmglee: I used Special:ExpandTemplates and tested how much code could be removed while preserving the effect.

<table><tr><td>
January
February
March</td></tr></table>

produces:

<table><tr><td>
January
February
March
</td></tr></table>

produces:

In the first case something (maybe mw:RemexHtml?) inserts a paragraph <p>...</p> in the HTML when </td> is on the same line as the last text. I don’t know why. You can for example avoid it by adding a line with <nowiki/> after the months in Wikipedia:Graphics Lab/Illustration workshop/Archive/2025, or by removing the newlines. PrimeHunter (talk) 13:45, 8 January 2026 (UTC)[reply]

<p>...</p> is also inserted in the second case but in the first case it ends before March</td>. PrimeHunter (talk) 14:03, 8 January 2026 (UTC)[reply]

(edit conflict × 2) The {{mbox}} template is producing output like this:

<templatestyles src="Module:Message box/ombox.css"></templatestyles><table class="plainlinks ombox ombox-notice" role="presentation"><tr><td class="mbox-image">[[Image:Vista-file-manager.png|100px|Archive]]</td><td class="mbox-text"><span style="font-size:105%;">'''Archives of 2025:'''</span><br/>
[[Wikipedia:Graphics Lab/Illustration workshop/Archive/Jan 2025|January]], 
[[Wikipedia:Graphics Lab/Illustration workshop/Archive/Feb 2025|February]], 
[[Wikipedia:Graphics Lab/Illustration workshop/Archive/Mar 2025|March]], 
[[Wikipedia:Graphics Lab/Illustration workshop/Archive/Apr 2025|April]], 
[[Wikipedia:Graphics Lab/Illustration workshop/Archive/May 2025|May]], 
[[Wikipedia:Graphics Lab/Illustration workshop/Archive/Jun 2025|June]], 
[[Wikipedia:Graphics Lab/Illustration workshop/Archive/Jul 2025|July]], 
[[Wikipedia:Graphics Lab/Illustration workshop/Archive/Aug 2025|August]], 
[[Wikipedia:Graphics Lab/Illustration workshop/Archive/Sep 2025|September]], 
[[Wikipedia:Graphics Lab/Illustration workshop/Archive/Oct 2025|October]], 
[[Wikipedia:Graphics Lab/Illustration workshop/Archive/Nov 2025|November]], 
[[Wikipedia:Graphics Lab/Illustration workshop/Archive/Dec 2025|December]]</td></tr></table>

For some reason, MediaWiki is transforming that by wrapping everything in between the first and last lines in a <p>...</p>. There are a few ways this might be fixed, for example changing the output to have the </td></tr></table> on a separate line avoids “December” being after the </p>. [4] suggests this isn’t a WP:THURSDAY thing, but might still be worth reporting in Phabricator. Anomie⚔ 13:55, 8 January 2026 (UTC)[reply]

Thanks, everyone, for the speedy resolution. Guess the simplest workaround for now is to add an empty item after December. Can someone please report on Phabricator, as I’m unsure how to phrase the technical problem? Thanks, cmɢʟee τaʟκ (please add {{ping|cmglee}} to your reply) 15:24, 8 January 2026 (UTC)[reply]

A better fix would be to use {{hlist}} or equivalent, such as {{cslist}}. That aside, this also happens with the div version of mbox (you can test with {{cmbox}}). I suspect it has to do with p-wrapping, which is always a fun discussion and so I suspect also that a new task would be closed duplicate. Izno (talk) 17:05, 8 January 2026 (UTC)[reply]

It can also be fixed by eliminating all those newlines. That is, instead of this:

|text=
<span style="font-size:105%;">'''Archives of 2025:'''</span><br/>
[[Wikipedia:Graphics Lab/Illustration workshop/Archive/Jan 2025|January]], 
[[Wikipedia:Graphics Lab/Illustration workshop/Archive/Feb 2025|February]], 
[[Wikipedia:Graphics Lab/Illustration workshop/Archive/Mar 2025|March]], 
...
[[Wikipedia:Graphics Lab/Illustration workshop/Archive/Oct 2025|October]], 
[[Wikipedia:Graphics Lab/Illustration workshop/Archive/Nov 2025|November]], 
[[Wikipedia:Graphics Lab/Illustration workshop/Archive/Dec 2025|December]]

use this:

|text=
<span style="font-size:105%;">'''Archives of 2025:'''</span><br/>[[Wikipedia:Graphics Lab/Illustration workshop/Archive/Jan 2025|January]], [[Wikipedia:Graphics Lab/Illustration workshop/Archive/Feb 2025|February]], [[Wikipedia:Graphics Lab/Illustration workshop/Archive/Mar 2025|March]], ... [[Wikipedia:Graphics Lab/Illustration workshop/Archive/Oct 2025|October]], [[Wikipedia:Graphics Lab/Illustration workshop/Archive/Nov 2025|November]], [[Wikipedia:Graphics Lab/Illustration workshop/Archive/Dec 2025|December]]

But it certainly indicates a bug in MediaWiki. —Redrose64 🌹 (talk) 23:10, 8 January 2026 (UTC)[reply]

Here is another simple example of the issue where F is rendered on a new line because it’s on the same source line as </td>. I have reported it in phab:T414152.

