Wikipedia:Edit filter noticeboard/Archive 12

Filter 803
From this recent false positive report, would the button to report false positives be needed? Most of the logs I can see are the filter working properly. Deauthorized. (talk) 14:30, 2 August 2023 (UTC)


 * From my experience, most of these reports come from IP editors who have forgotten that they've logged out of their accounts. We should therefore first instruct users to make sure they are logged in before filing a report. Partofthemachine (talk) 23:26, 2 August 2023 (UTC)
 * I don’t think the button is needed. Non-autoconfirmed editors shouldn’t be able to edit others userpages. End of story. We definitely still need to keep the disallow message, but let’s lose the button. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 23:57, 4 August 2023 (UTC)
 * I never suggested removing the message, just the button. Deauthorized. (talk) 18:19, 5 August 2023 (UTC)

possible MOS:ETHNICITY violation
What does this filter do? I have a vague idea of what it does but I'd like to know first. If it's what I think I'd like to propose an addition. Super Dromaeosaurus (talk) 10:10, 9 August 2023 (UTC)


 * It's coming from Special:AbuseFilter/982, which is related to this discussion. Its scope is intentionally fairly narrow. You are able to examine it to see its full logic, as it is public, if you'd like to suggest improvements. — Red-tailed hawk  (nest) 17:13, 11 August 2023 (UTC)
 * CC:, who's the maintainer of that filter. — Red-tailed hawk  (nest) 17:16, 11 August 2023 (UTC)
 * Is it really that narrow? The reason why I wanted to propose a change is because info regarding the Aromanian (a Balkan ethnic group living mostly in Greece, Albania and North Macedonia) ethnicity of some biographies is often deleted. Regarding Albanian biographies I have all of these examples:      , they're from different people I'd say, I already went to ANI and the help desk over this issue. This is normal among Balkan biographies however the Aromanians being the only stateless and disorganized major Balkan group (this being why you haven't heard of them before) there's nobody to revert and watch over these pages. This has also happened in biographies more associated with Greece             (all of these are actually the same page and the same guy, just different accounts and also different times)  (look at the history and scroll a bit down, exact same situation)  (again the same).
 * I'm worried because I've put effort into increasing info about them and their visibility and I don't want random IPs to undo my work throughout the years. Could a filter be added for removals of "Aromanian" and "Aromanians" or something like that? Maybe by any user that isn't extended confirmed? Just some ideas, I don't know what is usually applied in cases like these. Super Dromaeosaurus (talk) 17:38, 11 August 2023 (UTC)
 * What you're describing is definitely something very different - you can make a request at WP:EFR for creating such a filter. Galobtter (talk) 05:44, 12 August 2023 (UTC)
 * I did, thanks for the help. Super Dromaeosaurus (talk) 07:57, 12 August 2023 (UTC)

Filters 887 and 890
Following a reply to Suffusion of Yellow's comment on an earlier discussion, I decided to try my hand at rewriting the responses for these filters in a way which is more cordial and helpful. You may view them here. I chose to omit the "automated filter" bit of the usual message as - in my opinion - it seemed a bit too technical for new users, though you may edit my sandbox page and re-implement that if it's necessary. I also added a button which links to the WP:ACC website as it's the correct avenue for this rather than EFFP.

Feel free to make any changes that you wish, though I'd like to propose that the current system messages be updated if there's consensus here. I appreciate any feedback. Deauthorized. (talk) 02:49, 30 July 2023 (UTC)


 * I think there are definitely efforts we could make to make the messages more understandable. Whether this is the one, I’m not sure yet. I do like the possibility of linking to ACC, but to avoid creating a huge backlog for them, I’d remove the button and replace it with a simple link, after telling them to try a less repetitive or random username. We don’t want to discourage users from the very beginning! - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 12:11, 30 July 2023 (UTC)
 * Removed the button and added a wikilink. I also made the wikilink go to the ACC Wizard rather than directly to the form to perhaps give a chance for alternatives (like using a different username as the message says) before going to ACC. As per discouraging new users, it seems rather inevitable with a filter like this as it prevents you from creating an account in the first place, so my idea was to try and make the message less BITEy than the current one to hopefully alleviate that. Appreciate the response. Deauthorized. (talk) 14:29, 30 July 2023 (UTC)


 * The revised version as described above looks good to me. +1 for implementation. Taking Out The Trash (talk) 16:25, 30 July 2023 (UTC)
 * @Deauthorized @Taking Out The Trash I’ve made another change to sway users to create an account and to only encourage ACC use if they believe their username isn’t repetitive or random. Thoughts on that? Always a pleasure to chat with you on Discord @Deauthorized. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 22:54, 30 July 2023 (UTC)
 * I would prefer no link to WP:ACC. That "Wizard" contains lots and lots of text to read. This isn't the IRS and they don't have do deal with us if they don't want to; when their eyes glaze over, they'll just to go to another site. Instead, maybe remind people that usernames aren't permanent, and they can always pick a new name later. Compare MediaWiki:Abusefilter-disallowed-LTA-username. Otherwise, the new messages look good; thanks for getting the ball rolling. Suffusion of Yellow (talk) 00:32, 31 July 2023 (UTC)
 * I respect your opinion on ACC, but I think your view is definitely a minority. An ACC member reported about that the maximum wait is about 12 hours for acc. The message, after my modifications, asks you to try a different username first and to only use ACC if they believe the filter stop was incorrect. ACC exists a lot for this reason, so we might as well let them carry out their purpose. I welcome other thoughts and can definitely be persuaded here. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 00:52, 31 July 2023 (UTC)
 * Suffusion does sorta make sense. No matter the wait time for the ACC process, Special:CreateAccount will always be faster. As per which human-based process is faster (ACC or account rename requests), I'm not sure, but the account rename process seems like it'd be a better choice as you'd be able to get your proverbial foot in the door editing-wise before having the username you want. It seems like Changing username also has a bunch of words to read though, but so does Wikipedia as a whole I suppose.
 * Anyways, I made some changes to reflect Suffusion of Yellow's suggestion. Let me know what you think. Deauthorized. (talk) 14:01, 1 August 2023 (UTC)
 * I’m fine with the addition of possibly changing usernames later, but I’d prefer if the note about requesting an account wasn’t removed. We might as well let users choose which option they prefer. Also, there’s a weird line spacing with the new change as well. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 14:05, 1 August 2023 (UTC)

Removed the line spacing. Having both options might be a bit confusing for people trying to register, unless you can fit both in there without drawing it out too much. Deauthorized. (talk) 14:43, 1 August 2023 (UTC)


 * That’s a good point. Having both might be confusing/discouraging. I still think that the whole renaming process is much more confusing and complex then requesting an account. But if others think elsewise, I guess we can just not use ACC. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 14:49, 1 August 2023 (UTC)

Consensus?
This conversation has been inactive for a little while, and it seems like there's a rift between including a link to ACC or informing users that they can choose a new name at any time. Other solutions would be welcome as well, but I want to come to an agreement on an implementation. Deauthorized. (talk) 17:46, 6 August 2023 (UTC)


 * I am happy with your latest update to the messages. Suffusion of Yellow (talk) 21:37, 6 August 2023 (UTC)
 * Glad to hear that. I'll probably put it into an edit request later today. Deauthorized. (talk) 17:55, 10 August 2023 (UTC)
 * ...and I completely forgot to do so. Will do right now. Deauthorized. (talk) 04:44, 15 August 2023 (UTC)
 * Looks like it's been done. Appreciate the help, friends. Deauthorized. (talk) 09:04, 15 August 2023 (UTC)

(private)
no real hits since 2022. Should we disable? 0x Deadbeef →∞ (talk to me) 13:23, 18 August 2023 (UTC)


 * Seems reasonable to me. No need to add to the condition count if we're hitting less than once per year. — Red-tailed hawk  (nest) 19:33, 18 August 2023 (UTC)
 * Disabled. Feel free to revert and discuss here if anyone has objections. 0x Deadbeef →∞ (talk to me) 07:09, 20 August 2023 (UTC)

Impostor/attack accounts
The string "Harushiga" (User:Harushiga) should be added to the username blacklist as this user is frequently impersonated by trolls. Partofthemachine (talk) 00:44, 16 August 2023 (UTC)


 * You meant the socks of AudiGuy-1204? ToadetteEdit  (chat)/(logs) 18:30, 17 August 2023 (UTC)
 * Typically, per WP:DENY, it's correct to refer vaguely to "trolls" and "LTAs" instead of naming names if at all possible. casualdejekyll  21:19, 19 August 2023 (UTC)
 * Primarily, but they are sometimes impersonated by other LTAs as well. Partofthemachine (talk) 02:33, 21 August 2023 (UTC)

Removing button from MediaWiki:abusefilter-warning-empty-edit-request
Having a blue button will just direct people to create a false positive report without reading through the actual message. People are less likely to make false positive reports if we only have the link but not the button. 0x Deadbeef →∞ (talk to me) 04:58, 27 August 2023 (UTC)


 * It seems reasonable to remove the big blue button; we want people to fully read that one, and it probably actually makes it less likely that we'll ever see the content of that edit request. — Red-tailed hawk  (nest) 02:42, 28 August 2023 (UTC)
 * Agreed. casualdejekyll  23:55, 28 August 2023 (UTC)

FilterDebugger
For a while now, I've building a tool to replace Special:AbuseFilter/test with something more useful. It's been lurking on my local machine, slowly congealing more features as they occur to me, and it feels a bit selfish to leave it there. It needs some work still, but I'm curious as what y'all think of think of what I've done so far. Features:


 * See which part of edit matched the regex
 * See which part of the regex matched the edit
 * Immediate response as you type; no need to submit anything to a server and wait
 * Test any batch size that will fit in memory
 * Test from recent changes, older revisions, abuse logs, or a local file
 * Does not require EFH/EFM access; all the tests are run client-side, so anyone can use it

Of course, there are going to be lots of bugs. And I can't replicate every feature of PCRE with JavaScript regexes. But for most filters, it should produce reasonably accurate results.

Anyway, if you are on a modern, up-to-date browser, try installing User:Suffusion of Yellow/FilterDebugger.js, and navigating to Special:BlankPage/FilterDebug/filter/614. Please report bugs, and suggest improvements! Suffusion of Yellow (talk) 22:50, 3 August 2023 (UTC)


 * This is outstanding! Thanks so much for taking the time to improve a tool that, to be honest, has needed improvement for awhile. I’m going to install it now! - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 23:55, 4 August 2023 (UTC)
 * Update: I’m getting the error: Sorry, but your browser is too old to support FilterDebugger.
 * My browser is the latest chrome browser, so I’m not sure how it’s too old. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 23:59, 4 August 2023 (UTC)
 * Can you try again? I made that message a bit more detailed. (And what version exactly?) Suffusion of Yellow (talk) 00:45, 5 August 2023 (UTC)
 * Ooh, that looks super nice. Works for me. Galobtter (talk) 21:43, 5 August 2023 (UTC)
 * I have been using this for false positive reports and it has been working perfectly! Thanks for creating this amazing tool. 0x Deadbeef →∞ (talk to me) 02:53, 26 August 2023 (UTC)


 * Doesn't work on my iPad in either Microsoft Edge or Safari. Is this a desktop-only feature? Taking Out The Trash (talk) 18:13, 5 September 2023 (UTC)
 * Does your version of Safari support regex behind? That was only added a few months ago, and there's really no easy way to evaluate many PCRE regexes without it.  in JavaScript means an ASCII word boundary, so the  s are turned into (in the worst case)  . AFAIK Edge on iOS will use the same version of WebKit as Safari, so will have the same problem.  Adding: I briefly tested this with a a Linux WebKit-based browser and nothing was obviously broken, so it should in theory work with the latest versions of Safari. Suffusion of Yellow (talk) 22:15, 5 September 2023 (UTC)
 * Thank you for this, this is fantastic. ProcrastinatingReader (talk) 12:59, 18 September 2023 (UTC)

Should edit filter 971 have detected this?
I came across a log of filter 971 triggering and upon checking the user's contributions I noticed that they were the one who added the original image as well - the original image is also a missing file, however only the attempt to change it to a ".png" triggered the filter's warning, not the initial edit, because the filter does not check for ".webp" images.

Is adding a link to a missing ".webp" file something the filter should be detecting and warning for? – 2804:F14:80FB:2E01:9513:7B7D:14E3:DEDD (talk) 00:15, 19 September 2023 (UTC)


 * I've added the webp extension to the filter. 0x Deadbeef →∞ (talk to me) 00:28, 20 September 2023 (UTC)

Edit request for Efilter 1076
Would it possible to exclude admins from Efilter 1076? Admins generally should be trusted and don't need to be tracked when they move articles to Draft space. -- Sohom (talk) 16:15, 8 September 2023 (UTC)
 * , thanks for the message. On that ground, we might get a request to extend the argument to autopatrollers too, but it might not be prudent either way. The tracker, similar to page deletion logs, allows these actions to recorded for future history. This is just an EF tracker and does not impede any edit either way by warning or preventing. Thanks, Lourdes  04:48, 18 September 2023 (UTC)
 * Ah, good point, I hadn't thought about this from that point-of-view :) Sohom (talk) 14:39, 18 September 2023 (UTC)
 * On the other hand, that filter probably should be set to warn so people know they shouldn't do this, in which case exempting admins would make sense. * Pppery * it has begun... 23:11, 20 September 2023 (UTC)

WPWP Nigerian photo competition filter 1258
... Has been handily circumvented by Chinemeremprince, who I just blocked. Just letting people more experienced with edit filters know. Also pinging Materialscientist, who had some dealings with this user and has done things with filters. Graham 87 09:10, 27 August 2023 (UTC)


 * They aren't participating in WPWP if they don't use a hashtag. 0x Deadbeef →∞ (talk to me) 13:42, 27 August 2023 (UTC)
 * Yeah, this is intended. I do wonder exactly how much collateral there would be to disallowing "WPWP", though. (Don't think it'd be common enough to test.) casualdejekyll  00:04, 29 August 2023 (UTC)
 * They've just reached out and apparently thought they were using a hashtag, or something ... Graham87 (talk) 08:20, 21 September 2023 (UTC)

Request for helper permission:User:Seawolf35
Withdrawn. Seawolf35 (talk) 16:52, 29 September 2023 (UTC)

Hi, I would like to work on filters in the future but as of right now I don't trust myself with the ability to edit them because I would probably break something. I would like the ability to view all filters though. I would also like to get improved knowledge of filters because I would like to start working in SPI and LTA related areas. I have a good knowledge of regex in my opinion. Thanks, Seawolf35 (talk) 14:54, 29 September 2023 (UTC)
 * Oppose at this point. You've only been actively editing for a month, and it appears you don't have any experience at EFFP or EFR. This is a very high-trust permission, and I probably wouldn't support anyone with less than six months' experience. — Ingenuity (talk • contribs) 16:23, 29 September 2023 (UTC)


 * Oppose I'm not usually involved in filtering areas, but after seeing this poll, I get the impression that Seawolf35 is too inexperienced right now for this very sensitive user right. 2,000 edits and only 6 months tenure is not enough of a track record of trust to be granted the ability to view very private filters. Also the remark at the end of the poll from is not encouraging.  The Night Watch     (talk)   16:24, 29 September 2023 (UTC)

Filter 1253: False positive ratio
Filter 1253 has multiple problematic properties: At very least one of these three points needs to be fixed. I'm not sure which one, but as long as this filter is meant to remain a hidden "LTA" filter, its true positive ratio must improve and/or it needs to stop causing AIV reports. ~ ToBeFree (talk) 22:04, 4 October 2023 (UTC)
 * It is hidden,
 * it labels edits as "LTA" without actually being about a specific LTA user at the moment, and
 * it has many false positives that repeatedly appear at WP:AIV/TB2. Sometimes, the same user is re-reported immediately after I remove a report.


