Policy Technical Proposals Idea lab WMF Miscellaneous 
The technical section of the village pump is used to discuss technical issues about Wikipedia. Bug reports and feature requests should be made in Phabricator (see how to report a bug). Bugs with security implications should be reported differently (see how to report security bugs).

If you want to report a JavaScript error, please follow this guideline. Questions about MediaWiki in general should be posted at the MediaWiki support desk. Discussions are automatically archived after remaining inactive for five days.

Locator map showing Null Island not Western Australia

Henry Street, Fremantle shows a locator map consisting entirely of water. Zooming out it appears to be displaying the location of Null Island rather than Western Australia. There are no coordinates or explicit map call on the page that I can see. The article does transclude Template:Infobox Australian road, the documentation for which says that the locator map is automatic if the Wikidata item includes coordinates. d:Q16342680 does indeed contain coordinates (32°3'19.19"S, 115°44'39.06"E) but keen observers will spot that those are not the coordinates of Null Island. Thryduulf (talk) 22:30, 27 February 2025 (UTC)[reply]

I don't know what exactly causes it, but I figured out that it is possible to reproduce it just by {{#invoke:Infobox mapframe|auto|onByDefault=yes}}. I also found this displays the map well at Midland Highway (Tasmania), just like the infobox in that article. Janhrach (talk) 19:37, 3 March 2025 (UTC)[reply]
In User:Janhrach/sandbox5, I have a minimal example demonstrating the difference. Maybe the bug is caused by something in OSM. Here are the OSM links: [1][2]. Notice that the former link (not working) is linked from Wikidata, while the latter (working) is not, so I doubt this is caused directly by Wikidata. Janhrach (talk) 20:05, 3 March 2025 (UTC)[reply]
Maybe the error is caused by [3] being connected to WD instead of [4]? Janhrach (talk) 20:13, 3 March 2025 (UTC)[reply]
It does not. I put a wide varity of buggy maps into my sandbox. Janhrach (talk) 15:50, 4 March 2025 (UTC)[reply]
I have filed a report on Phabricator. Janhrach (talk) 16:16, 4 March 2025 (UTC)[reply]
Thanks for trying to figure this out. Thryduulf (talk) 17:11, 4 March 2025 (UTC)[reply]
I've seen this issue when the OSM listing doesn't contain both a Wikidata and a Wikipedia tag. I've added a Wikipedia tag to Henry Street, but it may take a few days for the OSM changes to propagate. --Ahecht (TALK
PAGE
)
18:50, 5 March 2025 (UTC)[reply]
I don't think so, e.g. Gumpendorfer Straße (third map in my sandbox) only has a dewiki link and works, while Gagarinova (penultimate map) has a skwiki link, yet it doesn't work. Janhrach (talk) 19:09, 5 March 2025 (UTC)[reply]

Hi can somebody use a tool to find how many stubs we have in total for Europe? ♦ Dr. Blofeld 15:40, 2 March 2025 (UTC)[reply]

A quick Petscan query shows over 760,000 articles. – DreamRimmer (talk) 18:28, 2 March 2025 (UTC)[reply]
Yikes!! Thanks! ♦ Dr. Blofeld 19:21, 6 March 2025 (UTC)[reply]

Archival of url and source update script

Hello, do you have some python or some other language code to archive an url and add it as parameter to a source template? I'd like to do it on another WMF wiki, wouldn't want to re-invent the wheel. Thank you in advance. Gryllida (talk, e-mail) 01:36, 4 March 2025 (UTC)[reply]

