If you want to report aJavaScript error, please followthis guideline. Questions aboutMediaWiki in general should be posted at theMediaWiki support desk. Discussions are automatically archived after remaining inactive for 5 days.
This tends to solve most issues, including improper display of images, user-preferences not loading, and old versions of pages being shown.
No, we will not use JavaScript to set focus on the search box.
This would interfere with usability, accessibility, keyboard navigation and standard forms. Seetask 3864. There is anaccesskey property on it (default toaccesskey="f" in English). Logged-in users can enable the "Focus the cursor in the search bar on loading the Main Page" gadgetin their preferences.
No, we will not add a spell-checker, or spell-checkingbot.
You can use a web browser such asFirefox, which has a spell checker.
If you changed to another skin and cannot change back, usethis link.
Alternatively, you can press Tab until the "Save" button is highlighted, and press Enter. Using Mozilla Firefox also seems to solve the problem.
If an image thumbnail is not showing, trypurging its image description page.
If the image is from Wikimedia Commons, you might have to purge there too. If it doesn't work, try again before doing anything else. Some ad blockers, proxies, or firewalls blockURLs containing /ad/ or ending in common executable suffixes. This can cause some images or articles to not appear.
Hardly the only supporter. FaviFake dangled a worm and quite a few editors took the hook, including yours truly. There is nothing wrong with dangling a worm. ―Mandruss☎ IMO.10:11, 12 October 2025 (UTC)[reply]
WaId, you have more patience with ridiculosity than I. My position is "Get over it." Last time I checked, we are the most adaptable species currently on the planet. The rest of the internet changes for most of us on a near-daily basis, and most of us absorb the change without missing a beat. Coddle people and they become weaker, but only in the environment where they are coddled. I oppose coddling. ―Mandruss☎ IMO.10:14, 12 October 2025 (UTC)[reply]
I know you're joking, but the ability to adjust this with a user script was seriously proposed. IMO the color of the box is so trivial that it's not worth fighting over. If folks want it to be purple, then let it be purple! Let them pick a new color each year, if they want. And if "I" want it to be brown, or green, or a more mauvy shade of pinky russet, then there's nothing wrong with me creating my own CSS code to do that(or, you know, asking one of y'all if you'd be willing to do that for me, because I don't know how).WhatamIdoing (talk)18:41, 12 October 2025 (UTC)[reply]
Another discussion on this page is proposing that every message box be designed to support being skinned. While it's technically possible for every design element to be written in a way that various visual features can be overridden in a skin or a personal style file, I don't think the benefit/cost ratio is sufficiently high that it should be mandated and backported for everything. If someone wants to voluntarily support skinability in a specific widget, sure, as long as it doesn't make it too difficult to find someone to maintain it in future. (It's not that hard for people who understand CSS, but there are plenty of editors who react with a "I can't do it!" when hearing the word "template", much less "CSS".)isaacl (talk)22:51, 12 October 2025 (UTC)[reply]
I would have argued this before LLM and autonomous agent coding tools became somewhat workable (with many caveats). Making a widget skinnable is one of the class of problems that automated coding agents can do competently and cheaply. I can flop a dark mode onto any existing thing with a single prompt using GPT-5. Yes, check it and make sure the LLM hasn't barfed up a hallucination, but while I would have argued themable UI was a nice-to-have prior to this generation of automation, at this point it is easy enough to do and low enough risk.Andre🚐23:12, 12 October 2025 (UTC)[reply]
As Izno says, it's pretty easy to do; no LLM necessary. The layout just needs to follow some basic CSS principles. Like I said, my concern is about mandating this for all new widgets and requiring all existing widgets to be retrofitted, for the benefit of what I think will be a small number of editors, and complicating support costs for them since their UI will differ from everyone else's.isaacl (talk)01:00, 13 October 2025 (UTC)[reply]
I add TemplateStyles on occasion solely because I want to make it easier to skin things. In some other cases it's enough just to add a class to a template or other transcluded item.Izno (talk)00:04, 13 October 2025 (UTC)[reply]
I'm going to interpret that as an endorsement of a Village Pump header that dynamically pulsates through the entire rainbow. Support.[FBDB]Sdkbtalk16:30, 16 October 2025 (UTC)[reply]
Its purpose is to make it easier to spot edits tagged as reverted; these are in contribs, diffs, history and watchlist. It works as intended in Firefox on Windows, and Safari on an Apple iPad: all three properties are active. Now according tothe CSS Text Decoration Module Level 3 documentation, this may be shortened to
It still works in Firefox, but not in Safari. Does anybody have any idea why it fails? I had considered that Safari only respects CSS Level 2, butthe doc for that does not list the separate properties - only the shorthand, with only one of my three values shown as valid. --Redrose64 🌹 (talk)21:56, 19 October 2025 (UTC)[reply]
Based on the charts at caniuse, the shorthand didn't roll out when the other properties were made stable in Safari, which happened in 2014, which is when the -webkit- versions were first supported. Kind of just sounds like it got forgotten to me.Izno (talk)22:34, 19 October 2025 (UTC)[reply]
Latesttech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you.Translations are available.