 * @ToBeFree It doesn't look like this filter is taking any actions. It could be removed from reporting to WP:AIV if it's not helpful there, but it's not tagging or disallowing edits as far as I can tell so I'm not sure it's particularly bothersome for the filter to be running. Sam Walton (talk) 22:09, 4 October 2023 (UTC)
 * In that case, I think the filter should be renamed and perhaps opened to public view. ~ ToBeFree (talk) 22:11, 4 October 2023 (UTC)
 * I'd be opposed to making it public, and there is a specific LTA it's aimed at though I'm not going into identifying them. But what am I missing - this filter has never been listed at Template:DatBot filters. Examples maybe? -- zzuuzz (talk) 22:39, 4 October 2023 (UTC)
 * Sorry, I seem to have mixed this up with filter 279 which is listed there and usually comes together with 1253. Examples are  . When I see such reports, I look at the LTA filter first; looking at 1253, I came to the conclusions listed above. 279 was even "automatically flagged as harmful" due to its huge amount of hits. I'm not sure if the filter system is really meant for this purpose, but I'll stop complaining as I can just remove 279 from Template:DatBot filters if it remains an issue. ~ ToBeFree (talk) 00:20, 5 October 2023 (UTC)
 * Looking again, a single edit at 2023-09-23T20:36:32 causing 12 (!) filter hits on two filters seems weird. Has the user clicked "Save" that often? ~ ToBeFree (talk) 00:26, 5 October 2023 (UTC)
 * Yes there seems to be something a bit weird going on there, with the ones I looked at all having the same timestamp. If it helps, I do think we can probably disable 1253 at this time. I'd wait for some more comments about that though (at the same time I can't see how it's the problem here). -- zzuuzz (talk) 00:32, 5 October 2023 (UTC)
 * Okay, I'm now confident we're dealing with a bug in the edit filter extension itself. Special:AbuseLog/36059143 happened for a single edit after 10 hours of no contributions. This isn't an issue with either filter 1253 or 279, but it may be caused by two filters interacting in a way they weren't expected to by the developers. Reported as T348237. ~ ToBeFree (talk) 12:10, 5 October 2023 (UTC)
 * Can you subscribe me to that task? I think there are two bugs being conflated here. (1) The "automatically flagged as harmful" warning is based on the number of hits before the throttle is applied. For example, has the same message there, seemingly permanently, because it matches about 12% of all edits, before the throttle settings are taken into account. But it really only matches fewer than 50 edits per day, so isn't "harmful" by any reasonable definition. (2) Every once in a while, a single edit is counted multiple times by the edit filter. This might not necessarily be entirely a problem with the edit filter extension; perhaps the browser is sending multiple POST requests for a some reason. (mumble mumble race condition) Whatever it is, it's hard to reproduce. Suffusion of Yellow (talk) 00:44, 6 October 2023 (UTC)
 * See also public filter ; curious to see what, if anything, turns up there. Suffusion of Yellow (talk) 00:49, 6 October 2023 (UTC)
 * Thank you very much! In today's weirdest coincidence, I subscribed Samwalton9 before seeing your same-minute subscription request. Thanks for creating the public 3/1 throttling test filter, that's good to have especially for discussions. I should probably have started this discussion on the editfilter mailing list, but I hadn't seen a need to subscribe before and thought the filter could probably be opened up anyway. ~ ToBeFree (talk) 00:57, 6 October 2023 (UTC)

Request for helper permission: Partofthemachine
Withdrawn. Partofthemachine (talk) 01:59, 13 October 2023 (UTC)

I am requesting this permission, partially because of my activity here and on WP:EFFP, but primarily because I do a significant amount of work in combating vandals and LTAs. Being able to view private edit filters would make this work easier. Partofthemachine (talk) 23:50, 10 October 2023 (UTC)
 * Support done some good work, I don't see a reason not to support, so therefore I support. Zippybonzo &#124;  talk to me   &#124;   what have I done   (he&#124;she&#124;they) 05:25, 11 October 2023 (UTC)
 * Hi, thank you for the work you do at Wikipedia. For my knowledge, may I please request you to explain how viewing private filters might make your work in combating vandals and LTAs easier? Thank you. Lourdes  06:09, 11 October 2023 (UTC)
 * Support as I see no issues for your good work. Signed, 64andtim  ( any problems? ) 00:39, 12 October 2023 (UTC)
 * Oppose, as I don't see a strong need for the permission at this point. You only have around 30 edits to filter-related pages, and I don't see any demonstrated knowledge of regex that would be useful for creating filters. EFH is a very sensitive permission - for example, it allows you to see emails and phone numbers, which are technically oversightable, and there are no logs for when you access filters - and it's really not that helpful for anti-vandalism work. — Ingenuity (talk • contribs) 01:26, 12 October 2023 (UTC)
 * Oppose. 13 edits to EFFPR according to xtools is quite low. Not sure how EFH could be used for anti-vandalism either.. If you are referring to patrolling the filter logs, most of the hidden filters are disallow so there's not much a non-admin can do except taking someone to AIV. But if a user has no edits except log entries there's not much real harm they are doing, so it's probably better to patrol cases that are actually actionable for non-admins. 0x Deadbeef →∞ (talk to me) 02:50, 12 October 2023 (UTC)
 * As with Ingenuity and 0xDeadbeef, I'm gonna have to oppose, as I'm afraid you might not have sufficient experience with edit filters (your responses to false positives are significantly low, unfortunately). Signed, 64andtim  ( any problems? ) 13:47, 12 October 2023 (UTC)
 * Oppose Your activity on WP:EFFPR is lacking currently, as 0xDeadbeef pointed out. I personally have 513 edits to that same page, and I'm still not sure it's enough to go for EFH. As for your vandalism work, you can see most of what you need already in the public filters, and for LTA work, I'm not entirely confident you need the entirety of the private filters. Something that has been pointed out in other discussions for LTA-specific grants is that EFMs can, at their discretion, send non-EFH/non-admins logs to singular private filters. This may be a better solution, depending on your use-case, but I'm much more likely to support someone who is either authoring regex, or helping at WP:EFFPR. EggRoll97 (talk) 19:49, 12 October 2023 (UTC)
 * Oppose per EggRoll97 and suggest a withdrawal. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 21:00, 12 October 2023 (UTC)

Setting Special:AbuseFilter/1182 to disallow
Would there be any objection if I set Special:AbuseFilter/1182 to disallow? This filter has been enabled for a while and appears to have a minimal false positive rate. The target of the filter has been active recently. cc as filter maintainer. Spicy (talk) 09:25, 15 October 2023 (UTC)


 * Sounds good to me. — Ingenuity (talk • contribs) 14:22, 15 October 2023 (UTC)
 * ++ Lourdes  06:54, 16 October 2023 (UTC)
 * I've implemented the change. Thanks, Spicy (talk) 08:33, 17 October 2023 (UTC)
 * Thanks. I'm usually a bit hesitant to set LTA filters to disallow; someone with enough time on their hands will usually find a way around, or at least find a brand-new way to disrupt, so log-only+DatBot often works better. But this seems like really specific obsession, so it's worth a try. Suffusion of Yellow (talk) 23:22, 20 October 2023 (UTC)

'Userspace spam'
In light of the following hit:
 * 17:33, 15 October 2023: 40.134.93.252 (talk) triggered an edit filter, performing the action "edit" on Talk:Stage hypnosis. Actions taken: Disallow; Filter description: Userspace spam

The filter description states 'Userspace spam' triggered on a non-user talk page. I suggest exclude non user talk pages from the filter or otherwise create a new filter. Toadette (let's chat together) 17:46, 15 October 2023 (UTC)


 * cc @Suffusion of Yellow: should we rename this? 0x Deadbeef →∞ (talk to me) 03:53, 16 October 2023 (UTC)
 * It was split from another filter for supporting some other non-user spaces. We could rename this, rather than removing 1. Lourdes  05:20, 18 October 2023 (UTC)
 * Renamed to "Userspace or talk page spam". See the filter notes for why I'm not using a more specific name. Suffusion of Yellow (talk) 23:28, 20 October 2023 (UTC)

Filter 712
If you look at this log, the IP was simply adding df=y despite the filter tagged the edits with "possible birth or death date change". Suggest add like  to prevent tagging of innocent edits. Toadette (let's chat together) 07:43, 16 October 2023 (UTC)


 * Fixed. Went with  We still want to tag edits which add or remove a   and change the date. Suffusion of Yellow (talk) 23:56, 20 October 2023 (UTC)

Setting Filter 149 to dissallow
I have been thinking that filter 149 could be changed to disallow due to the very low, if any FP rate and the fact that 99% of the edits it stops are spam. It is currently warn and tag but that has seemed to stop very few of the spammers. They are usually blocked very quickly but changing the filter to disallow could take a few blocks of the hands of admins. Just an idea, Seawolf35 (talk) 23:56, 21 October 2023 (UTC)
 * If the reason is to take a few blocks of the hands of admins, that may be incorrect, as one of the reasons for this filter is to capture promotional user names. Such names -- whether their edit goes through or not -- would need to be blocked with the filter-generated evidence. On the other hand, the decision to keep a weblink on the article page or not should rest with editors on the particular page. For example, an article on a club may witness a username named after the same club attempting to insert the club's official links into the page. We block the user; but allow the edits, for review by other editors... Thanks, Lourdes  05:45, 22 October 2023 (UTC)
 * I second @Lourdes‘s opinion here. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 16:24, 24 October 2023 (UTC)
 * Same. I agree with both of your opinions. 64andtim  ( talk to me ) 20:18, 24 October 2023 (UTC)

New filter for sports venue vandalism
Special:AbuseFilter/1272 is currently running without any actions. The intent is to match "team A now owns the stadium of team B" vandalism like. This is easiest to detect when it happens in the infobox, but I hope to expand the filter over time. Daniel Quinlan (talk) 03:17, 27 October 2023 (UTC)


 * I have also seen vandalism like this, and a filter is probably a good idea. This filter has a high FP rate though 4/5 of the hits are FPs. Hopefully we can iron out the issues and then turn on tag. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 23:31, 28 October 2023 (UTC)
 * I fixed the first two mistakes already, but it definitely needs more work. The hit rate is pretty low so I'm going to let it run for a week or two to build up some data and then I'll try to reduce the false positive rate further. Daniel Quinlan (talk) 23:38, 28 October 2023 (UTC)
 * Perfect. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 23:42, 28 October 2023 (UTC)

Set 1086 to disallow

 * ("Disruptive edit summaries", public)

As per the recent vandalism by an IP hopping vandal using... inflammatory edit summaries. Deauthorized. (talk) 18:12, 28 October 2023 (UTC)


 * This is one of those LTAs but who treats Wikipedia as a video game; a disallowing filter will be a new "level" and will encourage them more. Just WP:RBI until they get bored again. If someone wants to play their game anyway, please use a separate private LTA filter. (But I haven't checked the FP ratio of 1086 in a while; if the non-LTA hits justify setting it to disallow anyway, that's different.) Suffusion of Yellow (talk) 20:01, 28 October 2023 (UTC)
 * I suppose that makes sense. I took the liberty of looking through that filters logs (outside of the ip hopping) and it seems to be mostly vandalism. Though it's also caught by other filters as well. Do you think it'd be worth disallowing just to avoid having to request revision deletion? Deauthorized. (talk) 20:34, 28 October 2023 (UTC)
 * As to the LTA, no, they'll just find something else possible even more offensive to say. I doubt they are fixated on those specific words. As to the drive-by vandals, maybe, but I'd like to move stuff like "!!!!!!!!!!!!" out of this filter; there's nothing revdelable about that, and it occasionally comes with good edits like Special:Diff/1181421832. Also remember that a "bad" summary can sometimes draw attention to otherwise sneaky vandalism. Suffusion of Yellow (talk) 20:58, 28 October 2023 (UTC)
 * It looks like there's a new private filter that's set to disallow; see here. 64andtim  ( talk to me ) 18:33, 29 October 2023 (UTC)
 * General reminder, LTAs read this page too. Suffusion of Yellow (talk) 19:14, 29 October 2023 (UTC)
 * I understand now that I removed the filter name from my reply, and thanks for reminding me. 64andtim  ( talk to me ) 20:08, 29 October 2023 (UTC)

New disallow message proposal for filter 793
I would suggest a new disallow message for filter (Common spam/scam phone numbers, private). While I cannot see which disallow message the filter uses (I'm assuming MediaWiki:Abusefilter-disallowed), it would be nice if we could have a custom disallow message similar to MediaWiki:Abusefilter-disallow-email, the standard disallow message with additional information, or something like that. I'd suggest that the new disallow message for filter 793 should be called MediaWiki:Abusefilter-disallowed-793 or something else. Thanks. 64andtim ( talk to me ) 18:28, 31 October 2023 (UTC)


 * Could and edit filter helper/manager clarify what the current disallow warning message is? Thanks! - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 19:58, 31 October 2023 (UTC)
 * It's just the standard abusefilter-disallowed, what's the issue with that? There are a decent amount of non-spammer/scammer hits but we wouldn't really want to tell spammers/scammers what triggered the hit. Galobtter (talk) 01:01, 1 November 2023 (UTC)
 * I concur. We don’t want spammers to know that we’ve blocked their spam number or email. Then they’d attempt to get around it. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 01:04, 1 November 2023 (UTC)
 * While I believe that the standard disallow message is mostly used for filters that catch and stop vandalism, I concur too. 64andtim  ( talk to me ) 01:23, 1 November 2023 (UTC)

Filter 1273 and false positives
I understand some of why this filter was split out from 384. I appreciate your continued fine-tuning of the 1273 filter as false positives are encountered. Do you have any objection to me proactively incorporating some of the fine tuning for false positives from 384 into 1273? We've been getting multiple legitimate false positive reports over the past day, and you've been very responsive, but I wanted to check before I go and do this boldly. — Red-tailed hawk  (nest) 16:41, 1 November 2023 (UTC)


 * @Red-tailed hawk: sure, go ahead. — Ingenuity (talk • contribs) 16:50, 1 November 2023 (UTC)

Adding some kind of "moot" to EFFP
I wanted to float the idea of adding a "moot" option (or two) to EFFP, but figured it might warrant discussion first given how often we use it. Something to indicate that the filter hit reported wasn't a disallow, and the user made the edits, and then either the filter was right or wrong and might need to be changed. Something like Edit not prevented - original edit was saved successfully, and the filter is working properly and Edit not prevented though a change to the filter may be needed - edit was saved successfully. Thoughts?

@Taavi FYI since your bot uses the template. DannyS712 (talk) 05:42, 4 November 2023 (UTC)


 * Support as I have also seen times where this would be useful. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 13:39, 4 November 2023 (UTC)
 * I added the two options to Template:EFFP/sandbox under the moot (m) and mootefm (me) parameters. As of this edit, they are;
 * (moot)
 * (mootefm)
 * Let me know what you think. Deauthorized. (talk) 15:58, 4 November 2023 (UTC)
 * Though I should maybe add linking to the diff as well. Deauthorized. (talk) 16:01, 4 November 2023 (UTC)
 * Looks like a good start to me. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 16:20, 4 November 2023 (UTC)
 * Support That would be nice to have. Seawolf35 (talk - email) 02:21, 5 November 2023 (UTC)
 * Support as this would be a nice addition to the EFFP template. 64andtim  ( talk to me ) 03:12, 5 November 2023 (UTC)
 * @Deauthorized we can always link to the diff in the subsequent comment if needed. I think those are good to go, but lets wait for @Taavi to confirm that using them won't break the bot - we probably want  to act like, and   to act like  in terms of when to archive. I also propose having the short version of   be   to match   vs  , but thats a minor thing. DannyS712 (talk) 08:32, 5 November 2023 (UTC)
 * You can make that change yourself if you want. Deauthorized. (talk) 16:19, 5 November 2023 (UTC)
 * That came off a bit brash, I'm sorry. I added m! as a short code. Deauthorized. (talk) 18:03, 5 November 2023 (UTC)
 * I added configuration for  and   to act like  . I don't see any special handling for   so I haven't added any for  . Sorry for the delay on my side. Taavi (talk!) 20:50, 9 November 2023 (UTC)
 * I've added the statuses with Special:Diff/1184375338 (template) and Special:Diff/1184375608 (codes documentation). Thanks everyone --DannyS712 (talk) 01:21, 10 November 2023 (UTC)

False positives - backlog of pending filter changes
Hi all. For any EFMs that watch this page but don't monitor the false positives page, Edit filter/False positives/Reports currently has 5 sections that are marked as potentially needing a change to a filter. Would someone mind taking a look? Thanks, --DannyS712 (talk) 03:41, 8 November 2023 (UTC)


 * - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 11:49, 8 November 2023 (UTC)
 * I just had to restore some threads from the archive that were unresolved - do we need more EFMs? --DannyS712 (talk) 03:06, 9 November 2023 (UTC)
 * I think many are just busy right now. I know Deadbeef specifically is away for this week. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 03:08, 9 November 2023 (UTC)
 * backlog now cleared. 0x Deadbeef →∞ (talk to me) 13:28, 11 November 2023 (UTC)
 * @0xDeadbeef thanks! DannyS712 (talk) 00:15, 12 November 2023 (UTC)
 * In response to "do we need more EFMs", I think the answer is yes. And I was surprised to see that you are only an EFH and not EFM. Please request one! Given your continued involvement with edit filters, your help would be awesome. I'm sure there are way more examples, but this and User:DannyS712/EFFPRH clearly show that you would be very helpful as an EFM :) 0x Deadbeef →∞ (talk to me) 09:23, 13 November 2023 (UTC)
 * Suffusion of Yellow (talk) 02:47, 14 November 2023 (UTC)
 * definitely have the skills to be one. Galobtter (talk) 02:49, 14 November 2023 (UTC)
 * Thanks for the votes of support but given that I recently returned from a long bout of inactivity I think it would be premature to request advance rights without being back for a while DannyS712 (talk) 02:56, 14 November 2023 (UTC)

EFH for EggRoll97
Prior requests: 2018, 2019, 2020, 2020.