There's InternetArchiveBot. Graham87 (talk) 01:49, 4 March 2025 (UTC)[reply]
Interested in something I can run from my account Gryllida (talk, e-mail) 22:55, 4 March 2025 (UTC)[reply]
I.e. using a JavaScript gadget Gryllida (talk, e-mail) 22:55, 4 March 2025 (UTC)[reply]
There is no Javascript gadget for that. There are install instructions on https://github.com/internetarchive/internetarchivebot so you can run it on your own account. Even simpler still would be to ask InternetArchiveBot to run on your WMF project by asking on phab: with the tag InternetArchiveBot. You need a local consensus for that. In both cases you want to make sure your citation templates work with InternetArchive bot, and if they are CS1 templates, then make sure they are up to date, especially their support of the archive-url, archive-date and url-status parameters. You also need an dead link template and make sure InternetArchiveBot knows what that is. I have asked and gotten InternetArchivebot working on one project, so that is where this information comes from. Snævar (talk) 03:37, 6 March 2025 (UTC)[reply]
Ok, thanks, I will have a look. Let me know if gadgets becomes available. Gryllida (talk, e-mail) 07:44, 6 March 2025 (UTC)[reply]

Corrupted history at Westend Synagogue

Something is wrong with the history at Westend Synagogue. When I click history, it shows me my last three edits only, but not the previous history of creation and eight edits by User:Thriley, and two by User:TheChairmanOfTheBoard. (This from an old edit tab which I have not refreshed yet.) What is going on, here? Mathglot (talk) 09:24, 4 March 2025 (UTC)[reply]

And, the Talk page seems to be gone; previously held some wikiprojects, and one discussion. Mathglot (talk) 09:31, 4 March 2025 (UTC)[reply]
Okay, someone appears to have moved the Talk page to Draft talk, but somehow three edits of mine persist in mainspace history. How did that happen? Is some kind of main +Draft merge needed, and a histmerge? Or should I abandon mainspace, reapply the three edits to Drat, and move it back to mainspace? Mathglot (talk) 09:34, 4 March 2025 (UTC)[reply]
The page was moved to draft:Westend Synagogue without leaving a redirect, a minute before your first edit of the three in the history. So your edit was a page creation. Nthep (talk) 09:35, 4 March 2025 (UTC)[reply]
Nthep, okay, that makes sense up to a point, but there's a future Phab ticket here, somewhere. If the first was a page creation, I should have gotten an edit conflict, or some kind of indication. I was editing what to me was a valid page, and if it was yanked away before I could save, then I should have received some kind of notification of it. Mathglot (talk) 09:40, 4 March 2025 (UTC)[reply]
I admit, I'm surprised you didn't get an edit conflict notification. Must be the way the software works and page moves/deletions don't register as edits. Nthep (talk) 09:43, 4 March 2025 (UTC)[reply]
Nthep, Yeah, me too. I am familiar with Phab, but uncertain with what Phab project I should list this under. Any ideas, so it gets to the right folks? P.S., it's moved back, so the Draft should red-link now. Mathglot (talk) 09:49, 4 March 2025 (UTC)[reply]
This is phab:T25044 from 2010. PrimeHunter (talk) 12:10, 4 March 2025 (UTC)[reply]
Very helpful; thanks! I will add some thoughts there. Mathglot (talk) 20:57, 4 March 2025 (UTC)[reply]

Hello again, everybody. Has this happened to you, or might it trip you or someone else up if it did? Please add your name as an observer to Phab ticket T25044. More eyeballs are always good, and additional voices may eventually lead to a resolution. Thanks! Mathglot (talk) 21:21, 4 March 2025 (UTC)[reply]

Interesting. The steps to reproduce the issue is essentially a feature of suppressredirect in any other cases where the third step is something else, i.e. pageswap or creating disambiguation page at the now unoccupied title. The only way to see this however is that it is a form of edit conflict, and add this scenario into the edit conflict resolution process. – robertsky (talk) 12:01, 6 March 2025 (UTC)[reply]

References at Alpine marmot partially behind images

Does that happen for anyone else? Nobody (talk) 14:47, 4 March 2025 (UTC)[reply]