Updates for editors
To optimize how user data is stored in our databases, the saved preferences of users who haven't logged in for over five years and have fewer than 100 edits will be cleared. When those users return, default settings will apply.[1]
View all 20 community-submitted tasks that wereresolved last week. For example, there was a broken link from the GlobalContributions interface message to the XTools GlobalContributions page which has now been fixed.[2]
Updates for technical contributors
The work to reroute all traffic to API endpoints under therest.php route through a common API gateway is now complete. If any issues are observed, please file a phabricator ticket to theService Ops team board.
Edits to Wikidata references or qualifiers will now be shown in RecentChanges and Watchlist entries on other wikis less often, reducing unnecessary notifications. This will reduce the overall quantity of 'noisy' entries. Wikidata's own pages remain unchanged.[3]
What doesusers who haven't logged in for over five years mean, exactly? WithWP:SUL, when I log in here, and I visit another WMF wiki, I'm also logged in there. But in order to preserve my settings on other WMF wikis, do I actually need to (a) log in as a separate act on each WMF wiki in order to generate a fresh login cookie; (b) log in on en.wp and visit each WMF wiki in order to send an existing login cookie; (c) log in on en.wp and ignore the other WMF wikis? --Redrose64 🌹 (talk)21:13, 21 October 2025 (UTC)[reply]
It's not very clear, since the person responsible seems to be making several unwarranted assumptions as to how things actually work. But it does seem like if whatever they're looking at doesn't indicate activity on the specific wiki,they're intending to delete.Anomie⚔22:29, 21 October 2025 (UTC)[reply]
It appears from thiscomment on phabricator that the decision will be made wiki by wiki, so if there's a WMF wiki that you've not logged in to for five years then your settings there will get cleared. (Though global settings will still apply, of course.)DLynch (WMF) (talk)00:21, 22 October 2025 (UTC)[reply]
@DLynch (WMF):if there's a WMF wiki that you've not logged in to for five years - this is itself unclear. Please see my original question: is logging in on en.wp sufficient, or do I need to explicitly log in on each WMF wiki as a separate action? --Redrose64 🌹 (talk)17:42, 22 October 2025 (UTC)[reply]
So it's (a) log in as a separate act on each WMF wiki in order to generate a fresh login cookie. This seems to be requiring a great deal from users that want their settings to be preserved. --Redrose64 🌹 (talk)20:04, 24 October 2025 (UTC)[reply]
I'm Swedish, which means I'm obsessed with comparing anything Swedish to other countries, especially other Scandinavian countries. Recently I started to look into who are the translators of Swedish literature into Ukrainian. The articleSwedish literature on Ukrainian Wikipedia (uk:Шведська література, 72 kbyte) is very long and detailed, so I immediately compared it to the corresponding articles for Norwegian and Danish literature. The Norwegian (uk:Норвезька література, 108 kbyte) is also detailed, but the Danish (uk:Данська література, 3.7 kbyte) is a joke. It turns out, Swedish Wikipedia's article about Danish literature (sv:Dansk litteratur, 2.7 kbyte) is also really short and poor. Here's where I want to create a matrix or table with columns for Swedish, Norwegian, Danish, etc. literature, and rows for the articles in Ukrainian, Swedish, English, German, etc. Wikipedia. Is there already some tool that will generate this table for me? You'd guess thatWP:WikiProject Literature has done something like this, and that they would already have designed or requested a tool for it? But maybe not. But the need for such a tool is not limited to literature. Maybe friends of sports or music have already done this? Could Wikidata help me?LA2 (talk)20:46, 20 October 2025 (UTC)[reply]