Closing administrator of previous discussion:
 * Participants in previous discussion:

Hello all. I've come today to request access (again) to the edit filter helper right. I am aware in advance that this being my 5th request doesn't look good, though I believe I have mitigated factors brought up in my previous requests. I've spent the last few days in part writing this up aiming to hopefully address any previous problems, and I'm happy to elaborate on any concerns anyone has as well.

The main concern was my lack of experience. Since my most recent request, I am able to say that I believe myself to be, at this time, far more versed than my past self, and in hindsight can see my previous requests as shortsighted. In addition to other areas of the wiki I have dipped into, I've also begun to experiment with Suffusion of Yellow's FilterDebugger, and I've garnered a somewhat basic understanding of regex since my previous request (though I will admit wholeheartedly that I would not trust myself to try and write anything at the current time without constantly double-checking myself). As for the demonstrated need, I have frequently begun to run into false positives with private filters that even began going unanswered for such a length that one section was archived without a response. The main filter I ran into was 1273, though I've also ran into other filters as well recently. To address a concern made by Danny in the most recent request, I have mentioned a few times on the false positives page suggestions for improvements to filters to avoid false positives, such as here (This section might archive by the time I post this, but it's referring to the exclusion of categories from 380). EggRoll97 (talk) 11:00, 12 November 2023 (UTC)
 * Support. Thanks for your help in responding to false positive reports. 0x Deadbeef →∞ (talk to me) 11:23, 12 November 2023 (UTC)
 * Support. Net positive, has grown up since last requests, active in edit filter areas. I run into the same issue with private filters and can vouch for the need this user has. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 13:04, 12 November 2023 (UTC)
 * Support my concerns from the last discussion have been addressed --DannyS712 (talk) 15:11, 12 November 2023 (UTC)
 * Strong support as you're a trusted user, have relevant experience with EFFPR (550+ responses to reports), I see no issues here. As with Illusion Flame, I would have a need for the EFH right to help with EFFPR but have almost 200+ responses, definitely not enough for EFH in my opinion. Anyway, I give you a strong support for you to receive the right. 64andtim  ( talk to me ) 21:49, 12 November 2023 (UTC)
 * Support Full disclosure, we already discussed this at my talk page. As with the others, I am (finally) satisfied with EggRoll97's experience and long-term dedication. Suffusion of Yellow (talk) 02:32, 14 November 2023 (UTC)

"Possible template vandalism" creation?
While I was patrolling the abuse log, I noticed this:

17:55, November 15, 2023: Kadekeys (talk | contribs) triggered an edit filter, performing the action "edit" on Template:Taxonomy/Serpulaceae. Actions taken: none; Filter description: Possible template vandalism (diff)

While I will not give out details of the filter or what triggered it, this student editor above made a legitimate subpage of a template; that filter logged the creation as "possible template vandalism". Is this normal? — 64andtim  ( chat  •  see here ) 00:04, 16 November 2023 (UTC)
 * From a cursory review it looks fine, I see why the filter tripped but I'm not sure if it should have in this case. EggRoll97 (talk) 03:12, 16 November 2023 (UTC)
 * @EggRoll97: in this case, I believe it shouldn't have tripped because that editor was a legitimate user. — 64andtim  ( chat  •  see here ) 03:18, 16 November 2023 (UTC)
 * I wouldn't worry about it - I see why it was triggered, but that filter is deliberately very broad. There's a reason it is a log only filter, and the point of it is to cause a report to be made if someone triggers the filter a lot (which might indicate someone doing some nasty template vandalism that has to be caught quickly, which is an on and off issue). Galobtter (talk) 04:06, 16 November 2023 (UTC)

EFH for Illusion Flame
(Prior request: April 2023) Hello!

To those of you who don't know him, has been a meticulous patroller of the Edit filter false positive reports board over the past eight months. He has made over 500 edits in response to reported false positives in that time, and he has been one of the most active participants at the Edit filters noticeboard ever since he got into helping respond to edit filter false positive reports. I've been very impressed with this user's contributions in this space over the past several months—he has truly been an asset in helping with this area. I think that Illusion Flame's dedication would serve us very well if we were to allow him to help with reviewing false positive reports involving private filters. For these reasons, I'm happy to nominate Illusion Flame for the role of edit filter helper.

Outside of his dedication to helping with edit filters, Illusion Flame has also served the community in several roles with elevated permissions that require trust, including as a VRT agent, new page reviewer, and page mover. Since his last request in April, Illusion Flame has shown that he can be trusted with private information through other community roles, and has burgeoned into an experienced contributor to our project. I think that granting him the role of EFH is a no-brainer, and I encourage you to join me in supporting this request. — Red-tailed hawk  (nest) 01:10, 14 November 2023 (UTC)

Candidate, please indicate acceptance of the nomination here: Thank you RTH for the kind message and endorsement. As stated above, I plan to use the tools mainly to respond to private edit filter false positive reports. I graciously accept this nomination! - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 01:15, 14 November 2023 (UTC)

Discussion: EFH for Illusion Flame

 * Support as nominator. — Red-tailed hawk  (nest) 01:18, 14 November 2023 (UTC)
 * Also, Pinging you as participants in the prior discussion, as per policy. —  Red-tailed hawk  (nest) 01:22, 14 November 2023 (UTC)
 * Support clearly has trust and need. 0x Deadbeef →∞ (talk to me) 01:26, 14 November 2023 (UTC)
 * Support as he has a demonstrated need and is trusted. 64andtim  ( talk to me ) 02:38, 14 November 2023 (UTC)
 * Support Long-term dedication at EFFP and to the project in general. Suffusion of Yellow (talk) 02:46, 14 November 2023 (UTC)
 * Support very helpful and meets the criteria --DannyS712 (talk) 02:54, 14 November 2023 (UTC)
 * Support: Seems like we need more people with EFH and Illusion Flame has put in a lot of good work. They should be able to utilize the permission in a productive way and they're pretty receptive to feedback they receive, so I say yes. Hey man im josh (talk) 03:01, 14 November 2023 (UTC)
 * Support regularly helps with EFFP and meets the criteria. – DreamRimmer (talk) 03:19, 14 November 2023 (UTC)
 * Support Trustworthy and experienced, and has clearly demonstrated a need. EggRoll97 (talk) 05:26, 14 November 2023 (UTC)
 * Strong Support: I have no issues with the candidate, they are knowledgeable with the edit filter area, and I believe them having signed the NDA further shows their ability to be trusted with sensitive data such as emails or similar filtered by the private filters. After some exposure to the private filters I would encourage Illusion Flame to run for EFM (assuming they had regex experience obviously). Zippybonzo &#124; talk  &#124;  contribs  (he&#124;she&#124;they) 07:12, 14 November 2023 (UTC)
 * Pile-on support: I supported the last request—probably naïvely—but I'm happy to support again. Now with over six months tenure and significant contributions to this area, I don't see any immediate disqualifiers for this net-positive. All the best, Schminnte</b> &#91;<i style="color:#0645AD">talk to me</i>&#93; 08:29, 14 November 2023 (UTC)
 * Support They work in one or more of the VRT queues so knowing how to manage private data isn't an issue. I also communicate with them semi regularly and I've always gotten good vibes from them. Happy to support Deauthorized. (talk) 11:55, 14 November 2023 (UTC)
 * Support I would like to work with private filters as well someday, but I am not at that level yet, Illusion Flame is. Seawolf35 (talk - email) 15:40, 14 November 2023 (UTC)
 * Strong Support - Illusion flame is good candidate for this permission; this permission will give allowing to working with public and hidden filter.  αvírαm  | (tαlk) 13:12, 15 November 2023 (UTC)

Line 3 of Special:AbuseFilter/1239
I understand the rationale behind having this sort of phrase included in something like a tracker/warning filter, but I'm having a good bit of trouble understanding why that phrase would be included in a disallow filter. Is there a long-term pattern I'm missing here? — Red-tailed hawk  (nest) 03:18, 17 November 2023 (UTC)
 * I think I get the idea behind why it's in the filter, but it seems a bit broad, especially for a disallow filter. (Reasoning: Same type of people who would use that phrase would probably use the phrase at line 9 as well.) EggRoll97 (talk) 07:47, 17 November 2023 (UTC)
 * I agree. I’ll need to take a closer look at false positives, but it may be worth changing the filter to tag only. (Or simply moving line three to a separate tag filter) - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 11:22, 17 November 2023 (UTC)
 * As far as I know some LTAs really love using that phrase, and note that this was the email that prompted the creation of this. And by the way, this is the sort of discussion that could take place on a private wiki. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 12:09, 17 November 2023 (UTC)
 * Are edit filter helpers added to that mailing list? - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 12:15, 17 November 2023 (UTC)
 * You'd need to subscribe to it, and then a mailing list admin will need to review and approve your subscription. Yes, EFHs should have access. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 12:17, 17 November 2023 (UTC)
 * ✅, but I agree that a private wiki is a better place then a mailing list. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 12:21, 17 November 2023 (UTC)
 * Re-reading the mailing list info, it does say that subscription is limited to edit filter managers and administrators, however I did request and got access approved after I got EFH, so EFHs should de facto be able to request and get access. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 17:35, 18 November 2023 (UTC)
 * Having read the mail exchange, I still feel that line 3 is a bit trigger-happy for a disallow filter. Separately, I agree that a private wiki would be helpful in being able to specifically discuss what's going on, but I also worry about participation on it (I can't imagine it would get nearly as many EFH/EFM eyes that would regularly review it). — Red-tailed hawk  (nest) 22:06, 17 November 2023 (UTC)
 * Along these lines, and seeing no hard objections, I'm going to make a small tweak to line 3 so as to make this more narrowly tailored. — Red-tailed hawk  (nest) 19:06, 18 November 2023 (UTC)
 * Looks good. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 19:24, 18 November 2023 (UTC)

LTA/Private filters discussions wiki?
Not everyone who is an EFH/EFM/admin is subscribed to the mailing list, and not everyone is on IRC. It might be nice to have a private wiki for coordinating changes to/discussing contents of private filters, which can also have better documentation as compared to what we have at Special:AbuseFilter/1231. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 10:08, 10 November 2023 (UTC)


 * Seems like overkill to start an entire new project for that, just subscribe to the mailing list. — xaosflux  Talk 10:26, 10 November 2023 (UTC)
 * The mailing list is quite low activity and scales poorly when new requests for LTA edit filters can go unanswered for weeks and months. Composing a reply in one of those threads is quite different from leaving comments/messages on a private wiki. And when threads go unanswered, how might one know if action was needed but no one got to it or rather action wasn't needed at all? To my knowledge, there's little communication about private filters in semi-public or private places about private edit filters and LTAs. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 10:46, 11 November 2023 (UTC)
 * Xaosflux couldn't put it better, seems overkill to go to the effort of making an entire wiki because people don't subscribe to a mailing list. Zippybonzo &#124; talk  &#124;  contribs  (he&#124;she&#124;they) 07:14, 14 November 2023 (UTC)
 * I agree with @0xDeadbeef that a private wiki would be extremely helpful in coordinating edit filter changes, compared to an email thread. As someone who isn’t on IRC, this is very good idea in my opinion. I’m not sure how to go about requesting one though. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 12:22, 10 November 2023 (UTC)
 * Being an edit filter helper for a day now shows me how helpful this could be. For example, I’m still learning the ropes and want to ask questions about a few private filters. Rather than notify a whole mailing list who probably doesn’t care, I think a wiki type discussion format would be quite useful. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 11:45, 17 November 2023 (UTC)
 * Requests go through phabricator, but they will want to see more consensus that this is needed before creating it - I'm happy to help with the process but it would be premature to file a request now DannyS712 (talk) 13:01, 10 November 2023 (UTC)
 * I agree that it would be premature now, just curious what the process was. Thanks! [[Image:Face-smile.svg|20px]] - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 21:48, 10 November 2023 (UTC)
 * I'd prefer a slightly larger scope: edit filters and general discussion of LTAs that violates BEANS or DENY. Not every "tell" can be matched by a filter. Also, possibly, make it a global project, open to admins/EFMs from other wikis. Suffusion of Yellow (talk) 22:25, 17 November 2023 (UTC)
 * Excellent idea, I agree. It also kind of addresses the point made by User:Red-tailed hawk below. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 22:28, 17 November 2023 (UTC)
 * I'd be a bit hesitant about allowing access from admins on all wikis. wikt:User:Wonderfool/alternative accounts - LTA became an admin four times after being blocked the first time (first account was also the original admin). We also saw with Special:Diff/1183105704 that even our (enwiki's) screening of admins isn't perfect, and we probably have a lot more scrutiny than most wikis - a truly motivated LTA could probably become a sysop on another wiki given enough time. DannyS712 (talk) 23:24, 17 November 2023 (UTC)
 * If we plan to make a private wiki that allows discussion of hidden filters, we should probably keep it to the English Wikipedia only; should admins, EFMs and EFHs automatically get access, or is a subscription required? — 64andtim  ( chat  •  see here ) 00:22, 18 November 2023 (UTC)
 * Something to consider is that not every sysop is interested in, or actively involved with the edit filter. At the time of writing, there are 137 edit filter managers, because admins don't automatically have access. There are a further 21 edit filter helpers. I would presume there are a few admins who use the private filter viewing right bundled with sysop, but likely not a lot, so I'd add another 20 or so to the total. In total, that means there's probably around 200 people on the wiki who actively engage with private filters, and not all the administrators with EFM are actively editing filters. Realistically, it's a fairly small group of people, so it probably should be a subscription type of thing, where not everyone is automatically given access. EFHs and EFMs should probably be given access automatically, with admins able to gain access on request or by adding themselves to the EFM group, depending on their confidence in filter editing. EggRoll97 (talk) 01:11, 18 November 2023 (UTC)
 * Good points if the venue is only used for edit filter discussions. I do agree with Suffusion of Yellow that the scope could be larger, where the wiki could facilitate general discussion about LTAs, which means we could document the behavior patterns of LTA's while still in accordance with WP:DENY. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 04:34, 18 November 2023 (UTC)
 * IANAEFH let alone an admin, but to my knowledge most LTAs are either petty vandals or subtle abusers. The former wouldn't merit a private wiki (even having pages to document them is contentious), and the latter are monitored closely by admins and functionaries -- many of them will have much more useful documentation on the private checkuser wiki. Giraffer (talk·contribs) 14:06, 18 November 2023 (UTC)
 * I believe the idea is that this wiki would allow enwiki EFHs, EFMs, and Admins (if they wish) to discuss improvements to private filters that cannot be disclosed in public. It also would help us coordinate our filters about LTAs and have some documentation about them for future filter edits. And while I’m sure the CheckUser wiki does have more detailed info, most of the edit filter community cannot access it. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 14:26, 18 November 2023 (UTC)
 * I was responding to SoY's comment explicitly suggesting that the wiki's remit include LTA discussion, but regardless of whether it's included or not, one set of permission holders (EFH/EFM) on one wiki is an extremely small userbase to create a new wiki around. There are maybe 40 users actively involved with private filters across enwiki, and only 83 filters that have been modified this year. Absent a strong reason, I'm wary of reducing the transparency of a group already with little oversight. Giraffer (talk·contribs) 17:18, 18 November 2023 (UTC)
 * If we give access to admins (or like SoY's suggestion above, also to EFMs and sysops from other sites), I don't think this would be reducing the transparency of people working on private edit filters. The wikipedia-en-editfilters mailing list can be subscribed to by administrators, but it should be pretty obvious that not a lot of administrators are subscribed. (Not sure about the numbers, but it is possible that it has no non-EFM admins) Moving those mostly-dead discussions to a private wiki, though not sure whether it could revive some activity by being more accessible, should increase the transparency and not the other way around. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 17:33, 18 November 2023 (UTC)
 * Maybe I have an idea: for the English Wikipedia, EFHs and EFMs automatically get access, and administrators have the option to subscribe if they want to be interested about the edit filter discussion mailing list (not all admins get automatic access, but they have the option to subscribe without review). Although for the private wiki where we should keep the hidden filter discussions there and not on enwiki, we're gonna need much clearer consensus before we get started on developing that private wiki. — 64andtim  ( chat  •  see here ) 18:10, 18 November 2023 (UTC)
 * The mailing list infrastructure doesn't support doing this and we'd need a tool to automate that. You have to put an email address which gives mailing list owners a hard time verifying when you are not subscribed to any other mailing lists. (anecdote: I had to be contacted privately through discord verifying that the email was really me for getting access) <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 18:15, 18 November 2023 (UTC)
 * Oh, so for now everytime an EFH/EFM/admin subscribes to the current mailing list, it requires manual review; however, if we're not doing an automatic tool to automatically subscribe to the topic, how about that EFMs and admins can manually review and accept or decline so that this would mean a much faster response (they may have to sign the NDA first)? The mailing list only has two admins – this would mean a significantly slower response.— 64andtim  ( chat  •  see here ) 18:20, 18 November 2023 (UTC)
 * It wouldn't just be EFHs and EFMs who would be encouraged to join the private wiki; it would also be any admin who deals with LTAs. Which probably includes all the AIV and ANI "regulars". I think that's enough scrutiny. And the main advantage I see over a mailing list is organization. In a mailing list archive, information about LTA XYZ might be scattered across various (uneditable) threads named "Filter request", "Harassment" etc., but on a wiki, we can create something like WP:LTA but without worrying about the BEANS. Also some users might not like sharing their email addresses (and possibly IP addresses, depending on mail provider and client) with everyone else on the list, but on a wiki, it's only the person creating your account who can see your email address. Suffusion of Yellow (talk) 02:45, 19 November 2023 (UTC)
 * That would be nice, if there was a wiki in which we could properly write about WP:LTAs. Galobtter (talk) 03:55, 19 November 2023 (UTC)
 * But how would we restrict that discussion wiki to trusted users? I mean, sometimes I deal with those bad users and it would be nice if I came in to help. — 64andtim  ( chat  •  see here ) 04:25, 19 November 2023 (UTC)
 * I’m afraid that we’d probably have to restrict it to edit filter helpers/managers and admins, for security. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 11:17, 19 November 2023 (UTC)
 * Good idea, I concur with your statement. — 64andtim  ( chat  •  see here ) 16:13, 19 November 2023 (UTC)