The images displace the reference columns on my screen, but they don't overlap. CMD (talk) 15:13, 4 March 2025 (UTC)[reply]
This image shows how I see it. Nobody (talk) 15:18, 4 March 2025 (UTC)[reply]
For the specific case of Alpine marmot,  Works for me. But I've seen it elsewhere; and found that going into the "Inspect" feature of Firefox would fix it immediately; and when closing "Inspect" again, would stay fixed. I had suspected a z-index: issue, in combination with floating objects. But being transient makes it hard to nail down. --Redrose64 🌹 (talk) 17:13, 4 March 2025 (UTC)[reply]
I can reproduce it. It reminds me of a Firefox bug I reported a while ago: T294980#7483288, Mozilla bug 1739506 – but it's hard to tell whether it's the same problem. You might want to report your case too to give their devs a poke. Matma Rex talk 20:10, 4 March 2025 (UTC)[reply]

Tools for bundled XfD noms

Are there any tools to assist creation of bundled XfDs nominations?

I understand that there are two kinds of bundled noms: ones created under the title of one of the pages, with other articles then being added to the nom using {{la}} and tagged (this is recommended by Wikipedia:Articles for deletion#How to nominate multiple related pages for deletion), and noms that are not named after any of the nominated pages. For clarity, I am looking for tools for assisting any of these two kinds of nominations. Janhrach (talk) 17:10, 5 March 2025 (UTC)[reply]

There is User:Qwerfjkl/scripts/massXFD but it only supports CFDs and RFDs so far. CX Zoom[he/him] (let's talk • {C•X}) 18:19, 5 March 2025 (UTC)[reply]

There's a template about talk page behavior & I cannot remember what it is called...

I can remember what it does... Anyway, isn't there a template that we can put on a celebrity's or politician's article talk page in reply to when someone comes along and wants to get in touch or maybe they ask the politician what can they do about Social Security (as if the article is a way to ask the celebrity or politician a question, etc.) Thanks - Shearonink (talk) 22:54, 5 March 2025 (UTC)[reply]

{{Astray}}? * Pppery * it has begun... 23:10, 5 March 2025 (UTC)[reply]
I think the closest thing for article talk pages is {{not a forum}}. Astray is for user talk pages. Izno (talk) 23:24, 5 March 2025 (UTC)[reply]
Astray would almost work...I'm surprised there isn't something written up, that could be used on the article talk to reply since this poster is an anon on a politician's page. Oh well, thanks anyway. I just thought I couldn't find it but yay it doesn't exist... - Shearonink (talk) 01:44, 6 March 2025 (UTC)[reply]
Somewhere there's a template, intended to be WP:SUBSTed, that reads (or used to read) something along these lines:
I suspect, based on your question, that you found one of our over {{subst:#expr:(floor ({{subst:formatnum:{{subst:NUMBEROFARTICLES}}|R}} / 100000)) / 10}} million articles and thought we were affiliated in some way with that subject. Please note that you are at [[Wikipedia:About|Wikipedia]], the free online encyclopedia that [[Wikipedia:Introduction|anyone can edit]], and this page is for [[WP:TPG|discussion on how to improve]] the [[{{subst:PAGENAME}}]] article.
Maybe not exactly that, but maybe somebody knows how to do a proper search, using that as a clue. It was in use sometime around 2010. --Redrose64 🌹 (talk) 17:54, 6 March 2025 (UTC)[reply]
That would be {{Astray}}, which I pointed out above. * Pppery * it has begun... 18:16, 6 March 2025 (UTC)[reply]
I realise that, but the one that I've seen is about as long as my example above. Significantly, it lacked the second half (from Thus, we have no special knowledge onward). --Redrose64 🌹 (talk) 18:25, 6 March 2025 (UTC)[reply]
You mean this and this, Redrose64? Nobody (talk) 13:10, 7 March 2025 (UTC)[reply]

Citation generator is broken for Yonhap News

The automatic citation generator included in VisualEditor seems to be struggling with the Yonhap News website. It have been throwing an error for a long time. Since Yonhap News is the biggest news agency in Korea, it would be nice to get this behaviour fixed. Where or who can I contact to report this issue? Ca talk to me! 05:13, 6 March 2025 (UTC)[reply]