<table>
<tr><td>
A
B
C
</td><td>
D
E
F</td></tr>
</table>

produces:

The generated HTML shows the different placement of </p>, after C but before F:

<table>
<tbody><tr><td>
<p>A
B
C
</p>
</td><td>
<p>D
E
</p>
F</td></tr>
</tbody></table>

PrimeHunter (talk) 01:37, 9 January 2026 (UTC)[reply]

When working with Wikipedia:Database reports/Uncategorized categories, a user has the option of simply tagging the category as {{uncategorized}} if they don’t already know the appropriate categories to add it to, which also works to prevent the category from being picked up again on the next new run of the report.

However, if the category name has a slash in it, such as Category:Created or significantly improved through WikiPortraits/2026 Sundance Film Festival or Category:J/70, then the template fails to transclude the Category:Uncategorized from January 2026 category, and thus fails to actually put the page into the tracking category for uncategorized content — meaning that categorizing it doesn’t get addressed, and it just stays on the report through multiple runs, unless an editor manually adds Category:Uncategorized from January 2026 as a direct category declaration.

Is there any way that this can be fixed so that the template doesn’t overlook its tracking category on categories with slashes in their names? Bearcat (talk) 17:52, 8 January 2026 (UTC)[reply]

{{Uncategorized}} wraps its categorization by month code in {{subpage other}}, which may just be looking for a slash in the page path. I think this might be the culprit. – Jonesey95 (talk) 18:09, 8 January 2026 (UTC)[reply]

According to Special:Diff/1094666209, template {{Subpage other}} is there only to exclude /doc and /sandbox pages. This means that it can be easily replaced with {{Sandbox other}}, which excludes both /sandbox and /doc, despite its name. —⁠andrybak (talk) 19:41, 8 January 2026 (UTC)[reply]

Excellent. I have boldly swapped out one template for the other (sandbox other for subpage other), and I previewed an old version of Category:J/70 to ensure that it worked as designed. It did. There may, of course, be side effects, like other types of subpages being dragged into the “Uncategorized” categories, but that is now by design. If there are only a few, they should be cleaned up manually. – Jonesey95 (talk) 22:25, 8 January 2026 (UTC)[reply]

I’m trying to rescue a source I lost while faffing around during a major copy edit to Zero trust architecture. How do I invoke IABot to get it back? guninvalid (talk) 22:15, 8 January 2026 (UTC)[reply]

Go to [5] and type the article title. Be sure not to tick the optional choice to add archives to all links, because the page will time out. However, I just ran it and all links are shown to be live, so there must be a problem with something else. Maybe you accidentally deleted the ref? Szmenderowiecki (talk · contribs) 00:53, 9 January 2026 (UTC)[reply]

I used to tick ” optional choice to add archives to all links” if I want to archive all of the links on an article and add all archival links to the article. Is it not working now? ✠ SunDawn ✠ Contact me! 02:06, 9 January 2026 (UTC)[reply]

I tried to use that on a couple of my GAs, and it would consistently return a 504 error (timeout). I can’t tick that box for batch requests, which sucks. Szmenderowiecki (talk · contribs) 02:10, 9 January 2026 (UTC)[reply]

In monitoring page creations, I found that edits that create “be bold” user subpages (such as User:Aross05/be bold) as part of the Wikipedia:Training guided tour (preset) are flagged by as “may have problems” in Special:RecentChanges. –LaundryPizza03 (dc̄) 06:23, 9 January 2026 (UTC)[reply]

I do not find that unexpected. They are short and generally by noobs. The filter doesn’t know some random people wrote some text in a wiki page, telling noobs to do this, makes it non-problematic entries. —TheDJ (talk • contribs) 10:38, 10 January 2026 (UTC)[reply]

I’m not talking about my current one, but I’m trying to change it to this in wikitext but it says it’s misnested, but I have no clue what that means and the help page doesn’t help me in anyway either!