Additional filter parameter
Hey all, I’ve seen some vandalism with “pp”, usually proceeded by “suck my”. Can an EFM add this term to filter #? - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 16:23, 30 October 2023 (UTC)


 * The regex of 'suck my (dick|pp|penis)*' should cover all that I believe. Zippybonzo &#124; talk  &#124;  contribs  (he&#124;she&#124;they) 20:28, 6 November 2023 (UTC)
 * Hmm. I do see that now. I guess I need to stop editing when sleep deprived - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 20:34, 6 November 2023 (UTC)
 * I never said it was in the filter, it possibly is but I can't bother to look, but it could be added to solve it. Zippybonzo &#124; talk  &#124;  contribs  (he&#124;she&#124;they) 07:18, 14 November 2023 (UTC)
 * I think this is already part of filter 260. Nobody  ( talk ) 07:30, 14 November 2023 (UTC)
 * Nope it's not as far as I believe, suck my (dick/penis/cock (take your pick)) is included but not suck my pp. Zippybonzo &#124; talk  &#124;  contribs  (he&#124;she&#124;they) 09:47, 18 November 2023 (UTC)
 * Probably could be added right after the entry for "PRICK" (I'm sure it's just nice fellows adding that to articles, right?), so it would read as follows (had to sub out | for - so the template works, but I think the point is made)? EggRoll97 (talk) 00:20, 25 November 2023 (UTC)

BLP vandalism LTA
See the recent history of Stan Mataele and Dan Blumenthal. This is the same crap that was being posted to AN/ANI the other night. Looks like it's already being tracked but we really need to start disallowing this nonsense. Taking Out The Trash (talk) 00:14, 27 November 2023 (UTC)
 * I think there's a couple filters that tag, and if I recall there's a disallow filter somewhere. I'm not able to see the contents of the edits in the history, though, since they appear to have been revdel'ed already. From the history though, it doesn't seem like a severe problem, and seems to have been caught quickly by RC patrollers. EggRoll97 (talk) 02:06, 27 November 2023 (UTC)
 * It looks like because of the severity, I have made a case about them. Apparently, for both biographies, we can't see the public or private filter hits for those pages in question.And as Suffusion of Yellow told me, "LTAs read this page too". – 64andtim  ( talk to me! ) 03:52, 27 November 2023 (UTC)

Filter 1159 exception
Despite Filter 1159 aiming to prevent mentions of Christine W. Chandler from being added to Wikipedia, a mention was present on the "List of outsider artists" until I removed it: https://en.wikipedia.org/w/index.php?title=List_of_outsider_artists&diff=prev&oldid=1187547794 Given that there will inevitably be people with one of the names Chandler has had throughout her life mentioned on Wikipedia (including the New Mexico politician linked) the filter can't block all of them, but this does show that it's not perfect. Alexschmidt711 (talk) 23:21, 29 November 2023 (UTC)


 * I believe we may need to change 1159 on the first line so that it will also apply to all namespaces except templates, the Wikipedia and MediaWiki namespaces, user (and user talk) pages, and pages in userspace (except the sandbox in said userspace).
 * And by the way, filter 1159 does not disallow such mentions, and instead it tags edits; a private filter probably does the job at disallowing mentions. – 64andtim  ( talk ) 01:19, 30 November 2023 (UTC)

Filter 1112
Would it be possible to check and disallow links added to disambiguation pages? I'm requesting this after seeing this edit go through. Nobody ( talk ) 08:33, 29 November 2023 (UTC)


 * Good idea, we should allow articles of people who have said articles about them on notable people lists while simultaneously disallowing disambiguation pages whose titles have peoples' names too. – 64andtim  ( talk ) 18:38, 29 November 2023 (UTC)
 * This doesn't seem possible to do via an edit filter. Perhaps a bot would be better suited for this task? Deauthorized. (talk) 10:45, 5 December 2023 (UTC)
 * All you need to do is check if Template:Human name disambiguation, Template:Disambiguation or one of their redirects, is on the article that they're trying to add. Can't that be done by a filter?  Nobody  ( talk ) 12:31, 5 December 2023 (UTC)
 * Edit filters can't retrieve article content from a title. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 13:16, 5 December 2023 (UTC)
 * Then a bot task would be needed for this. You think it would pass WP:BRFA 0xDeadbeef? Nobody  ( talk ) 13:21, 5 December 2023 (UTC)
 * I guess? You could make a bot tagging it as potential vandalism or another tag to get people's attention. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 14:00, 5 December 2023 (UTC)
 * I created a request here. Let's see how it goes. Nobody  ( talk ) 14:29, 5 December 2023 (UTC)
 * Incorrect, as links to disambiguation pages have  which can be identified by looking at new_html (as the filter already does to determine if the linked added is a redlink); see the HTML here. I support this change and it should be an easy one to make. Uhai (talk) 18:30, 5 December 2023 (UTC)
 * Yep, that will work. Let's see what (log-only) catches. Suffusion of Yellow (talk) 20:00, 5 December 2023 (UTC)
 * Gah, I clearly need to RTFM more. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 22:49, 5 December 2023 (UTC)
 * @Suffusion of Yellow How you think it's looking? All edit's catched were either reverted or had the correct link added in a following edit. Nobody  ( talk ) 08:26, 15 December 2023 (UTC)
 * Good enough. Merged the changes into 1112. Suffusion of Yellow (talk) 22:05, 15 December 2023 (UTC)
 * Looks good enough to everyone. – 64andtim  ( talk ) 01:40, 17 December 2023 (UTC)

Edit filter helper request for 64andtim

 * Not ending before 17 December 2023 04:54 UTC

Hello, everyone. I am self-nominating myself for the edit filter helper user right because I would like to help out in reducing the backlog of false positives tripped by private filters, to view their logs to help me identify certain LTAs and their behavior as part of my anti-vandalism work, and to suggest fixes for those private/LTA filters on the mailing list (not on-wiki) when possible. I probably have at least a (very) basic understanding of regex as I use Suffusion of Yellow's FilterDebugger script to know which parts of regex tripped filters, but I do not have the intention to author such filters.

My account has two-factor authentication enabled, I have never been blocked or sanctioned even for the first time, I have sufficient (near-native) ability of the English language to understand details and notes about private filters, and I'm currently an active, extended confirmed editor on the English Wikipedia. I am likely trusted to handle sensitive information (EFH allows me to see emails, real names, phone numbers, etc.), and probably have the confidence to not spill the beans about private filters.

And despite my EFFPR responses being over 400 or more, I'm pretty competent enough in my anti-vandalism/spam/LTA activities, if I'm not mistaken.

Thank you for your consideration in advance. – 64andtim  ( talk ) 04:54, 15 December 2023 (UTC)
 * Weak Oppose It's ironic that I'd oppose a request here, given that I made five of them before eventually getting the right, but I'm concerned when you say, or . This isn't so much a reflection on your work as much as it is a need for absolute certainty here that you can refrain from sharing filter content with others. I didn't particularly understand the reasoning until I actually had access to the vast amount of data that private filters do have. While your work is good, qualifiers to whether you can be trusted to not spill the beans can't cut it, because abuse of the EFH right is virtually undetectable, and the level of trust required is generally on par with sysop. Oppose Changed to a full oppose based on Mdaniels5757's comments. I've seen this around too and that kind of recklessness is how actual false positives get ignored. EggRoll97 (talk) 19:52, 15 December 2023 (UTC)
 * About "abuse" of the right, if I'm not wrong, I believe that includes disclosing private filter details, notes and code here or off-wiki, such as Discord. If anything, I will do my best to not to spill those beans involving private or LTA filters to anybody (including very trusted users), unless if they are an admin, an EFM, or an EFH. When there are suggestions to fix private filters because of false positives involving those filters in EFFPR, maybe I could take it to the mailing list since LTA filters involve "intense and sensitive" data that power such private filters, and it would be no benefit, and all risk, to disclose here. So it's probably a better idea to remain safe than sorry. –  64andtim  ( talk ) 05:54, 16 December 2023 (UTC)
 * Weak support. You will probably get opposition because you said that you don't want to author filters, (perhaps also implies not intending to help with authoring filters?) and EFHs without intending to become an EFM in the future is of limited help to the project. On the other hand, I have seen your contributions to patrolling the EFFP page, and I believe you can be trusted with this user right. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 11:58, 16 December 2023 (UTC)


 * Support They are actively involved here and I believe they have enough experience, as is evident from their contributions. I am also concerned about the privacy of sensitive information, but I think they can be trusted with private data. – <b style="color:black; font-family: Tahoma">DreamRimmer</b> (talk) 13:24, 16 December 2023 (UTC)
 * Support helpful with dealing with false positives --DannyS712 (talk) 00:46, 17 December 2023 (UTC)
 * Oppose Has been processing requests when they can't see the private filters (e.g. Special:Permalink/1190426564), which suggests either recklessness or lack of understanding of the purpose of EFFPR to me. Even though the grant of EFH would of course make it so they can see the private filters, the underlying recklessness or lack of understanding will not change with the EFH bit. &mdash; Mdaniels5757 (talk &bull; contribs) 21:08, 17 December 2023 (UTC)

flood of reports about lta 1164
false reports page is getting a bunch of reports about filter lta 1164. not an edit filter manager, can't look into it. what's going on? ltb d l (talk) 15:13, 19 December 2023 (UTC)
 * @Ltbdl Its Filter 1164 that Ohnoitsjamie just changed. Nobody  ( talk ) 15:15, 19 December 2023 (UTC)
 * Disabled for now. OhNo itsJamie Talk 15:17, 19 December 2023 (UTC)

Hiding of filter log entries
Is it possible to redact details for abuse log entries concerning public filters? I was prompted by the abuse log of, who made an edit at Chink that is now RD2'd and which triggered filters 50 and 491. (The edit incites genocide against Jews and Chinese people.) –<b style="color:#77b">Laundry</b><b style="color:#fb0">Pizza</b><b style="color:#b00">03</b> ( d c̄ ) 05:07, 19 December 2023 (UTC)


 * Currently, edit filter log entries cannot be revision deleted, and can only be suppressed by oversighters. – 64andtim  ( talk ) 05:44, 19 December 2023 (UTC)
 * See also T115530. Suffusion of Yellow (talk) 20:17, 19 December 2023 (UTC)

Emoji in cite web template in an edit
On this edit, an IP added some exclamation marks and a Moai emoji (🗿) in a cite web template parameter, which managed to bypass filter 680. I have since reverted it, and can an EFM modify the filter to prevent additions of emojis to reference templates? Thanks. – 64andtim  ( talk ) 19:15, 19 December 2023 (UTC)


 * References are excluded intentionally; emojis sometimes occur in the title, at least. Suffusion of Yellow (talk) 20:24, 19 December 2023 (UTC)
 * Oh, I guess that the emoji filter intentionally excludes emojis in references, anyway. – 64andtim  ( talk ) 22:54, 19 December 2023 (UTC)

About filter 247's warning and disallow messages
For the current disallow message and previous warning message used by filter 247 ("Adding emails", private), I have a suggestion:

Should we delete MediaWiki:Abusefilter-warning-email (because the filter only prevents users from adding emails to pages, and not warn them anymore) and move MediaWiki:Abusefilter-disallow-email to MediaWiki:Abusefilter-disallowed-email (to maintain consistency with the MediaWiki:Abusefilter-disallowed series), or do one or the other? If you have suggestions or if you disagree with my suggestion, let me know. Thanks. – 64andtim  ( talk ) 18:04, 7 December 2023 (UTC)
 * I don't see any problems with this. EggRoll97 (talk) 20:11, 7 December 2023 (UTC)
 * @EggRoll97: what I meant to say was that the warning message is unnecessary, but we should move the disallow message to MediaWiki:Abusefilter-disallowed-email to maintain consistency, or not. This might be a controversial action so we probably need some consensus before proceeding such actions. – 64andtim  ( talk ) 20:26, 7 December 2023 (UTC)
 * As in, I don't see any problems with this change. Might add the disallow seems to be fine, but I don't have any issues with moving it for consistency. Support. EggRoll97 (talk) 20:27, 7 December 2023 (UTC)
 * LGTM - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 00:25, 13 December 2023 (UTC)
 * Notified the creators of both MediaWiki pages involved, as well as tossed notices on the talk pages of both of the pages, and notified WP:VPT. EggRoll97 (talk) 00:00, 23 December 2023 (UTC)
 * ✅. -- zzuuzz (talk) 23:07, 23 December 2023 (UTC)

EFM attention to 3 mailing list threads?
If there's any EFMs who have a spare few minutes, I've proposed multiple filter tweaks here, here, and here. I'm not able to implement these myself, of course, since I'm not an EFM, but I've tested all three of them against the original false positive and they seem to work fine. EggRoll97 (talk) 03:51, 24 December 2023 (UTC)


 * Thanks, answered all three, and made two changes. Suffusion of Yellow (talk) 23:27, 25 December 2023 (UTC)

Edit filter helper request for Philipnelson99
Hi everyone, I'm self-nominating myself for EFH. I realize I do not have many contributions to WP:EFFP, but I have started to work there recently. My main reason for requesting this is that I have a thorough understanding of regular expressions due to my technical background and I believe that makes me a good candidate for EFH. I have done quite a bit of antivandalism work and think that I could aid in honing filters to be more accurate including the private filters. Speaking of the private filters, I don't intend to discuss them outside of the context of the mailing list and other EFHs and EFMs. Thanks for considering me to be an EFH. Philipnelson99 (talk) 22:22, 24 December 2023 (UTC)