@Ca The automatic citation generator uses mw:citoid, which in turn is based on an open source piece of software called zotero. To add support for a website you would need to write a "translator", which tells zotero how to extract data from the page, instructions on how to do this are at mw:Citoid/Creating Zotero translators. 86.23.109.101 (talk) 08:42, 6 March 2025 (UTC)[reply]
It seems like Yonhap News has blocked WMF servers for making too many requests. If I try to use zotero without WMF servers then it works without an error. Citoid is supposed to use archive.org in that instance, so you get the information from there. See phab:T374831 which requires you to have an citation template parameter for archive url, archive date and specify that to citoid as explained in the link. See also phab:T362379. Snævar (talk) 11:07, 6 March 2025 (UTC)[reply]

Display of contributions and "Current" label

When displaying contributions, a bold "Current" tag is added to the end of the edit when it is the latest edit to that article, talk page, etc. Is there a method or gadget to move it to a point where it is in a uniform column? I do a lot of wp:RCP and editor will often repeat their edits without correcting the problem. I will view the "Current" tag in my list of contributions to see if the article has been changed. Trying to determine if articles have been edited or not is rather cumbersome with just the list as the "current" tag is not in a uniform vertical column. I could add it to my watch list with a short expiration. I usually do a text search on my contributions page for "current" which highlights it. Highlighting makes it easier, but I'm looking for a better method. Example of listing in my contributions:

  • 19:08, 6 March 2025 (diff hist) (−6) Iran men's national basketball team (Undid revision 1279103344 by 217.218.49.102 (talk) unexplained change of header) (current) (rollback: 1 edit) (Tag: Undo)

What I would like to see:

  • 19:08, 6 March 2025 (diff hist) (C) (−6) Iran men's national basketball team (Undid revision 1279103344 by 217.218.49.102 (talk) unexplained change of header) (rollback: 1 edit) (Tag: Undo) --OR--
  • (C) 19:08, 6 March 2025 (diff hist) (−6) Iran men's national basketball team (Undid revision 1279103344 by 217.218.49.102 (talk) unexplained change of header) (rollback: 1 edit) (Tag: Undo)

Thank you, Adakiko (talk) 19:32, 6 March 2025 (UTC)[reply]

@Adakiko: "Search for contributions" on a contributions page has the option "Only show edits that are latest revisions". User:Markhurd/hidetopcontrib can do the opposite. Does that work for your purpose? PrimeHunter (talk) 20:35, 6 March 2025 (UTC)[reply]
@Adakiko: Here are two CSS rules:
/* For current rev in contribs, add "(C)" before the size change */
li.mw-contributions-current span.mw-changeslist-links+span.mw-changeslist-separator:after {
  content: "(C)";
  font-weight: bold;
}
/* For current rev in contribs, suppress the "(current)" marker */
li.mw-contributions-current span.mw-uctop {
  display: none;
}
The second rule is optional. They go in Special:MyPage/common.css, since it works in all skins. But personally I use a rule that I wrote some months ago:
/* highlight current edit in contribs */
li.mw-contributions-current {
  background-color: #eeffcc;
}
I created it at somebody's request, don't recall who for. --Redrose64 🌹 (talk) 21:07, 6 March 2025 (UTC)[reply]
@Redrose64: Thank you. added it to my CSS and it works great! Now all I have to do is to look for the "C" on the left. I go into shock each time I look at my contributions page and don't see "current". Cheers Adakiko (talk) 19:10, 8 March 2025 (UTC)[reply]
@Adakiko: As I mentioned, the second rule is optional - you can have both "(C)" and "(current)" displayed, by just using the first five lines, down to (and including) the first closing brace }. --Redrose64 🌹 (talk) 20:03, 8 March 2025 (UTC)[reply]
@Redrose64: I prefer the lack of "Current". I need to retrain myself. Can I train an old dog a new trick? Thanks again! 20:05, 8 March 2025 (UTC)[reply]

Parsoid error

I've been trying to edit my sandbox page in the visual editor, but when I click 'submit' it returns 'parsoid error'. Please could someone tell me what to do with this. I looked it up and none of the questions posted about this address my issue but instead go into a lot of detail about other kinds of error which are very technical. I can't paste an image here but the error essentially says: "Something went wrong [!] Parsoid error" and gives the options to 'dismiss' or 'try again'.