This is what it should look like:

MrDenjiLover ^🗣️ (Non-binary)

[[User:MrDenjiLover|<b style=”background-color:#ffd29c;border:2px solid orange;padding:2px;color:#ffb75e;”>MrDenjiLover]]</b> ^{[[User talk:MrDenjiLover|🗣️]]} (Non-binary) (talk) 15:12, 9 January 2026 (UTC)[reply]

MrDenjiLover, the closing b tag (</b>) should be inside the link text, not after it. Move it to before the ]]. — Qwerfjkltalk 15:16, 9 January 2026 (UTC)[reply]

oh. my. god.

🤦

HOW DID I MESS THAT UP MrDenjiLover ^🗣️ (Non-binary) 15:18, 9 January 2026 (UTC)[reply]

MrDenjiLover, I’d strongly suggest you change the background and/or the text because without highlighting I can barely see your username/nickname even though I have no problem with colours. It’s about accessibility because there’s almost no contrast whatsoever. Szmenderowiecki (talk · contribs) 15:31, 9 January 2026 (UTC)[reply]

fixed
MrDenjiLover ^🗣️ (Non-binary) 15:36, 9 January 2026 (UTC)[reply]

Still terrible. I mean, I’m not an admin but we should be able to easily identify your signature “A customised signature should make it easy to identify your username”. I’m not saying what you should change your signature to but you should make your signature at least moderately visible. The webpage has sliders saying what contrast is good enough. You should at least be in AA level of compliance, preferably AAA; so at least 4.5:1 but 7:1 is recommended. Szmenderowiecki (talk · contribs) 15:41, 9 January 2026 (UTC)[reply]

fixed again

this is starting to annoy me though
MrDenjiLover ^🗣️ (Non-binary) 15:50, 9 January 2026 (UTC)[reply]

wait it says this isn’t required
MrDenjiLover ^🗣️ (Non-binary) 15:54, 9 January 2026 (UTC)[reply]

Still not fixed. Accessibility is not just a nice-to-have. Please fix either the foreground or background color. Thanks. – Jonesey95 (talk) 16:08, 9 January 2026 (UTC)[reply]

oh my god

its not my fault I don’t want my signature to look like TRASH with a repulsive color scheme
MrDenjiLover ^🗣️ (Non-binary) 16:10, 9 January 2026 (UTC)[reply]

DO I WANT TO USE BLUE AND VOMIT GREEN??? NO!
MrDenjiLover ^🗣️ (Non-binary) 16:11, 9 January 2026 (UTC)[reply]

We did not ask you to use blue or vomit green. This would be fixed if the background was white, not orange. From the “it’s starting to annoy me”, something like: [[User:MrDenjiLover|<b style="border:2px solid #FA6000;padding:2px;color:#FA6000;">MrDenjiLover</b>]] <sup>[[User talk:MrDenjiLover|🗣️]]</sup> (Non-binary) displaying as MrDenjiLover ^🗣️ (Non-binary).

Separately, do not use <br /> in your signatures either. Izno (talk) 16:50, 9 January 2026 (UTC)[reply]

WHY can’t I use line breaks? It makes it look better
MrDenjiLover ^🗣️ (Non-binary) 16:52, 9 January 2026 (UTC)[reply]

DONE.

IT LOOKS GARBAGE.

ARE YOU HAPPY
MrDenjiLover ^🗣️ (Non-binary) 16:57, 9 January 2026 (UTC)[reply]

AND IF I USE THIS IT LOOKS LIKE THE HUB
MrDenjiLover ^🗣️ (Non-binary) 16:59, 9 January 2026 (UTC)[reply]

IM TRYING TO AVOID CUSSING BUT THIS IS PISSING ME OFF.
MrDenjiLover ^🗣️ (Non-binary) 17:01, 9 January 2026 (UTC)[reply]

IS THIS GOOD
MrDenjiLover ^🗣️ (Non-binary) 17:03, 9 January 2026 (UTC)[reply]

The colors are fine, thank you.