 * Support - one of our most active anti-vandalism patrollers, who is active in helping to combat LTAs. I've spoken to them off-wiki and I fully trust them to use this permission wisely. Their experience with regex will also be an asset. — Ingenuity (talk • contribs) 22:52, 24 December 2023 (UTC)
 * Oppose You have one request for a new edit filter (which isn't a bad thing, it's good you've requested one) that doesn't propose any possible filter regex to include, and you haven't, as far as I can tell, given any potential fixes for false positives on filters, which means I can't, in my opinion, substitute any knowledge you possess of regex for your lack of demonstrated experience. In the absence of filter tweaking proposals or requested edit filters, generally requests here require either being a sysop on another wiki interested in learning from our filters, or around five hundred edits to the false positives page, over the course of multiple months, to give you a perspective of the type of time and experience needed before EFH is considered. You have 25 edits to the false positive page over around a single month. This is a good start, but insufficient. The level of trust for EFH is generally speaking on par with sysop, far higher than almost every other unbundled userright. Unfortunate as I am to say so, you're requesting this far too soon. Having gotten EFH and seen the private filters, there's very good reason we're all quite cautious with this right, so please don't take it as a personal condemnation, just a judgement that you probably aren't quite ready for this yet. Support based on discussion below. EggRoll97 (talk) 23:09, 24 December 2023 (UTC)
 * I understand your objections. As far as the lack of edit filters I've requested, I haven't suggested any changes or modifications primarily because I have not seen any that needed a modification reported at WP:EFFP that are not private and not already handled. Additionally, I've privately requested changes directly to EFMs. As far as trustworthiness goes, I've made over 40,000 edits to Wikipedia over a period of around 10 years primarily doing antivandalism work. Philipnelson99 (talk) 00:27, 25 December 2023 (UTC)
 * Looking through the mailing list archives and searching for you, I see one other requested filter modification. I could be missing some since you mention you requested more directly to EFMs. 40,000 edits of antivandalism over 10 years is great work, but anti-vandalism doesn't prove trust with filters. Regardless, I'm willing to support based on a requested one on the mailing list. EggRoll97 (talk) 01:14, 25 December 2023 (UTC)
 * They have over 40k edits, including 883 at AIV, and no blocks. What does a bunch of edits calling out (usually) obvious vandalism at EFFP, a page that's already well-patrolled, have to do with trust? Uhai (talk) 00:30, 25 December 2023 (UTC)
 * Anti-vandalism doesn't generally require the private filters, and edit count isn't really a factor. Experience either at EFFP or EFR shows at least some competency with the public filters before moving onto the private ones. EggRoll97 (talk) 01:14, 25 December 2023 (UTC)
 * EFHs cannot modify filters so the only application of trust in this scenario has to do with the user not leaking the details of private filters, for which edit count can definitely be used as a factor for determining. If you are concerned about their usefulness as an EFH given lack of participation in relevant areas, no demonstrable regex experience, or that they do not have a strong need for the permission, that's fair, but concerns over trust do not apply when the user has such aforementioned qualifications (edit count, significant anti-vandalism experience, and no blocks) and have already indicated in their self-nomination they understand the importance of not leaking private filter details.
 * And I wholeheartedly disagree with you: users with significant anti-vandalism experience are great assets for working on private filters as they have an understanding of abuse patterns and LTAs. I would be supportive of someone like this for EFH with zero regex experience and near-zero EF participation just for the expertise they could bring to private EF discussions. Uhai (talk) 22:32, 26 December 2023 (UTC)
 * This was closed as early recently – would an uninvolved admin please grant and close this? Thanks. – 64andtim  ( talk ) 22:24, 27 December 2023 (UTC)
 * Support Trusted user, has need. Uhai (talk) 00:38, 25 December 2023 (UTC)
 * Support I haven't seen you acquiring any unnecessary rights or making requests for them. Although you have not handled sufficient requests here, your efforts at AIV are enough for me, and that's why I'm happy to support.– <b style="color:black; font-family: Tahoma">DreamRimmer</b> (talk) 00:51, 25 December 2023 (UTC)
 * Support as you're a legitimate, trusted user who can beat vandalism in a nutshell, and have great experience with regex to suggest additions to private filters. I do have faith that you can be trusted to view private data. – 64andtim  ( talk ) 02:50, 25 December 2023 (UTC)


 * Support – A trusted user with no concerns or issues. Toad ette  ( Merry Christmas, and a happy new year ) 08:01, 25 December 2023 (UTC)
 * Support user can be trusted and is active with anti-vandalism --DannyS712 (talk) 00:50, 26 December 2023 (UTC)

Edit filter helper for Zippybonzo
Hi everyone, I know I've not been the most active lately, but I've been active in anti vandalism for nearly 2 years by now, and have been helping out around edit filters for about a year off the top of my head and I'm pretty good with PCRE and I'd like to help out with working on fixes for filters and creating regexes for them and optimising the existing filters. I've got a pretty good understanding of how LTA's work from my experience as a New Page Patroller. I completely understand if you're going to oppose based on my activity because it's not been great but I'm working back into it and I think I would be good with working with filters from experience with regex. Full disclosure: I learnt most of my regex from regex101 and sleepless nights :) -- Zippybonzo &#124; talk  &#124;  contribs  (he&#124;she&#124;they) 17:19, 26 December 2023 (UTC)
 * Weak oppose because while I do appreciate your anti-vandalism activities and your good understanding of regex, your work in the EFFP page appears to be very limited which you made almost 100 responses; for the latter, I believe your EFFP work could use a bit more experience with responses reaching anywhere between the 400 to 500 range. Ironically, I did my first nomination, but I had to withdraw after receiving quite a bit of backlash (see my withdrawn nom above a couple of sections). – 64andtim  ( talk ) 18:54, 26 December 2023 (UTC)
 * In my honest opinion making 500 responses turns into a monotonous task given that I know enough about filters and anti-vandalism that making 500 edits to a page is basically irrelevant given I know how to handle such reports. EFFP is far from everything to do with edit filters and whilst I've never requested a filter being that I've never had a need to, I do know how to write the regexes to make a filter work efficiently. Zippybonzo &#124; talk  &#124;  contribs  (he&#124;she&#124;they) 09:50, 27 December 2023 (UTC)
 * Oppose regretfully, because I think you've got the understanding needed, but I, like others before me, consider each request with a heavy standard of demonstrated need, and I'd put the experience needed at EFFPR around 500 or so edits with no problems, or high-quality contributions to existing/new filters, especially private filters. EggRoll97 (talk) 23:20, 26 December 2023 (UTC)
 * Support Zippy really knows his stuff, and honestly, I haven't noticed any issues. His 96 responses at EFFPR give me complete confidence in supporting him. – <b style="color:black; font-family: Tahoma">DreamRimmer</b> (talk) 04:23, 27 December 2023 (UTC)


 * Support: I have been following Bonzo since 2022, and I personally like his contributions. I feel that he can be trusted with EFH. Best of luck . Maliner (talk) 05:19, 27 December 2023 (UTC)
 * Oppose. Aside from their last edit to EFFPR being over a month ago, their comment at this report doesn't inspire a lot of confidence that they understand the purpose of edit filters fully. I'm not seeing a lot of demonstrated need for the user right and please come back after you've helped process some non-trivial reports (ideally showing your knowledge of regex and understanding of edit filters better). <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 10:01, 27 December 2023 (UTC)
 * I understand the purpose of edit filters perfectly fine I just worded it in a particular way rather than repeating exactly what was said above. Zippybonzo &#124; talk  &#124;  contribs  (he&#124;she&#124;they) 10:20, 27 December 2023 (UTC)
 * The demonstrated need is more for helping out with working on private and public filters and false positive reports. Zippybonzo &#124; talk  &#124;  contribs  (he&#124;she&#124;they) 10:25, 27 December 2023 (UTC)
 * That report is probably a nit pick, but I think the proxy filter is intentional in catching it for everyone, so your suggestion for only flagging people with <100 edits is a bit misdirected. But that plus the low activity levels and not a lot of record of helping with authoring filters is a concern for me. I really appreciate the intention to help, but I don't think having EFH is necessary for you to continue helping in this area. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 11:07, 27 December 2023 (UTC)
 * Weak opppose per Deadbeef’s comments. Low recent involvement and below the status quo of ~500 EFFP reports handled. My advice would be that if this is a right you feel you need, become more active within the process to show you really do need it. - 🔥𝑰𝒍𝒍𝒖𝒔𝒊𝒐𝒏 𝑭𝒍𝒂𝒎𝒆 (𝒕𝒂𝒍𝒌)🔥 12:58, 27 December 2023 (UTC)


 * Oppose, per Deadbeef. There's simply not enough involvement and it comes off as hat collecting to me. A user of Zippy's experience should be able to recognize the lack of a need in this instance. Also, I believe the NPP experience is overstated in this instance based on my experience. Hey man im josh (talk) 14:04, 27 December 2023 (UTC)
 * Where I said, 'I know I've not been the most active lately' that implied at EF too, as for the need, the need is not fully demonstrated yet, however I always intended for EFH to be a step up towards becoming an EFM rather than jumping in at the deep end so to speak. Zippybonzo &#124; talk  &#124;  contribs  (he&#124;she&#124;they) 14:13, 27 December 2023 (UTC)

bot triggering filters
Hello. I'm currently on mobile, and my bot is archiving entries of Template:In the news. The template is currently fully/admin-protected for a reason. When I saw the page In the news/Posted/November 2004, it was (still is) full of profanity, and objectionable images. I will clean-up these pages once I am back on computer, which might be on Monday. I tried to see if the bot is tripping/triggering edit filters, as there is an unexpected delay between bot's edits. But as I am on mobile, I couldn't check it at all. Would someone kindly help me? User:KiranBOT. Thanks a lot in advance. —usernamekiran (talk) 19:18, 30 December 2023 (UTC)
 * (correctly) blocked the bot for adding older vandalism/profanity, but I would still like to have answer to my questions. —usernamekiran (talk) 19:24, 30 December 2023 (UTC)


 * I just removed the most egregious stuff from that page. Definitely a "WTF??" Moment for me. Taking Out The Trash (talk) 19:28, 30 December 2023 (UTC)
 * No edit filter trips. Bots are generally exempt from almost all the edit filters, since they're generally only performing community-approved tasks (meaning they have no need to be constrained by edit filters). The bot, as far as I can tell, is actually working as intended, just without a filter? The bot seems to be trying to archive this edit though, from an "AppleWorks hacker" trying to deface the Main Page. See here. Also see here (warning: extremely NSFW). The reason this got through was that the original edits to the ITN template were from when the template wasn't actually admin-protected, from back in the olden days of 2004. EggRoll97 (talk) 08:57, 31 December 2023 (UTC)
 * thanks guys. while I was testing the bot, I tested it thoroughly on the months February, March, and April of 2004, and then on "current month". They had very mild form of vandalism, I was not expecting such level of vandalism. I am currently creating the archives in txt files, I will remove the vandalism, and lint errors that I can, and then I will upload the files as archive pages. courtesy ping to . —usernamekiran (talk) 09:13, 31 December 2023 (UTC)

Edit filter manager for DannyS712


A while back some of us encouraged DannyS712 to request EFM, and he has finally taken up the offer. Danny is a volunteer MediaWiki developer, an admin on multiple other projects, and has made hundreds of edit filter modifications on meta and elsewhere. I have lately been in the habit of copying his suggestions at WP:EFFPR with minimal testing, and fully trust him with this right. Suffusion of Yellow (talk) 22:22, 8 January 2024 (UTC)


 * Support as he's a great asset to configure the filters that stop the vandals and LTAs, and is also a trusted user on other projects, such as Wikidata. – 64andtim  ( talk ) 22:32, 8 January 2024 (UTC)
 * Support, trusted user, easy support, no concerns from me. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 23:19, 8 January 2024 (UTC)
 * Support – <b style="color:black; font-family: Tahoma">DreamRimmer</b> (talk) 00:58, 9 January 2024 (UTC)
 * Support because they are a trusted and experienced user – PharyngealImplosive7 (talk) 01:49, 9 January 2024 (UTC)


 * Umm, I don't really like the idea of trying to foist this upon another user that hasn't requested it. At the very least I'd want to see DannyS712 accept this "nomination". —  xaosflux  Talk 01:24, 9 January 2024 (UTC)
 * I probably should have mentioned that we discussed this by email., you might want to acknowledge this in a way more explicit than this, though. Suffusion of Yellow (talk) 01:40, 9 January 2024 (UTC)
 * @Xaosflux Sorry I didn't note it earlier - I accept the nomination --DannyS712 (talk) 04:35, 9 January 2024 (UTC)
 * Support They clearly have the knowledge, experience, skills required and are trusted to do good with this right. Nobody  ( talk ) 08:55, 9 January 2024 (UTC)
 * Easy support looked over their wikidata, specious and commons edit. Trusted and competent user ~ 🦝 Shushugah (he/him • talk) 10:33, 9 January 2024 (UTC)
 * Support Easy support. We need more EFMs, and Danny clearly far surpasses what I would consider the requisite experience and trust necessary. EggRoll97 (talk) 22:37, 9 January 2024 (UTC)


 * Support Philipnelson99 (talk) 02:28, 10 January 2024 (UTC)
 * Support. Has plenty of experience with edit filters outside of enwiki. Would make a great EFM here.—*Fehufangą (✉ Talk · ✎ Contribs) 03:39, 10 January 2024 (UTC)

EFH/EFM Needed
There are 5 unanswered private reports on WP:EFFPR. Someone (who can see them of course) should answer them. –  Pharyngeal Implosive7  (talk)  12:29, 24 January 2024 (UTC)


 * Looks like they have all been resolved - in the future I suggest waiting until the reports have been waiting for at least 24 hours, EFHs and EFMs actively monitor that page but might not be online at the exact time of the report. Thanks, --DannyS712 (talk) 17:07, 24 January 2024 (UTC)
 * Ok. I'll wait from now onwards. –  Pharyngeal Implosive7  (talk)  17:17, 24 January 2024 (UTC)

Adjusting filter 1,248
I've seen Special:AbuseFilter/1248 get triggered by edits like Special:AbuseLog/36791674 (which fixes the table). I was wondering if it might be worth adjusting it to avoid  and   changes (or maybe separating it to its own filter). Anyone have any thoughts on that? &#8211;<span style="font-family:CG Times, times"> MJL &thinsp;‐Talk‐☖ 17:59, 16 January 2024 (UTC)


 * Seems like a good idea to me. –  Pharyngeal Implosive7  (talk)  22:11, 16 January 2024 (UTC)
 * No objections here. –  64andtim  ( talk ) 22:49, 16 January 2024 (UTC)
 * This has been open for quite a few days with no objections. Would an EFM care to implement this or raise any problems with this filter change? –  Pharyngeal Implosive7  (talk)  17:06, 21 January 2024 (UTC)
 * If there is a specific suggested edit I can try it out, otherwise we need to wait for someone who can also design the exclusion from the filter --DannyS712 (talk) 17:09, 24 January 2024 (UTC)
 * Seems like !(new_wikitext irlike "rowspan|colspan") could work just added as a new line based on some quick examination I've done on the filter log mentioned. (Would also need to add an & at the end of line 10). EggRoll97 (talk) 21:48, 24 January 2024 (UTC)
 * That sounds like that would work pretty well. –  Pharyngeal Implosive7  (talk)  22:23, 24 January 2024 (UTC)
 * ✅, see Special:AbuseFilter/history/1248/diff/prev/30861. Just checking  would fail to flag any numerical change to any article that uses colspan or rowspan anywhere. But the same   hack can be used to detect changes to only the colspan/rowspan. This should reduce the number of hits by about 3%. Suffusion of Yellow (talk) 02:21, 25 January 2024 (UTC)

New script: User:DannyS712/AbuseFilterMessageLinks.js
I just created a new user script, User:DannyS712/AbuseFilterMessageLinks.js, that adds links to the current `warning` and `disallow` system messages (that get updated when you change the selection/input) so that its easier to navigate to the message to make changes/edit requests. Hope you find this useful! --DannyS712 (talk) 09:07, 25 January 2024 (UTC)

Rule 1286: German company slander
Hi!

I propose to activate Special:AbuseFilter/1286 which I imported from dewiki a few minutes ago. Reason is crosswiki spamming, see (German) discussion at de:WP:Administratoren/Anfragen.