This error is occurring across multiple devices when I try to edit my sandbox. When I try to switch to the source editor it returns simply 'Parsoid error' and this time just gives me the option of 'OK'.

I don't know whether this is linked, but I recently had another (very minor) technical concern which I researched, and was told to create a common.css page with my username, in which I added the following: .mw-parser-output span.cs1-maint {display: inline;} /* display Citation Style 1 maintenance messages */. I don't know whether it had anything to do with this and I doubt it does, since I did this yesterday and was editing throughout the day yesterday with no issues even after pasting that in, but it might be useful context. Krimzonmania7078 (talk) 21:51, 6 March 2025 (UTC)[reply]

If it helps, I'll explain what I'm trying to do in my edit session. I'm working on a new article and decided to copy and paste another article into my sandbox to use as a 'template' of sorts (which ended up being a pointless exercise because I've basically written this article from scratch anyway). I think the issue might stem from that decision somehow. The reason why I'd like to get this fixed is that now I've been editing the article in my sandbox for a while and basically have no way to save my progress Krimzonmania7078 (talk) 23:21, 6 March 2025 (UTC)[reply]
I've actually found a workaround which is good enough (copy-pasted the article into another session of the sandbox with the first sentence missing) - that seems to have fixed it. Probably something right at the top of the article I was trying to use as a 'template' which I mindlessly copied was causing it. Krimzonmania7078 (talk) 23:39, 6 March 2025 (UTC)[reply]
Just report it on phab:. Your wikipedia account will work there. Include the name of the article you where copying in your bug report. The developers can look up specifics about your edit attempt, users can not do that. Snævar (talk) 13:10, 7 March 2025 (UTC)[reply]

CAPTCHA Security check

adding a internal link on Double Dutch (jump rope) requires a CAPTCHA Security check

there are pages where adding external links does not require a CAPTCHA Security check

suprising....

69.181.17.113 (talk) 11:53, 7 March 2025 (UTC)[reply]

Double Dutch (jump rope) hasn't been edited since 18 February. Your IP address has never edited it. Please always post a diff if you report a perceived issue with an edit. PrimeHunter (talk) 22:40, 7 March 2025 (UTC)[reply]
@PrimeHunter: Well, if you do a null edit on the page they linked as an IP you get a captcha, so I'm guessing it's some issue like Wikipedia:Village pump (technical)/Archive 212#unwarranted captcha (though it doesn't appear there is a news: link in that page, even transcluded). I purged the page and it still happens so I don't think it's from a change in a template causing a new link to be added... though maybe it's still the fault of a template? What else could it be? – 2804:F1...90:4783 (::/32) (talk) 16:25, 9 March 2025 (UTC)[reply]

Modifying system message on Petscan's interwiki page

When you type :petscan: in the search bar, you are taken to Special:GoToInterwiki/petscan:. The link on that page is https://petscan.wmflabs.org/?psid= which upon clicking takes us to an error page from where it is not possible to navigate back to petscan's main page. Is it possible to modify the system message so that there is a direct link to petscan's main page: https://petscan.wmcloud.org/? CX Zoom[he/him] (let's talk • {C•X}) 14:23, 7 March 2025 (UTC)[reply]