Unfortunately, it does not seem to have worked, and I can't findinnerstyle anywhere inWikipedia:Extended_image_syntax. However, 'style="width:auto;background:black"'is in the HTML produced, and apparently in the right place. Even adding!important to it does not achieve the desired effect. This is presumably a CSS issue. —The Anome (talk)17:58, 21 October 2025 (UTC)[reply]
It looks like it has some sort of transparency, if opened in the media viewer. This image seems to be jinxed in some way: I will try uploading it in a different format and see if it's better. —The Anome (talk)18:17, 21 October 2025 (UTC)[reply]
I re-uploaded it as a PNG and all is good. I think support for transparency in TIFF is not uniformly supported across browsers; PNG works fine for me. —The Anome (talk)18:26, 21 October 2025 (UTC)[reply]
Looking at their website, I don't think there should really be transparency at all. It seems their design is against a dark blue background. It's just transparent in the header, so that they can also use it atop of a gradient of blue and with opacity on top of the scroll element.. I'm guessing they don't have an official design page published detailing things like this, but that puts us a bit in a bind, as that makes us have to guess how they want others to use their logo. (maybe they want the text flipped to black when it is shown on a white background, we can't really tell). —TheDJ (talk •contribs)18:28, 21 October 2025 (UTC)[reply]
You can't with TemplateStyles. This is intentional, we want the page title to remain generally visible. You could hide it for yourself only by adding appropriate CSS toUser:Harringstars/common.css, but everyone else would still see it.Anomie⚔17:14, 21 October 2025 (UTC)[reply]
@PrimeHunter I'm kind of surprised that's allowed by the software. I know it's been filtering out "display", "user-select", and "visibility" for the past decade or so, but there are plenty ofWP:BEANS things you can do with the remaining allowed CSS.--Ahecht (TALK PAGE)14:55, 22 October 2025 (UTC)[reply]
Silly me, turns out they're grayed out if they're in source mode, since they only make sense in visual mode. I've been trying to figure out how the background between ref tags turns green..A diehard editor (talk |edits)02:52, 22 October 2025 (UTC)[reply]
Oh yes, I knew it was syntax highlighting, what I meant was that I was trying to figure out how I could get green ref tag backgrounds on my own wiki. Turns out I'll just have to wait for the next MW release. I created a common.css there which works for now.
Has some setting changed? Or is there something in preferences that I’ve accidentally switched or something? Or is this just a problem for me? Am on iPhone 14 latest iOS version using Chrome.FOARP (talk)16:51, 22 October 2025 (UTC)[reply]
Yeah, I can force it to serve the mobile version by adding the “m” to the URL but otherwise I get the desktop version by default. My browser settings are set to preferring the mobile version but I’m not getting that.FOARP (talk)17:41, 23 October 2025 (UTC)[reply]
Typing this on my ipad as always logged in but on my PC it wants me to enter a verification code to log in. Has now sent me near a dozen codes by email, none of which work. Can't get into my account on my PC now, where I do most of my article work. ♦Dr. Blofeld18:04, 22 October 2025 (UTC)[reply]
Gawk! That's awful. Don't know any other way to gain access. Ergo, recommend: email the WMF per Xaosflux and include link to this conversation. --Rosiestep (talk)18:40, 22 October 2025 (UTC)[reply]
Go to the login page, wait for the verification code form to show up, wait for the verification email and don't do anything else until it arrives, then enter the code.