Examples for spamming at enwiki: Global filter rule: If there no objections, I'd activate the rule in ~24h. -- seth (talk) 00:25, 3 January 2024 (UTC)
 * Special:Contributions/2001:9E8:600::/42
 * Special:Contributions/2a02:3036:262::/42
 * Special:Contributions/89.247.160.0/20
 * m:Special:AbuseFilter/335
 * If you are subscribed to the edit filter mailing list, please see here. If not, please let me know, and I can forward you the email. Not a massive concern, but given it's a private filter that's being jumped straight to disallow, might be best that it be ironed out now. Also, not sure about the global rule, since I don't have global abuse filter helper, but I presume it's the same as (or similar to) the content in 1286? EggRoll97 (talk) 06:49, 3 January 2024 (UTC)
 * For local EFH/M: yes, the global filter is identical. —*Fehufangą (✉ Talk · ✎ Contribs) 07:00, 3 January 2024 (UTC)
 * Finally, a new filter to stop some kind of LTA that targets random pages and says this type of German nonsense. Support. – 64andtim  ( talk ) 07:37, 3 January 2024 (UTC)
 * Hi!
 * I'm not yet subscribed to the mailing list. I've sent a request right now.
 * The global rule is not fully identical. I imported the dewiki rule de:Special:AbuseFilter/407 which potentially blocks more edits (because of less conditions) than the global rule.
 * -- seth (talk) 10:28, 3 January 2024 (UTC)


 * @EggRoll97: Today I got an answer to my request:
 * 'Your request to the wikipedia-en-editfilters@lists.wikimedia.org
 * mailing list
 * Subscription request
 * has been rejected by the list moderator. The moderator gave the
 * following reason for rejecting your request:
 * "[No reason given]"'
 * What shall I do?
 * -- seth (talk) 18:00, 5 January 2024 (UTC)
 * I had forgotten to respond back on this noticeboard, my apologies. Fehufanga cleared it up on the mailing list that I was misunderstanding part of the filter, though I do have a different question that I sent through on the mailing list. It's not urgent, though, so I don't see any reason not to enable the filter unless there's other objections. As for the mailing list, the two list admins are Samwalton9 and MusikAnimal, so if you drop a message on their talk page they should be able to set you up. Not sure why it was rejected, though, given that you are a sysop. EggRoll97 (talk) 18:05, 5 January 2024 (UTC)
 * @Lustiger seth We get a number of no-context subscription requests with no good way to figure out who the filer is, so it looks like we assumed yours was a spam/ineligible request to join. Could you request again and email me to confirm your email address? That way we can confirm you're you and approve your request. Sam Walton (talk) 18:19, 5 January 2024 (UTC)
 * Thanks, works! :-)
 * And i activated the rule today. -- seth (talk) 00:02, 6 January 2024 (UTC)
 * -- seth (talk) 18:00, 5 January 2024 (UTC)
 * I had forgotten to respond back on this noticeboard, my apologies. Fehufanga cleared it up on the mailing list that I was misunderstanding part of the filter, though I do have a different question that I sent through on the mailing list. It's not urgent, though, so I don't see any reason not to enable the filter unless there's other objections. As for the mailing list, the two list admins are Samwalton9 and MusikAnimal, so if you drop a message on their talk page they should be able to set you up. Not sure why it was rejected, though, given that you are a sysop. EggRoll97 (talk) 18:05, 5 January 2024 (UTC)
 * @Lustiger seth We get a number of no-context subscription requests with no good way to figure out who the filer is, so it looks like we assumed yours was a spam/ineligible request to join. Could you request again and email me to confirm your email address? That way we can confirm you're you and approve your request. Sam Walton (talk) 18:19, 5 January 2024 (UTC)
 * Thanks, works! :-)
 * And i activated the rule today. -- seth (talk) 00:02, 6 January 2024 (UTC)

This filter has had zero hits since it was enabled. Do we really need a filter for this? * Pppery * <sub style="color:#800000">it has begun... 19:20, 25 January 2024 (UTC)


 * Indeed. The global and the german rule had no further triggering edits for 3 weeks now, too. I'll deactivate the rules. -- seth (talk) 20:38, 25 January 2024 (UTC)
 * ✅. -- seth (talk) 20:43, 25 January 2024 (UTC)

Merge filter 874 and filter 1196 into filter 102
(Moved from Wikipedia talk:Edit filter)

Hi Everyone! While going through and trying to locate a filter that blocks username creation given a list of patterns, I found that filter 102, filter 874, filter 1196 are constructed exactly the same way. The only difference is the name of the variable containing the regex patterns to check for, and the username patterns listed in that variable. The construction, layout, condition statements, actions, template used when a positive hit is blocked, edit filter privacy settings, etc are all exactly the same. It just seems silly to have separate filters with the same purpose, code structure, actions, and settings separated into three filters. I went and took the conditions listed in the username pattern variable on 874 and 1196 and added them to 102.

I know that there are scripts out there (such as bots in IRC and likely scripts around here) that reference one of these other edit filters, and it'll take a bit of effort to modify their code to use filter 102, but I think it's worth doing this so that we can have one filter performing the checks for these conditions instead of three. Essentially, outside of modifying any scripts pointing to 874 and 1196 to now point to 102, we just have to turn off those two filters; the migration part is already done. Can I get input on this? Any thoughts, concerns, or opinions? I think this is a no-brainer, but I wanted to get input in case I missed something important that would cause things to break if we were to move forward with this. Thanks everyone!  ~Oshwah~  (talk) (contribs)   07:14, 19 January 2024 (UTC)


 * Now obviously, LGTM. –  64andtim  ( talk ) 17:44, 19 January 2024 (UTC)
 * No objection in principle, but I'd prefer to see some recognition that some filters use norm, and some user ccnorm, and they're quite different things. It seems clear at first glance that some of the things are written for one function and not the other. -- zzuuzz (talk) 18:03, 19 January 2024 (UTC)
 * Zzuuzz - ccnorm is DA MANNN!!! ;-)  ~Oshwah~  (talk) (contribs)   04:42, 20 January 2024 (UTC)
 * 1196 was split by from 874 during this discussion in 2022 when the filter was a mess and we were trying to sort out the false positives. Sometimes, the only value to multiple filters is just being able to sift through the log. That said, shortly after the split, I did a significant pruning of both filters; no objection to re-merging at least those two. Suffusion of Yellow (talk) 02:04, 25 January 2024 (UTC)

You're now using ccnorm on patterns that were originally intended for norm. I'm getting about 20% false negative rate when I test the "new" 102 against the combined log of 102, 874, and 1196. Suffusion of Yellow (talk) 21:01, 27 January 2024 (UTC)
 * Suffusion of Yellow - I've re-opened this discussion in light of your observations. As of right now, everything is rolled back to the way they were before (874 and 1196 are re-enabled, and 102 is only looking for conditions that it was originally set to). Thanks for the heads up and for letting me know. :-) I'll take a look and begin separating each condition into bucket A or bucket B...  ~Oshwah~  (talk) (contribs)   21:11, 27 January 2024 (UTC)
 * Update: Okay, this should be fixed now with the variable separation and changes I just made to edit filter 102. I've applied the changes on 102, and have disabled 874 and 1196 again in order to test and keep an eye on logs. Suffusion of Yellow, can you help keep an eye on this deployment and let me know if more issues come to light? I thank you again for catching that 102 implemented ccnorm and not norm. I must've obviously missed that when comparing the three edit filters against one another...  ~Oshwah~  (talk) (contribs)   21:22, 27 January 2024 (UTC)
 * Looks good to me; this matches everything from all three filters (after the 2022 refactor). I'll keep an eye out for FPs. Suffusion of Yellow (talk) 21:33, 27 January 2024 (UTC)

Partially Done Templates for WP:EFFPR
Recently, I have come along some reports where I implement some of the actions the user wants and don’t for some other actions. One example is here, where I thought that the user added too much weight with little sources to one artwork, but had an image worth adding. I would suggest the templates would look something like this:  and. If you have any questions or concerns, I would be happy to respond. –  Pharyngeal Implosive7  (talk)  15:27, 27 January 2024 (UTC)


 * Probably looks good to me. –  64andtim  ( talk ) 02:18, 28 January 2024 (UTC)
 * This has been open for a while. Would anyone mind implementing this or continuing the discussion? –  Pharyngeal Implosive7  (talk)  18:44, 1 February 2024 (UTC)
 * While i understand why you want it, I don't think it happens that often that it's own template is needed. The current ones give enough options I believe. If you add most of an edit then you can just use done and write why not all of it and if you add a only a small part of it you can just use either: Not done if a part shouldn't be added, comment if you're not sure if it should be added or request on talk with refs if the part was unsourced. Nobody  ( talk ) 18:53, 1 February 2024 (UTC)
 * Yeah. I understand that it would happen in edge cases only. Maybe it is better to just use not done and comment templates together, but creating a new template is always a possibility. –  Pharyngeal Implosive7  (talk)  18:59, 1 February 2024 (UTC)

Edit filter 602
Hi - I recently notified an editor of a contentious topic, but because I forgot to substitute the alert/first template, it didn't generate a filter log entry. AnomieBOT then substituted the template for me, but because the edit filter is set to exclude bots, this didn't generate a log entry either.

I just wanted to leave this message to check if this is by design, and that everything's working as it should be :)

All the best, &zwj;—&zwj;a smart kitten[ meow] 20:07, 26 January 2024 (UTC)


 * The filter intentionally excludes bots, but warns editors if they want to leave an alert or not. We should also include non-substituted contentious topic alert templates to the regex. – 64andtim  ( talk ) 20:28, 26 January 2024 (UTC)
 * How about

page_namespace == 3 & "confirmed" in user_groups & !( "bot" in user_groups ) & /** do not warn for subpages **/ !( "/" in page_title ) & (   substituted := "-- derived from template:(?:ds|contentious topics)/alert(?:/first)?(?:/DS)? --";    unsubstituted := "{{(?:(?:ds|contentious topics)/)?alert(?:/first)?(?:/DS)?";    comment := substituted + "|" + unsubstituted;

/** cancels out most edits **/ added_lines irlike (comment + "|subst:") & /** now check added_lines_pst, which is the more expensive operation **/ added_lines_pst irlike comment & !( removed_lines irlike comment ) )

as a check for the unsubstituted alerts? --DannyS712 (talk) 20:47, 26 January 2024 (UTC)


 * But also, why do we ignore bots? @AGK you added that in Special:AbuseFilter/history/602/diff/prev/12049 but without any notes, do you remember why? (I realize that its been over 10 years though...) --DannyS712 (talk) 20:50, 26 January 2024 (UTC)
 * Is the  necessary or not since it just warns all human editors? –  64andtim  ( talk ) 21:00, 26 January 2024 (UTC)
 * Wouldn't it then trigger for archival bots? And we wouldn't want to warn bots. Logging is fine I guess but it should log the human who who gave the alert. Galobtter (talk) 04:01, 27 January 2024 (UTC)
 * @Galobtter the filter already excludes subpages which is presumably where the archiving would be done (if you are talking about bots being shown the filter warning). If you are talking about people trying to warn bot accounts, that is already allowed. Not a big deal, not important to remove the check, just figured that I would ask. --DannyS712 (talk) 05:05, 27 January 2024 (UTC)
 * Still, in what case would we want bot getting a filter warning? Galobtter (talk) 05:33, 27 January 2024 (UTC)
 * {{reply|A smart kitten|DannyS712|Galobtter}} Luckily, I still remember the details of creating these templates! We needed the log to clearly show the users who both gave and received the alert. It was unsatisfactory that substitution bots were being recorded as giving alerts, so the filter at the time incorporated {{tl|require substitution}}. An unsubstituted alert would then fail to print the alert text on the user talk page and users were expected to see that, remove the template, and try again. That filter condition was introduced so that, the failed alert having not logged, the substitution bot's edit would not then be logged as an alert. I see that the equivalent current template no longer requires substitution, so I think that the condition should be removed from the filter, ensuring that an alert is never printed on the user's talk page while not being logged due to non-use of substitution. AGK (talk) 12:48, 27 January 2024 (UTC)
 * @AGK okay, if we change the filter to what I suggested the filter will warn both for the substitution and the transclusion. If a bot comes and substitutes the template the filter will not trigger, but if the user themselves realizes the mistake and switches to substitution it'll show up twice. Is that an acceptable situation? --DannyS712 (talk) 22:52, 27 January 2024 (UTC)
 * Just a note - I may have proposed the above text but I'm not comfortable deploying it given how sensitive the filter is - if there are false positives people might get sanctions they didn't know about, so I'd prefer another (more experience) EFM take this on --DannyS712 (talk) 01:34, 2 February 2024 (UTC)

Turn off "give the user a warning" for this filter?
Notified: Template talk:Contentious topics, Wikipedia talk:Arbitration Committee. – Novem Linguae (talk) 12:53, 27 January 2024 (UTC)
 * While we're here, is anyone interested in changing this filter from "warn" to "log only"? The warning is an unnecessary extra step when I'm leaving these. I'm not sure preventing an occasional double warning is worth the hassle of warning every single time someone attempts to place an alert. Thoughts? – Novem Linguae (talk) 00:23, 27 January 2024 (UTC)
 * Maybe in the middle of yes and no, NL. My suggestion above can still work without the  and/or the bot exemption after I debugged a random log entry of filter 602 made by a human editor. Opinions? –  64andtim  ( talk ) 01:12, 27 January 2024 (UTC)
 * The filter still needs to trigger for everyone, for logging purposes. Just to be clear, my suggestion is to not have it display a visible warning that stops the attempted edit. So basically, unticking the box that says "Trigger these actions after giving the user a warning" – Novem Linguae (talk) 01:16, 27 January 2024 (UTC)
 * For that, I think we could use some more input and opinions before implementing such a controversial change to the filter. – 64andtim  ( talk ) 01:33, 27 January 2024 (UTC)
 * Would need to to talk to ArbCom/ArbCom clerks about that. Galobtter (talk) 03:58, 27 January 2024 (UTC)


 * Thanks.
 * The formal definition of "awareness" and the rules around alerts have changed significantly when "discretionary sanctions" became "contentious topics".
 * old:
 * new:.
 * It may be arguable – not speaking for the Committee – that this removes the need for the warning notice. It may, however, also be arguable that a filter warning is more needed than before, to occasionally remind those used to the old system that things have changed and placing {{subst:DS/alert}} on user talk pages is likely not what they wanted to do. ~ ToBeFree (talk) 13:10, 27 January 2024 (UTC)
 * Note that MediaWiki:Abusefilter-warning-CT includes a customized message if the user has contentious topics/aware already on their page. If the warning is disabled for 602, then we should at least have a separate warning filter for pages with this template. Suffusion of Yellow (talk) 03:10, 28 January 2024 (UTC)
 * I quite like the warning message because it brings up the four places that one can check before actually issuing the template. In terms of workflow while placing the template, it's a lot easier to just paste it into a new section, have the warning bring up the four places one can check to see if the warning is unnecessary, and then check those places. My understanding is that CTOP still does care about those places; footnote M (the footnote after but may attempt to refute this presumption on appeal does give a list of per se ways to be presumed aware of a topic outside of receiving a notification, although that presumption is rebuttable.
 * I feel like the result of removing the warning will be duplicate alert messages proliferating more than anything else. It's not strictly necessary to have a warning if we want to build scripts to replace the links in the warning message with things that display on-screen, but I don't think that the benefit in terms of workflow having to click one less button outweighs the cost in workflow of having to go and manually check several places before placing the template. — Red-tailed hawk  (nest) 20:52, 28 January 2024 (UTC)

Filter 102's message modification proposal
For filter (LTA or abusive account username creation, private), I would suggest these changes to its disallow message which slightly makes the "This username is unavailable" text smaller since it appears too big on the mobile version of the English Wikipedia, and it removes the "Report error" button since it is redundant.

In addition, I have decided proposing the removal of the remaining filter2 and filter3 parameters per this discussion below. Courtesy ping to since he merged the remaining LTA username-catching filters' conditions of  and  to 102, as the former two may become disabled later. I believe that this should be implemented when 874 and 1196 become disabled, and/or when there is consensus to do so.