That is probably due to the interwiki prefix on meta:Interwiki_map. That interwiki map is used on WMF wikis, not just english wikipedia.
Changing the system message from the tool itself would be done with a bugreport on the tool itself - https://github.com/magnusmanske/petscan_rs/issues Snævar (talk) 15:29, 7 March 2025 (UTC)[reply]
Yeah, and the interwiki map format doesn't support doing something different with an empty prefix. It possibly should, but that's another Phabricator ticket (which in my experience would languish forever). * Pppery * it has begun... 16:13, 7 March 2025 (UTC)[reply]
meta:Interwiki map assumes there is a page name after the colon. petscan:$1 goes to https://petscan.wmflabs.org/?psid=$1. If $1 is empty then you get the bad https://petscan.wmflabs.org/?psid=. Special:GoToInterwiki/petscan: uses MediaWiki:Gotointerwiki-external. It's possible to change it. I made a test at testwiki:MediaWiki:Gotointerwiki-external. It discovers if petscan: or acronym: were used with nothing after the colon and selects a different page, https://petscan.wmflabs.org or https://www.acronymfinder.com. It works (see testwiki:Special:GoToInterwiki/petscan:) but it may confuse users if interwiki prefixes don't behave the same way at Wikimedia wikis so I'm not sure we should do it. It's also confusing if entering petscan: in the search box doesn't lead to the same page as a wikilink petscan: which goes directly to the interwiki map target without using Special:GoToInterwiki. This also means "petscan:" and the displayed url at testwiki:Special:GoToInterwiki/petscan: have different targets. My test also has code for mail: but it's not treated as an external wiki so Special:GoToInterwiki isn't used for that. At phab:T309558 I suggested a redirect for mail: to https://lists.wikimedia.org/postorius/lists/. I also made a post saying "A more advanced solution would be adding a way for the interwiki map to specify where a link with an empty $1 should go". This would give a nice consistent solution across Wikimedia wikis with the search box and wikilinks behaving the same way. PrimeHunter (talk) 16:18, 7 March 2025 (UTC)[reply]
I think your solution solves the problem at hand, but I agree with the concerns you have with its implementation. Maybe the text could read something like: "Continue to https://petscan.wmflabs.org/?psid= or go to Petscan's homepage". It will keep the default destination intact, while also adding an additional link to the homepage for convenience. CX Zoom[he/him] (let's talk • {C•X}) 16:43, 7 March 2025 (UTC)[reply]
Actually, I'm proposing that an additional line of text be added at Special:GoToInterwiki/petscan: with a direct link to Petscan's main page (if feasible with system messages). CX Zoom[he/him] (let's talk • {C•X}) 16:21, 7 March 2025 (UTC)[reply]
Of course, changes to GoToInterwiki would only fix pages that go there, not direct links like petscan:. * Pppery * it has begun... 16:32, 7 March 2025 (UTC)[reply]

Privacy of sockpuppets and sockmasters

I have seen this in past but did not tell.

Check users are not allowed to disclose IP address of socks and sockmasters unless they do it themselves or long time offender or intentional IP socking.

But when I check administrator block log of check users, then check users often do this. First they block the large number of sockpuppets then after that sometimes they also block the IP range without mentioning the name of sockmaster in the IP block.

But those with experience can know that if the IP range is blocked immediately after blocking the sock accounts, then it belongs to the sockmaster.

I saw this when Mike V, Bbb23 used to block socks. 2409:40E1:100C:D174:78FA:B94D:2E28:A22A (talk) 07:33, 8 March 2025 (UTC)[reply]

See the policy about this which explicitly excepts that. — xaosflux Talk 10:35, 8 March 2025 (UTC)[reply]

Help:Creating a bot: Revision history - Wikipedia Revision history 13:23, 8 March 2025 Anomie talk contribs 31,536 bytes +118 The rendered HTML is exactly the same after your edits, all you did was make the wikitext uglier. Please don't do this sort of thing again, and consider asking at Wikipedia:Village pump (technical) whether your ideas will actually work


How does reducing the bytes of the document not increase processing speed? In a sample from this: Simon E Spero Analysis of HTTP Performance problems www.w3.org @ "HTTP Illustrated" "This page is 1668 bytes long" indicates bytes is the determination - @"Very nice, but what does it all mean": "Another important metric is the bandwidth of the connection. This is a measure of how many bits per second our connection can carry." this would indicate that reducing the bits decreases bandwidth load. Understanding latency developer.mozilla.org: "the HTML includes requests for multiple CSS, scripts, and media files. The greater the number and size of these requests, the greater the impact of high latency". In my imagining (this is to say I don't know exactly how the server scans the data for retrieval): if a server scans a page with gaps the motion is from the 1st byte to the last byte in sequence: like a sensor which receives the imprint at each position- although the whitespace has no textual element the scan would need to pass through the byte of the space to reach the last datum in the stored document. Onemillionthtree (talk) 17:12, 8 March 2025 (UTC) ⇐ this is from my Talk page not here[reply]