The verification logic is not very sophisticated so e.g. if you restart login and get a second code, the first code won't work anymore. I suspect what happens is that the email takes a long time to arrive (might be an issue on our side or with the email provider) and people restart login, refresh the page, try logging in in multiple windows etc, and then they receive multiple codes (one for each time they loaded the form) most of which aren't valid anymore.
Alternatively, if you make sure to use the same IP address on your PC that you used on your iPad (probably it has some sort of hotspot option and the PC can connect through that?), that should skip the email verification.
Seems to be OK now. Yes it was odd, it seemed that every time I tried to enter a code it would email me another one! Thanks all! ♦Dr. Blofeld13:26, 23 October 2025 (UTC)[reply]
Having edited on mobile using the desktop site for a couple of years I would say the exact opposite. The mobile site is barely usable on any device. Is the header small on Chrome or just Firefox? --LCUActivelyDisinterested«@» °∆t°18:55, 23 October 2025 (UTC)[reply]
Which browser version of Firefox mobile are you using? Which skin (I believe this might be Vector but I'm not 100%)? Do you get the same problem insafe mode? If not, which gadgets do you have enabled? Could you also share what your Firefox accessibility settings are? (those arE: Automatic font sizing, font-size, zoom on all websites and available via the settings).
The firstroot parameter is normally meant to point to the page whereUser:MiszaBot/config is located. Since RFPP uses a different bot,Cyberbot I, which is hardcoded to run on this page and does not use a config, the first root parameter should be removed. This parameter usually provides a readable summary of the archive configuration, like"This page has archives. Sections older than 20 days may be auto-archived by Lowercase Sigmabot III if there are more than 5". In this case, using it will not work and will cause an error. –DreamRimmer■14:52, 23 October 2025 (UTC)[reply]
Apologies, this one's my fault - was recently updating a module decently deep into the chain of what this template uses, and I forgot to consider the edge case of non-existant pages in my testing. I've already made an edit request for this, so it should go away soon. There's nothing wrong (afaik) with how the template is being used on that page.Aidan9382(talk)15:05, 23 October 2025 (UTC)[reply]
I'm pretty sure it used to be that space characters in an anchor were replaced with a plain underscore character (low line:_; U+0095). That seems to have recently changed and doesn't match the uri encoding:
{{anchorencode:string with spaces}} →string_with_spaces → string_with_spaces
{{urlencode:string with spaces|WIKI}} →string_with_spaces → string_with_spaces
So now the anchor encoding does not use the single character_ but instead uses the html numeric entity string_. Note that the anchor encoding does not match the url encoding.
It does link. Because the target template is a wrapper template around{{cite web}} and because the{{NHLE}} template doesn't expose the information necessary for Module:Footnotes to identify its anchor ID, Module:Footnotes emits an error message. We get round that by use of a whitelist. The whitelist links a template name with aCITEREF anchor ID. In this example case, the whitelist lists{{NHLE}} with the anchor ID lua pattern{'Historic_England%d+'}. That pattern was added toModule:Footnotes/whitelist atthis edit 1 December 2024. The whitelist pattern expects a simple underscore in place of the space character in 'Historic England'.
So, was the change to use the html numeric entity string_ intentional? Is it permanent? Why was it done?