Any inputs, concerns or opinions on this? Thanks. – 64andtim  ( talk ) 22:12, 21 January 2024 (UTC)
 * I don't see any reason to object this proposal.  ~Oshwah~  (talk) (contribs)   16:00, 22 January 2024 (UTC)
 * Seems good to me also. –  Pharyngeal Implosive7  (talk)  21:42, 22 January 2024 (UTC)
 * ✅. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 15:46, 27 January 2024 (UTC)

Impersonations
How on earth did these accounts get created? Don't we have a filter to block impersonation accounts (IIRC it's like 100 or 101 or something). The first one should've been blocked internally by AntiSpoof too. Taking Out The Trash (talk) 18:56, 4 February 2024 (UTC)


 * Might be filter 102 that can prevent such impersonations, and that's all what I'm going to say. Feel free to email the edit filter management list or a local EFM. – 64andtim  ( talk ) 21:25, 4 February 2024 (UTC)


 * Can non-EFH/EFM actually use the mailing list? Last I checked it was closed off and private for only rights holders. Taking Out The Trash (talk) 18:22, 5 February 2024 (UTC)
 * You can, your request gets on hold and needs to be accepted by one of the list owners first. Nobody  ( talk ) 18:40, 5 February 2024 (UTC)
 * Anyone can send to the list, but only admins, EFHs and EFMs can subscribe. I don't know if extra spam-filtering is applied to non-subscribers, though. Suffusion of Yellow (talk) 00:06, 6 February 2024 (UTC)

deprecated source filter
Can this change by be reverted as the closure establishing EI as deprecated was overturned? They havent edited in a few weeks so asking here, though if there is somewhere else to request edit filter changes I would appreciate the pointer. Thanks,  nableezy  - 16:03, 13 February 2024 (UTC)


 * ✅ ProcrastinatingReader (talk) 17:57, 13 February 2024 (UTC)

EFH for PharyngealImplosive7


Hi guys. I am requesting the EFH right today. I meet all criteria mentioned in the guidelines for obtaining this perm and believe that it would help me when answering FP reports on WP:EFFPR. I have edited that page more than 500 times (which seems to be the status quo here) and that can be seen in xtools (I don't have the link currently but I'm sure you guys can search manually). I also have experience with RegEx and the edit filter's code syntax, and have suggested some changes (with RegEx) to various filters including adding  to one of the meme filters. I have also helped with some bad word substitutions and provided code there. I also believe that I am trustworthy enough to not leak the contents of any private filters, and have good account security. Overall, I think that I am a good candidate for EFH, and thank you for reading this proposal. –  Pharyngeal Implosive7  (talk)  01:40, 16 February 2024 (UTC)
 * Support as you're helpful in dealing with false positives. – 64andtim  ( talk ) 02:03, 16 February 2024 (UTC)


 * You started editing this page 40 days ago and have already responded to over 500 reports in this short time.That’s good, I'm curious about your experience with RegEx. Can you: 1. Match a string containing a valid IPv4 address? 2. Match a string containing a valid email address with an optional subdomain? 3. Match the following string: ? – <b style="color:black; font-family: Tahoma">DreamRimmer</b> (talk) 03:06, 16 February 2024 (UTC)
 * Thanks for the conplement. While I am not an expert, I can try to provide solutions for you (hopefully I didn't mess something up with all the typing and copying/pasting).
 * 1. This should match valid IPv4 addresses and has the flags of g and m:.
 * 2. This would be a pain to make. The closest I've got is  but it allows invalid emails like  . It has the flags of g, m, and i.
 * 3.  should work with flags g and m.
 * In all, I do have a moderate understanding of RegEx, but I'm not an expert, and I'm sure I can help author and definitely read filters. –  Pharyngeal Implosive7  (talk)  05:16, 16 February 2024 (UTC)
 * Thanks for the solutions; these are all correct. I am happy to support, as you are well aware of what vandalism is, how filters work, and also have knowledge of regular expressions. – <b style="color:black; font-family: Tahoma">DreamRimmer</b> (talk) 05:31, 16 February 2024 (UTC)
 * I am changing my vote to neutral based on the rationales of the opposing votes. Please don't be discouraged and keep up your good work; you are valuable to us. – <b style="color:black; font-family: Tahoma">DreamRimmer</b> (talk) 16:13, 16 February 2024 (UTC)


 * @PharyngealImplosive7 @DreamRimmer without evaluating the actual request yet, I don't think the first answer is correct - separate from the questionable use of the multiline flag, it doesn't match things like  which I'm pretty sure is valid. The regex needs that any 2xx part of an IP have both of those digits be 0-5 but actually the value just needs to be in the range 20-55, so things like 209 are valid. --DannyS712 (talk) 08:15, 16 February 2024 (UTC)
 * @DannyS712, I agree with you. I know that even though it can match many valid IPv4 addresses, it might also match some incorrect ones or even overlook some correct ones because of its limitations. But honestly, I didn't notice any issues with their work. Overall, their responses were correct. They stated that they have a moderate understanding of regex, and I am satisfied with their solutions. I believe they possess sufficient experience to be considered for an EPH role. – <b style="color:black; font-family: Tahoma">DreamRimmer</b> (talk) 09:17, 16 February 2024 (UTC)
 * Would work?  Nobody  ( talk ) 12:19, 16 February 2024 (UTC)
 * It should work, but it would not match the 201 to 255 range in the fourth segment. – <b style="color:black; font-family: Tahoma">DreamRimmer</b> (talk) 13:06, 16 February 2024 (UTC)
 * Tried to fix it and ended up doing this:  But this just looks to long.  Nobody  ( talk ) 13:24, 16 February 2024 (UTC)
 * Yeah that was incorrect. The 200 part could be fixed to .–  Pharyngeal Implosive7   (talk)  16:20, 16 February 2024 (UTC)
 * Neutral While they do have the usual requirements (and are better at RegEx than me). This, this and this don't make me believe that they've been around for long enough to be ready for that right. Same with this discussion. I'm also not sure I agree with this comment. Nobody  ( talk ) 06:25, 16 February 2024 (UTC)
 * 1. Pinning was fine, it wasn't disruptive.
 * 2. I don't understand the issue. PharyngealImplosive7 didn't respond to a report that was responded to. Perhaps wasn't aware that there is a bug with the script that the script responds to one section above the section you have selected to respond to?
 * 3. we should encourage more coordination and communication instead of "silently figuring out everything on your own", the latter admittedly can be a good quality, but has no relevance to EFH. Asking people should be normal.
 * 4. I don't understand why a reasonable proposal is bad?
 * 5. Approaching with caution makes them more qualified for EFH, not less. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 06:46, 16 February 2024 (UTC)
 * 1. If you know of MajavahBot/EFFP Helper Configuration, then you know that not answered reports stay for 3 days before being archived. Pinning them after just 30 minutes isn't disruptive I agree, but I'd say it's unnecessary. 2. This shows it more clearly, not a script issue, but a edit conflict. 3. While I agree with your take, it doesn't take long to see how much the average respond time for private filter reports is. (I'm assuming that if he knew about MajavahBot/EFFP Helper Configuration he wouldn't have made that section, but that also shows that they haven't been around for long.) 4. That's just my opinion feel free to disagree. 5. Caution is good I agree. But those filter hits don't look like a legitimate editor at all to me. But since you can see those private filter hits. Are the edits promotional or Autobiography as I'm assuming? Nobody  ( talk ) 07:18, 16 February 2024 (UTC)
 * I'm not evaluating this request at the moment but I'm aware of the bug. The diff that is linked on section 2 here shows the issue better. For what it's worth after discussing responding to reports that have been answered with  this hasn't been an issue. Additionally, they haven't pinned anything hastily after I asked them not to and explained how  works. Philipnelson99 (talk) 13:06, 16 February 2024 (UTC)
 * There was a time when EFH was scarcely granted at this board. Most EFH users were SPI clerks or EFMs who gave up that right, and (if I recall correctly...) candidates seeking it for other reasons were often suggested to go to RfA, as the level of trust was considered similar. (I can't quite remember if that was just EFM or EFH as well, I think both.) In particular, activity at EFFP was generally not considered to be sufficient for EFH. I think the bar was perhaps too high then, but I'd be cautious of it swinging in the other direction too much, and EFH being granted like the WP:PERM rights. I think your help at EFFP is useful and immensely appreciated, but I'm not sure I can support especilly on the back of ~4 months of effective activity. ProcrastinatingReader (talk) 12:01, 16 February 2024 (UTC)
 * Oppose per PR. I'm afraid that only a few months of consistent activity isn't enough for such a sensitive permission. — Ingenuity (talk • contribs) 13:36, 16 February 2024 (UTC)
 * Oppose 8 month old account with 4 months of recent activity. Less than 3000 total edits across the wiki. Only 13 edits to AIV, 0 to UAA, and 5 to SPI case pages. I'm not calling the requester an LTA, but if an LTA were to register an account and work towards the EFH right, it would look like this. 508 edits to EFFP in such a short time and creating this request immediately after passing the 500 edit threshold that some users (not I) consider as a or the requirement for EFH is not a good look.
 * I've said it before and I'll say it again: a bunch of edits calling out usually obvious vandalism at EFFP is not a suitable criterion for EFH eligibility. It's unclear what the requester will bring to the table with this role as they don't need it to continue helping with public filters at EFFP. Additionally, their answers to regex questions #1 and #2 are incorrect, and #2 in particular shows a lack of consideration for which are valid vs. invalid characters in an e-mail address.
 * Granting this request at this time would be extremely reckless. I hope the requester understands the utmost importance of trust for this role and I thank them for their contributions and help at EFFP and would encourage them to keep it up and to make a request again no sooner than 1 to 2 years of activity to eliminate concerns over the security of the private filters. Uhai (talk) 14:10, 16 February 2024 (UTC)


 * Oppose per and  who both have legitimate reasons for not supporting. Philipnelson99 (talk) 14:34, 16 February 2024 (UTC)
 * Oppose Uhai sums up my concerns fairly well --DannyS712 (talk) 14:52, 16 February 2024 (UTC)
 * Oppose unfortunately, per everybody else. I appreciate your help at EFFPR, but right now you might be a little inexperienced at the moment. As EggRoll97 said, "abuse of edit filters are virtually undetectable". Also, responding to reports that only involve private filters could potentially mean that you act reckless or you don't understand the meaning of EFFPR, and that's what happened to my first (and failed, yet withdrawn) nomination; I have learned that the hard way. In addition to the EFFPR response count, consider requesting this again in a year or two to demonstrate that you really need this highly-sensitive right. – 64andtim  ( talk ) 15:48, 16 February 2024 (UTC)
 * I understand everyone's concerns and am withdrawing this request. This has been a learning experience for me and I may request this perm in the future when I'm more experienced and seem more trustworthy. Thank you all again. –  Pharyngeal Implosive7  (talk)  16:23, 16 February 2024 (UTC)

EFH for 64andtim

 * (Previous request)

Hello! For those of you who don't know him, is very active around edit filters in many aspects. He has responded to around 670 WP:EFFPR reports so far and has helped implement various changes to filters such as here and here. I believe that he would be a good asset to the edit filter team with EFH. They are quite familiar with various LTAs and have been a target of one (ex. ) in particular for a while. I believe he has proven he can be trusted with private data and the security of private filters. v/r - Seawolf35 <sup style="font-family: Times New Roman; color: #006400;">T--C 15:49, 16 February 2024 (UTC)

Candidate: Please indicate acceptance here: First, I have written a formal apology below.

''For those who have opposed my previous nomination, I am writing this apology to accept my mistakes in which I have responded to reports that only involve private filters, and that was not a good idea. I apologize for being so reckless and irresponsible at EFFPR, and I will not do that again like how a user said that I have acted so reckless/didn't understand the meaning of EFFPR, and I will change my ways and note this for future references. Please forgive me and provide me another chance to prove that I can be trusted with this very-trusted role. I shall be very kind and grateful for your consideration. Thank you.''

As per above, I hereby apologize to those who have opposed my first nomination for the edit filter helper role, and I accept this nomination (this was discussed via email). Thanks. – 64andtim  ( talk ) 16:01, 16 February 2024 (UTC)
 * Weak support again per the last request. Help is limited if no technical understanding of regular expressions or intention to help with authoring filters. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 16:06, 16 February 2024 (UTC)
 * I know, but on the other hand, I do have a basic understanding of the edit filter syntax language. – 64andtim  ( talk ) 16:08, 16 February 2024 (UTC)
 * Weak Support: You have put in lots of hard work and effort into responding to false positives. I'm willing to give you a second chance and I'm pretty sure that you didn't mean to be reckless. However, I agree with in that you have limited technical experience.–  Pharyngeal Implosive7   (talk)  16:57, 16 February 2024 (UTC)
 * Oppose. I am very unsure about a couple of things. The first is your early editing history; you created an alternate account (64andsomeone) after making only two edits, which is very unusual. Immediately upon becoming autoconfirmed you installed RedWarn. The second thing is that I do not feel that you understand WP:DENY. User:64andtim/List of the most wanted users I know is a very bad idea; it makes you a huge target for LTAs. Jamie's comments here are very relevant. This is a very high-trust permission - any abuse would be undetectable - and if you don't plan on authoring filters I do not think it is worth the risk to grant this permission. — Ingenuity (talk • contribs) 17:23, 16 February 2024 (UTC)
 * I understand why you believe making the alt account was very suspicious, but is it really a problem? The OP wasn't socking or anything like that. Also, why would installing redwarn immediately after getting autoconfirmed be bad? It seems that they just fight vandalism and have been doing that since the start of their account creation. I do agree that they don't understand WP:DENY though. –  Pharyngeal Implosive7  (talk)  19:02, 16 February 2024 (UTC)
 * Oppose I don't see the nominee's need for the role, especially since they don't intend to step up to EFM. My bigger concern is how often I've seen them respond to obviously frivolous reports, including Special:Diff/1208268011 7 hours after this request was made. Just check the filter log then revert; don't give them the satisfaction of receiving a response to the report. This goes along with the WP:DENY concern already raised. Edit count is okay, but the account is less than a year old and I share Ingenuity's concerns regarding risk/reward in granting this role: there's little potential benefit and significant potential harm. It's also only been two months since the last request—if this request also fails, I highly recommend waiting a year before requesting again and coming up with a more concrete explanation for requiring the role. Perhaps reconsider the stance on EFM and get involved in technical work in the meantime (and not necessarily just with the abuse filter). Uhai (talk) 02:16, 17 February 2024 (UTC)
 * Neutral, with the good being the activity on WP:EFFPR which seems okay recently (I have no issue with the handling of the diff linked above) as well as satisfactory communication skills, and the bad being a thorough misunderstanding of WP:DENY (with Special:Undelete/User:64andtim/How to deal with a specific LTA § For administrators being particularly absurd) and accepting this nomination a bit too soon following the previous one with an odd statement promising to do better in the future. To stand on my soapbox for a bit, I don't understand why commenters expect the requestor to be EFMs-in-waiting – the vast majority of EFFPRs for private filters are one that could be understood with no RegEx knowledge whatsoever. Most of those that can't, can be understood using User:Suffusion of Yellow/FilterDebugger, and the rest can be deferred to those that do understand RegEx. The first two questions in the section above are things that go way beyond a basic understanding of regular expressions—which is at Edit filter helper when not intending to author filters—and certainly not something I'd want to pull out of a hat without the use of Google and a debugger. DatGuyTalkContribs 02:45, 17 February 2024 (UTC)
 * I am hereby withdrawing this, again, because I now understand your major concerns about me, and also the fact that I regretted making those essays that do not deny recognition. In fact, I'm not sure if I can come up with a more solid response to request this sensitive right, nor am I sure that I might try again next year, but I might or might not do so if I'm on the leaning side of being trusted. Thank you for your comments. – 64andtim  ( talk ) 03:39, 17 February 2024 (UTC)


 * Some points I want to raise after the closure: I don't understand why commenters expect the requestor to be EFMs-in-waiting. I think our backlog is healthy and kept low with the amount of EFHs and EFMs that we have. So having another EFH would not give us a lot of benefits unless they want to help with authoring filters. (which having EFH could help them demonstrate their technical contributions better I suppose) Other EFMs have also suggested that EFH doesn't make much sense and we should just tell people to request EFM directly, because sometimes we apply a higher requirement for EFH candidates than admin candidates, and unbundling it would just make EFM requests harder making EFH a prerequisite. As an aside, I am curious what the quotes in the summary of the withdrawal mean. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 04:01, 17 February 2024 (UTC)
 * Interestingly, one of the three most recent grantees of EFH has been essentially MIA at EFFP since receiving it, and things don't seem worse for the wear.
 * And the "'major' concerns" comment rubs me the wrong way and makes me question the nominee's judgment and ability to accept criticism. An account less than a year old requesting arguably the third most sensitive permission after EFM and sysop is a major concern. If you were to go to RFA with an account 9 months old it would be a WP:SNOW early close, WP:NOTNOW, WP:NQY, yadda yadda. The diff I linked was perhaps a little nitpicky, however, as I said, was one of multiple instances that I noticed, including where the user felt the need to message me about reverting a frivolous false positive report that had "nigro" in the section header, presumably a report they were about to respond to. Uhai (talk) 14:19, 17 February 2024 (UTC)
 * Agreed on the comment, it rubs me the wrong way too. I was curious about myself and another request, so see quarry:query/80483. I don't know though, is EFH being granted really correlated with increased involvement with edit filters? Perhaps conferring the title boosted confidence? <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 15:39, 17 February 2024 (UTC)
 * fwiw, for me it wasn't a confidence boost to get EFH. I simply wanted EFH because I was interested in helping with the private filters specifically. I've coordinated a lot with Ingenuity directly about the private filters since receiving EFH. Eventually I will go for EFM. But I think my bump in responses at EFFPR is also correlated with an increase in editing overall.
 * See my xtools link here.
 * Philipnelson99 (talk) 15:57, 17 February 2024 (UTC)
 * About that, I apologize if I was unable to accept criticism or about my judgment in question, I think this was also not a good idea doing a nomination when my account is less than a year old. About these "major concerns", I had to address that I didn't understand much about WP:DENY, and that I'm probably not trustworthy yet doing such a nomination too soon for this right. – 64andtim  ( talk ) 16:28, 17 February 2024 (UTC)
 * This is the second time you are using quotes. They may be interpreted as scare quotes. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 16:37, 17 February 2024 (UTC)
 * I think we should assume good faith and encourage 64andtim, as I believe they are trying their best to acknowledge their mistakes and improve. – <b style="color:black; font-family: Tahoma">DreamRimmer</b> (talk) 16:45, 17 February 2024 (UTC)
 * well yes, this is just to be up front about communication. I am assuming good faith that perhaps 64andtim does not know of the potential negative connotations of using quotes around the word "major", then the phrase "major concerns". <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 16:48, 17 February 2024 (UTC)
 * I understand now that I will not use such scare quotes next time, and I too am assuming good faith to everyone. – 64andtim  ( talk ) 17:08, 17 February 2024 (UTC)
 * Not trying to be a downer but stating a fact, AGF generally turns into CIR or NOTHERE eventually. Zippybonzo &#124; talk  &#124;  contribs  (they/them) 02:59, 18 February 2024 (UTC)