Onemillionthtree, you've just edited the wikitext. The html will remain exactly the same unless you make an edit that is not cosmetic. — Qwerfjkltalk 17:49, 8 March 2025 (UTC)[reply]
The bytes are reduced so the retrieval is easier? Each time someone requests the page the complete bytes of the page is retrieved. If less bytes less to retrieve. Onemillionthtree (talk) 18:10, 8 March 2025 (UTC)[reply]
The page is parsed (converted from wikitext to HTML) once when you save the edit, then the HTML is cached and stored for every other person who visits it. Following that the page is reparsed once every 30 days to check for any updates.
Parsing the page automatically removes extra whitespace, new lines, etc, there is absolutely no point in removing them by hand. 86.23.109.101 (talk) 18:17, 8 March 2025 (UTC)[reply]
Converting the parsed version into the editorial page version: the retrieval isn't different but the conversion rate is different - the cached version is the same in both but the product of reconverting from the cached have different bytes value; so there is a time difference in the conversion. Onemillionthtree (talk) 18:25, 8 March 2025 (UTC)[reply]
Although the store amounts are the same - the bytes sent to the requestor each time is more with whitespace; that's obvious Onemillionthtree (talk) 18:29, 8 March 2025 (UTC)[reply]
What you seem to be stating is that when the request is made the server delivers a page - in both versions the bytes value is the same - but if an editor requests the editorial screen the server delivers coding screen with whitespace extra bytes. So if the whitespace makes no difference then it would be helpful for editors to maximize whitespace on the coding version? Onemillionthtree (talk) 18:46, 8 March 2025 (UTC)[reply]
There is a technical issue (is removing wikitext whitespace useful?) and a social issue (does removing wikitext whitespace irritate people?). We could debate the former but the answer to the latter is yes. There have been numerous cases where editor A adds whitespace because it looks good, and editor B removes it because it doesn't. That is not sustainable and the community has decided that existing stuff should be left alone unless there is a good reason to change it. Saving inconsequential amounts of time or storage is not a good reason. Please stick to the established procedure of not changing existing style (whether it be spelling, date formats, reference templates, or whitespace). Johnuniq (talk) 01:49, 9 March 2025 (UTC)[reply]
@Onemillionthtree: If you remove blank lines above a section heading (as you did here), then the very next time that somebody edits the section preceding that heading, the MediaWiki software will automatically restore that blank line. Secondly, some editors with less-than-perfect eyesight have previously stated that some of these gaps - particularly blank lines - have been shown to aid the editing experience, and so their removal can create an accessibility issue. Third, removing 118 bytes from the wikitext of a page that previously had 31,536 bytes is a reduction of about 0.37% - a miniscule saving. Fourth, WP:DWAP. --Redrose64 🌹 (talk) 08:25, 9 March 2025 (UTC)[reply]

Removing copyrighted text

This probably isn't the right place to ask this, but I User:141.154.49.21 has been cutting-and-pasting huge pieces of text into articles. I ran two articles through "Earwig's Copyvio Detector" and it was a complete cut-and-paste.

I'm not sure how to "rev del" the additions. A bit of help would be great. Thank you! Magnolia677 (talk) 11:28, 9 March 2025 (UTC)[reply]

You can always contact an admin at Category:Wikipedia administrators willing to handle RevisionDelete requests. I have started revdeling those edits, including one you hadn't caught, yet. I've also dropped a warning on the IP's talk page. - Donald Albury 13:38, 9 March 2025 (UTC)[reply]
Thank you! Magnolia677 (talk) 14:06, 9 March 2025 (UTC)[reply]
I've revdeled copyvios in 6 articles. I don't see any other actionable copyvios. Donald Albury 14:09, 9 March 2025 (UTC)[reply]
No tags for this post.