No, you may not use line breaks. See WP:Signatures for other requirements. Izno (talk) 17:05, 9 January 2026 (UTC)[reply]

I submitted an AfC draft in my user sandbox. I’m not autoconfirmed so I can’t move it. Could someone please move User:FoyleView/sandbox to Draft:Guildhall Press? Thanks FoyleView (talk) 15:27, 9 January 2026 (UTC)[reply]

 Done
MrDenjiLover ^🗣️ (Non-binary) 16:13, 9 January 2026 (UTC)[reply]

Thank you – my first attempt at submitting an article so still a bit unsure as to how it all works! ~2026-18670-7 (talk) 16:55, 9 January 2026 (UTC)[reply]

Am attempting to find a deleted article, the Signs of the coming of Judgement Day. Not sure it was a good idea to delete it and would like to read what the arguments were on its deletion pro and con, and of course see how valid they are by checking the deleted article. Normally or formerly you used to find the redirect page where the the deleted article and its talk page lived, and you would find that by typing in the old title in the search box, but that no longer works. Using a acer aspire 3 with the latest version of AVG Secure Browser. Thanks for any help. Louis P. Boog (talk) 19:39, 9 January 2026 (UTC)[reply]

@Louis P. Boog Check edit history at [6], seems this was a WP:BLAR, and you can read earlier versions. Gråbergs Gråa Sång (talk) 19:45, 9 January 2026 (UTC)[reply]

{{ec}} The history of that article was not deleted, so anyone is welcome to look at what was there before it was made into a redirect, and the article’s talk-page (Talk:Signs of the coming of Judgement Day) is still intact. Another approach is to look at all article talk-pages that point to that article to see if a discussion happened elsewhere. DMacks (talk) 19:45, 9 January 2026 (UTC)[reply]

@DMacks . thank you — Preceding unsigned comment added by Louis P. Boog (talk • contribs) 21:45, 9 January 2026 (UTC)[reply]

re-pinging @DMacks:. Graham87 (talk) 02:45, 10 January 2026 (UTC)[reply]

You’re welcome! DMacks (talk) 03:18, 10 January 2026 (UTC)[reply]

I just ran across the user whose account creation date is not displayed on their Special:Contributions page. As far as I’m aware, for the past few years, the aforementioned page displays the account creation date of the user which it is currently being utilized on. Prior to engaging the editor directly to enquire if they know why it’s happening, does anyone know if is this a setting, a glitch, or something else? Steel1943 (talk) 19:47, 9 January 2026 (UTC)[reply]

Some very old accounts (prior to about 2005) don’t have an account creation date saved in the database, so there’s nothing to show. * Pppery * _{it has begun…} 19:50, 9 January 2026 (UTC)[reply]

Based on the very limited information I know about this user (literally 15 minutes ago marks the first time I’ve ever run across them), that presumption probably checks out. I decide to go ahead and engage them directly because this is a very curious case that I’d like to know more about it so I can know more about Wikipedia and its history in general. Either way though, if there is an exact date for when the account registration date started being recorded, and that date is posted somewhere on Wikipedia, please let me know as I’d like to look at it. Steel1943 (talk) 20:01, 9 January 2026 (UTC)[reply]

The oldest entries in the user creation log [7] are from 7 September 2005. PrimeHunter (talk) 20:22, 9 January 2026 (UTC)[reply]

It’s not quite that simple. The user registration field in the database is tracked separately from the user creation log (although they may have been populated at the same time). At one point old accounts’ registration dates were backfilled using their first edit date, which of course couldn’t be done for users who hadn’t edited. * Pppery * _{it has begun…} 20:26, 9 January 2026 (UTC)[reply]

user_registration, the relevant field, began being recorded in MediaWiki 1.6 (mw:Manual:User table#user_registration). That version’s public release was on 5 April 2006 (mw:MediaWiki 1.6#MediaWiki 1.6.0), though it may have been deployed to WMF wikis earlier. Users who registered before that had their user_registration field backfilled with the timestamp of their earliest edit, if they’d made any, with mw:Manual:fixUserRegistration.php. That was likely run only once, around the time 1.6 was deployed; I know for certain its last run was no later than 14:44, 24 August 2006, since that’s when the user I first researched this for first edited, having registered about a year and a half prior. —Cryptic 20:57, 9 January 2026 (UTC)[reply]

New MediaWiki versions were deployed first to Wikipedia before being publicly released even then (compare Signpost coverage of MediaWiki 1.5. Also see earlier archived discussion about this topic at Wikipedia:Village pump (technical)/Archive 220 § null user registration. Graham87 (talk) 02:40, 10 January 2026 (UTC)[reply]

Hi all, prompted by a question in The Guardian’s Saturday quiz[8] (“What element has the lowest boiling point?” – I was close with hydrogen), I found Boiling points of the elements (data page) where the table sadly does not sort as you might expect. In fact, the results are downright weird. I tried adding ! data-sort-type="number" | [[Kelvin]] etc. at the top but that didn’t work, unsuprisingly: and that’s the end of my expertise with tables. I can see what’s wrong, but can anything be done? I’d be happy to do the work myself given some explicit instructions, or maybe it’s beyond help. :> MinorProphet (talk) 14:12, 10 January 2026 (UTC)[reply]