Does the private edit filter wiki exist?
I was looking through the archives of this page and I found a discussion about an edit filter private wiki (see : the discussion is quite interesting). While I know that I wouldn't be able to participate, did this wiki ever happen, or did we just stick with the mailing list, because it seems that the archived discussion never reached a conclusion? –  Pharyngeal Implosive7  (talk)  21:01, 6 February 2024 (UTC)


 * I think we're stuck to the mailing list for now, and we're gonna have to restrict it to admins, EFHs and EFMs for security purposes. So it looks like we (both) cannot join until we have the EFH role at the very least. – 64andtim  ( talk ) 21:23, 6 February 2024 (UTC)
 * I know, but I was just curious about its existence. I know that we both can't get it yet, but if one of us gets EFH at some point, we could be in it, and it could be useful anyhow. –  Pharyngeal Implosive7  (talk)  23:40, 6 February 2024 (UTC)
 * I don't see anything that looks like "edit" or "filter" on the private wiki list. – Novem Linguae (talk) 01:26, 7 February 2024 (UTC)
 * I didn't know that there was a place you could see all private wikimedia wikis. That answers my question. –  Pharyngeal Implosive7  (talk)  01:37, 7 February 2024 (UTC)
 * This does not exist. I'd be in favor of such a thing, but given the overwhelmingly negative reaction to this similar proposal, it looks like it would be an uphill battle to get it created. Suffusion of Yellow (talk) 02:42, 7 February 2024 (UTC)
 * It might still be a good idea to try, as others have pointed out before: the mailing list is generally inactive, and organization might be easier on a private wiki. It seems to me that it would be a good idea, even though I can't join. –  Pharyngeal Implosive7  (talk)  03:27, 7 February 2024 (UTC)
 * Is the mailing list inactive though? It's had 4 threads (10 messages) in the last month. – Novem Linguae (talk) 04:15, 7 February 2024 (UTC)
 * It isn't possible for me to know whether it is active currently (I'm not in it nor can I be in it even if I wanted to) but people in the previous thread said it was generally inactive. –  Pharyngeal Implosive7  (talk)  04:27, 7 February 2024 (UTC)
 * Not that it is inactive, but it is just hard to track. There are requests, concerns, and issues that can go unanswered for months, which would be much better suited if we had an actual ticket system, which the private wiki could help with. (at least gives you an idea of what issues were resolved and what were not)
 * In terms of personal preference, we are already used to the threaded discussion format here on Wikipedia. We don't use mailing lists to handle WP:EFFPR requests do we? Making it a wiki makes it a better tier of communication.
 * The need is clear. EFMs/EFHs/Admins need a way to communicate about LTAs, vandals, and contents of private edit filters. This is currently done through a non-ideal form of communication (mailing list, private messages, IRC, etc.) that doesn't get the job done as efficient as a private wiki. Sending an email to a mailing list requires considerable effort, and creates friction that potentially ward off seemingly trivial/simple questions that can otherwise turn into a useful discussion.
 * Moving the activity on a mailing list to a private wiki is just better for effectiveness in handling matters related to private filters and LTAs. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 15:24, 14 February 2024 (UTC)
 * Could a discord server also work? I know it's limited what you can do there, but probably easier to setup then a wiki. Nobody  ( talk ) 15:35, 14 February 2024 (UTC)
 * But this would be more than a struggle for us to propose this on Meta, and have our time wasted if most of that community simply rejects that proposal.
 * Also, what about the thing that you send an email to the edit filters mailing list and you have to wait maybe days or months until your request gets accepted. Maybe we need a better way to "accept" emails from non-list/wiki members if their requests are related to creations of/additions to private or LTA filters. – 64andtim  ( talk ) 16:45, 14 February 2024 (UTC)
 * Honestly, this proposal has its pros and cons. However, it seems to me that most people who actively deal with edit filters here are onboard. If we're ready, I would be ok with supporting a proposal for a private edit filter wiki on meta. –  Pharyngeal Implosive7  (talk)  21:39, 14 February 2024 (UTC)
 * I agree it is hard to track. I am not going to commit to writing an RfC for this yet. But if no one does in the next week or so I'll try to. Haven't written an RfC before. Philipnelson99 (talk) 15:02, 16 February 2024 (UTC)
 * Huh. Maybe I spoke too soon when I said "uphill battle". There apparently is a private wiki for Italian admins, at https://sysop-it.wikipedia.org. This was created with minimal fuss at T256545 after it:Wikipedia:Bar/Discussioni/Coordinamento delle informazioni sui vandalismi Was that ever even discussed on meta? Are we allowed to create a private wiki if there's only local consensus? Suffusion of Yellow (talk) 20:31, 15 February 2024 (UTC)
 * No idea if we have to discuss on meta first, but it seems unlikely that we have to discuss if that sysop wiki could form without any discussion. If that's the case, I would be all for a phab ticket. –  Pharyngeal Implosive7  (talk)  20:57, 15 February 2024 (UTC)
 * We'd definitely need local consensus for creating such a private wiki. Anyone willing to draft a Rfc? I can maybe help edit. <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 05:37, 16 February 2024 (UTC)
 * I'm too tired to draft a request for comment because of all of my college studies, but I will edit it too first thing in the morning. – 64andtim  ( talk ) 05:52, 16 February 2024 (UTC)
 * I also have a bunch of work to do and haven't ever made or commented on an rfc so I'm not the best choice either. It's also late in the night already in my timezone. I'm happy to comment on an rfc though in the morning. –  Pharyngeal Implosive7  (talk)  05:59, 16 February 2024 (UTC)
 * I actually don't know how much use a private edit filter wiki would be. Mailing lists aren't particularly friendly, but at least they go to an inbox that (presumably) the person otherwise monitors. A separate wiki might get some initial activity due to novelty, but I suspect it'd die out? In part because the community of EFM/EFHs used to be quite small (I'm not sure about now?). That said, I suppose it'd make discussing filters easier for WP:BEANS reasons. Right now, we'd have to just email each other. A ticketing system might be useful, but sounds like a private phab board serves better at that than a wiki. ProcrastinatingReader (talk) 12:07, 16 February 2024 (UTC)
 * A private phab broad would only be useful for edit filters, and it doesn't cover the discussion aspect well. (since they have different tickets with different threads, while a wiki can centralize discussion under a single page) Suffusion of Yellow also mentioned the idea of expanding it to discussion about LTAs per WP:DENY and WP:BEANS at the last discussion. A wiki could have more benefits than a phab broad. I suspect it'd die out? Maybe, or maybe not. I think it would be good if we can try. If people enable emails from notifications on the private wiki it could be remain active. (We can do something like @EFMs for time-sensitive issues) <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 11:03, 17 February 2024 (UTC)

Reactivated 1286
See Edit_filter_noticeboard/Archive_12.

I reactivated the rule now, because the person is back, see e.g.. @64andtim, thanks for hint. -- seth (talk) 23:23, 20 February 2024 (UTC)


 * In addition, with my suggestion, I've asked them to de-specify the filter's name. Thank you. – 64andtim  ( talk ) 00:37, 21 February 2024 (UTC)

Set filter 1285 to disallow?

 * ("Removal of short description", public)

See WP:EFR, courtesy ping.

Very low FP rate. While this might seem like a trivial thing; it's almost always accompanied by something worse. This is usually either just partial blanking that slips past the other filters, or users who have something to say, click the first available edit link at the top so they can share it with us, and remove anything that they don't understand. When reviewing, remember to view the log with saved changes only to see the new edits we'll be disallowing; this filter overlaps quite a bit with some blanking filters. Suffusion of Yellow (talk) 20:22, 11 February 2024 (UTC)


 * In addition, see Special:AbuseLog/36985288; an IP address has removed the last two brackets of the SD template. It turns out they were trying to add "and YouTuber" but they didn’t realize that can break the short description template.
 * Also, Special:AbuseLog/36984552 recently catched an IP address adding something non-constructive in lieu of the template. – 64andtim  ( talk ) 20:37, 11 February 2024 (UTC)


 * Support as requester. Filter has a very low rate of false positives thanks to iteration by Suffusion of Yellow. Template:Short description's typical placement at the very top of article wikitext results in it being a common target for vandalism and disruption of the template can serve as an indicator of unconstructive edits that are not caught by other filters. Setting it to disallow would reduce the workload of recent changes patrollers and avoid the rarer cases of vandalism being missed altogether. Uhai (talk) 07:49, 12 February 2024 (UTC)


 * I'd suggest switching to a warning first and seeing the extent to which that resolves the issues --DannyS712 (talk) 19:24, 12 February 2024 (UTC)

✅ Set to disallow with the standard message, no prejudice against changing it back. All the best: Rich Farmbrough 21:01, 21 March 2024 (UTC).
 * I'd support setting it to disallow (or warning, if preferred). Low FPs and catching a lot of nonsense. ProcrastinatingReader (talk) 13:49, 14 February 2024 (UTC)
 * I also will support. Disallow seems pretty good to me but we could try warn or warn+tag first to see if that resolves the issue. –  Pharyngeal Implosive7  (talk)  21:27, 14 February 2024 (UTC)
 * In addition to what I said above, since those log entries are "accompanied with something worse", I support setting this to disallow. – 64andtim  ( talk ) 14:05, 15 February 2024 (UTC)
 * Oppose disallow for one reason. If someone adds a SD by accident and then tries to remove it, it wouldn't work. (Example) (Which I don't think can be fixed using the filter). But I support warning. Nobody  ( talk ) 15:09, 15 February 2024 (UTC)
 * Even if they add it by accident, they should change it to something appropriate as the template shouldn't be removed once it's added to a page per Short description: "all mainspace articles should have a short description". The example you linked is actually perfect for something that should have been disallowed as it was a suitable short description for the page. Furthermore, if the edit summary contains revert/rv/undid (like if they click the "undo" button) then the edit will be allowed to go through. Uhai (talk) 15:18, 16 February 2024 (UTC)
 * Right, so maybe we should include some instructions, like if you wanted to leave an intentionally blank description, use ? <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 15:48, 16 February 2024 (UTC)
 * @0xDeadbeef The warning message above already contains these instructions. Uhai (talk) 16:19, 16 February 2024 (UTC)
 * I can't read. (I've said this too many times) <span style="font-family:Iosevka,monospace">0x Deadbeef →∞ (talk to me) 16:20, 16 February 2024 (UTC)
 * I was more thinking along the lines of "They're not sure what the right description is, based on our standards, so they're trying to leave it empty for someone who does." Nobody  ( talk ) 16:20, 16 February 2024 (UTC)
 * It's a fair point but I'm not sure the situation where a non-confirmed, good faith editor accidentally adds the short description template and then tries to remove it (without undoing) is common enough to warrant concern over the filter disallowing. In the absolute worst case scenario, they could follow the link in the warning message to report to EFFP and have a patroller there remove the template from the page for them. Uhai (talk) 17:38, 16 February 2024 (UTC)
 * So set to disallow and see how many reports come in? Sounds good. Nobody  ( talk ) 18:44, 16 February 2024 (UTC)
 * @1AmNobody24 Not sure if this was meant to come across as sarcastic but I can say that I went through 1500 hits and didn't see a single instance where the removal was constructive. Even the example you linked I would not consider a false positive as the user should not have removed the short description and I have since added it back. It's unclear why they did this—maybe they thought removing it was the same as it being "none", which the warning message would clarify. Setting the filter to warn may stop some vandals, sure, but plenty just ignore or don't see the warning messages and click "publish" again. Often even good faith editors miss seeing the message and just think that they missed clicking the publish button or something.
 * I would never advocate for a filter being set to disallow as a test to see the volume of reports. Most disallow filters will have some number of false positives and I believe the benefits of this filter being set to disallow would greatly outweigh the harm from false positives and that the volume of false positives and non-frivolous false positive reports should be very low. Uhai (talk) 19:16, 16 February 2024 (UTC)
 * Not sarcastic at all. Just wanted to know how this type of Situation would be handled. Nobody  ( talk ) 19:26, 16 February 2024 (UTC)
 * No problem. My apologies for the misinterpretation. Uhai (talk) 19:30, 16 February 2024 (UTC)
 * It has been about a month and it seems like most people support disallow. Any progress in making the filter disallow? –  Pharyngeal Implosive7  (talk)  01:16, 16 March 2024 (UTC)
 * Nope. Still log-only. Codename Noreste  🤔  talk  01:12, 20 March 2024 (UTC)
 * Yeah I know but I was asking the EFMs if they could change the filter to disallow because consensus clearly supports this change. –  Pharyngeal Implosive7  (talk)  01:19, 20 March 2024 (UTC)


 * For now, until a custom disallow message is made, the generic disallow (and somewhat bitey) message is currently being used. Codename Noreste  🤔  talk  00:47, 22 March 2024 (UTC)
 * I have made a request for an interface page to be created. –  Pharyngeal Implosive7  (talk)  01:14, 22 March 2024 (UTC)
 * The interface page has been created.  You can change the message used now and we'll be done. –  Pharyngeal Implosive7   (talk)  13:23, 22 March 2024 (UTC)
 * Thanks, however the page name is incorrect. We need to move MediaWiki:Abusefilter-disallow-short-description to MediaWiki:Abusefilter-disallowed-short-description, I think. All the best: Rich Farmbrough 20:30, 22 March 2024 (UTC).
 * My mistake. You are correct. Here is the request for the move. –  Pharyngeal Implosive7  (talk)  22:34, 22 March 2024 (UTC)
 * The mediawiki page was moved to the correct title. It is ready to become the disallow message for this filter. –  Pharyngeal Implosive7  (talk)  17:57, 24 March 2024 (UTC)
 * And this should be done as quickly as possible as the disallow message currently is undefined. –  Pharyngeal Implosive7  (talk)  18:02, 24 March 2024 (UTC)
 * could you please change Special:AbuseFilter/1285's message to MediaWiki:Abusefilter-disallowed-short-description? The current one does not exist anymore. &mdash; Mdaniels5757 (talk &bull; contribs) 01:54, 25 March 2024 (UTC)


 * ✅. Might be nice for the AF-editor interface to check for that sort of nit (or maybe even autocomplete?). DMacks (talk) 03:47, 25 March 2024 (UTC)
 * ...also, what's the difference between "System message to use for disallowing" (popup menu) and "Page name of other message" (textbox)? DMacks (talk) 03:49, 25 March 2024 (UTC)
 * DMacks - I've only seen the textbox version of this option, which just displays a textbox below the edit window to notify or warn a user. I would assume that the "pop-up menu" version/option either opens a pop-up or new window in the browser displaying a notification to the user that their change is disallowed, or uses java to display a notification in front of the editor to a user. Don't quote me, though; I've never seen this before. Is it a new thing, or am I just crazy? lol  ~Oshwah~  (talk) (contribs)   06:30, 25 March 2024 (UTC)
 * DMacks - I'm not seeing the "(popup menu)" drop-down list at all. Do you have a screenshot?  ~Oshwah~  (talk) (contribs)   06:33, 25 March 2024 (UTC)
 * Abuse-filter_actions_screenshot.png
 * DMacks (talk) 13:17, 25 March 2024 (UTC)