I uploaded a new image and immediately sent it toWikipedia:Graphics Lab/Illustration workshop because it contains some problems. And for this reason, I don't want to use it in the article yet. Now I'm waiting for the fixed version so that I can use it in the article and fill in the file parameters with{{Non-free use rationale logo}}, however, a bot has already tagged that image with{{untagged}} - this means that it should be deleted after seven days. Can you advise me what to do in such a situation? Thanks,Maiō T. (talk)19:23, 23 October 2025 (UTC)[reply]
@Maiō T. Looks like this particular issue was resolved, but in the future when posting to the illustration workshop you can just link to where the file exists on the web instead of uploading it first.--Ahecht (TALK PAGE)19:12, 24 October 2025 (UTC)[reply]
I'm curious.'' and''' used to output<em> and<strong>, respectively, but somewhere in December 2004, that output was changed to<i> and<b> (for example, look at "Browse Wikipedia" text in the page's source codehere (20 December 2004) andhere (28 December 2004)). Why?sapphaline (talk)19:52, 23 October 2025 (UTC)[reply]
When I just tried to check mywatchlist, I found it blank; this has never happened before. There's just a notice saying, "No changes during the given period match these criteria". I'm using the latest version of Firefox and Windows 11, and have made no changes to my watchlist preferences. Any help appreciated. All the best,Miniapolis21:10, 23 October 2025 (UTC)[reply]
@Sarsenet Are you talking about the Watchlist itself (Special:Watchlist) or about the listing of pages on your watchlist (Special:EditWatchlist)? -- If the former, that's a new and distinct bug, please start a new Topic for it and share more details (and ideally a screenshot) or file a bug on Phabricator. -- If the latter, that might be related to the ongoing changes to fix the bugs/timeout issues for users with very large watchlists (phab:T41510) where some things have been changed for performance reasons, including no longer showing "temporarily watchlisted pages" at the top of each section.Quiddity (WMF) (talk)22:30, 23 October 2025 (UTC)[reply]
In that link, you've got the filters for both "Unseen changes"and "Seen changes" selected. Thatshouldn't have any effect as they should combine to just show all changes [i.e. cancel each other out] (I'll file a bug-report:phab:T408167), but it does seem to currently; so to temporarily (and permanently!) fix it, just unselect both those 2 filters.Quiddity (WMF) (talk)21:56, 23 October 2025 (UTC)[reply]
Too good to be true, apparently; my watchlist was blank again today until I deselected "unseen changes" (that and "seen changes" were back). Unseen changes have been in bold, which was useful. All the best,Miniapolis19:33, 24 October 2025 (UTC)[reply]
(←)Thanks,Quiddity (WMF), but I saved "seen changes" as a default filter after removing "unseen changes"; when I left the page and returned, unseen changes was back and the watchlist was blank. When I removed it, my changes returned as seen; I miss the bold unseen changes, but it's better than a blank watchlist. I've been here a while, and this used to be a no-brainer; frankly, I'm surprised it hasn't been reported by others. Thanks for your help and all the best,Miniapolis22:40, 24 October 2025 (UTC)[reply]
I recently created templateUser:Mathglot/sandbox/Templates/Uh-page-numbers, and added substability. The template still works fine unsubsted, but substed, it emits an error similar to this:
with the blue 'help' link targeting local section #Errors on the page where it is substed. See a substed example atUser:Mathglot/sandbox5. What could be causing this? Note: this error also appears onthese 121 pages. I suspect I added safesubst protection incorrectly, but I don't see it. Thanks,Mathglot (talk)07:34, 25 October 2025 (UTC)[reply]
For the last 72 hours or so, when I get notified about talk page message there's a block of solid colour over the message with an information icons and a "reply…" tag.I have to click this to see the text. The "i" icon takes me to the image page… which I suppose is a type of information about the icon.
This may, of course, have something to do with my settings. Even so it's a change in behaviour, so we should be able to track down the proximate cause. All the best:RichFarmbrough21:09, 25 October 2025 (UTC).[reply]
Two galleries in an article make images in the second one unclickable
So inLe_Touquet#The_peak_years_(1902–1940) there are two galleries created using an HTML-esquegallery tag. The first one works fine, but on desktop (not tested on mobile) the subsequent ones are not clickable and the images are skipped when you go through preview of all images. The behaviour is replicated across major browsers. All tags are appropriately closed.ETA: The twelfth image in the first gallery also doesn't work for whatever reason.
It's not because there's two galleries. It's because all the images in the second gallery are specifically set unclickable, by including|link=|.Fixed. —Cryptic21:47, 25 October 2025 (UTC)[reply]
