User:AmandaNP/UAA/Blacklist is permanently protected from editing because it is a page that can expect to need this level of protection on a multi-year time scale. Substantial changes should first be proposed and discussed here on this page. If the proposal is uncontroversial or has been discussed and is supported by consensus, editors may use {{edit template-protected}} to notify an administrator or template editor to make the requested edit.
Instructions
For Admins: If there is no objections to code change requests below, please implement them and mark it as no objection if it's an edit request. Other administrators should also feel free to revert such changes without explanation. Admins can also edit the list at their own convenience. Requests should be able to match the format so they can be easily changed, if they are not, automatically decline them.
For Template editors: Right now, this group's editing privileges are being tested. While you are able to edit the page, you must propose changes here first for a period of:
24 hours, requiring 1 support and no objections
72 hours, with no objections
72 hours, if there are more than 3 comments, with a majority supporting
If you are not proposing an entry, you may make any modification that meets the above criteria, and where your opinion would count.
For Non-admins/template editors:
Please propose your additions or removals in this manner (changing anything in bold):
;YOUR REGEX/STRING:FLAGS
Only insert the colon if you are invoking a flag
Separate each flag with a comma
Possible flags:
WAIT_TILL_EDIT - Waits till the user edits to file the report
SOCK_PUPPET(USER) - Directs patrollers to file an
SPI
about the user.
LABEL(TEXT) - Overrides placing "Target: {{REGEX}}" with "Target: {{YOUR TEXT}}"
NOTE(TEXT) - Adds a note with the text appended
LOW_CONFIDENCE - Adds a note that the filter is not considered to be highly effective
NO_SIM_MATCH - Do not use the similar characters to match any obfuscation (This is rarely needed to be used)
We're concerned that this is far too broad and is returning loads of false positives - could it at least say what bit of the name it is picking up on. I think we're basically ignoring these. Also: Spunk spunk _is_ slang for semen, but not always. Secretlondon (talk) 03:56, 5 March 2012 (UTC)[reply]
Edit request on 15 December 2012
This edit request has been answered. Set the |answered= or |ans= parameter to no to reactivate your request.
Recently the username Pepernickel (talk·contribs) was caught, and I can't see why. It's *close* to matching the p[e3][nm][i1!][sc5z] rule, but being close shouldn't be enough to put it on the list. I'm worried there might be a bug in the code, and if there is, there might be more that we're missing. I'm not much for regex though so I could be missing something that to others is obvious. —Soap—04:49, 25 January 2013 (UTC)[reply]
Nathandrex (talk·contribs) shouldnt have matched either. Is there some other function running on top of the blacklist that catches names that "almost" fit? e.g. I could see a rule that treats drex as a variant of dreck as a good way to catch vandals who obfuscate their names to escape detection. pernic could be an odd deliberate misspelling of penis. If that's all that it is, it's probably good, but if it's not, then there's a bug. —Soap—04:58, 25 January 2013 (UTC)[reply]
Ahhh, I think I figured it out. pernic is seen as "pemis", drex is "drecks", both of which match the list items exactly. —Soap—05:02, 25 January 2013 (UTC)[reply]
Use on wiki farms
I was just wondering if this list could be pulled and used on private wikis, in the fashion of TitleBlackList, Spamblacklist are used. I guess I asking if this can be incorporated in AntiSpoof actions.
Since, I'm (very) bad at regex, I'll not suggest but rather just give the words:
madarchod (Stands for motherfucker), lund (stands for penis), choot (stands for fuck), peshab (stands for piss).
Discussion at
choot (stands for fuck) phuddi (stands for pussy) behnchod (stands for sisterfucker) alternative spellings "bainchod or bainchoad or behnchoad" moot (stands for piss) lun (stands for dick) lull or lul (stands for dick) lorra or laora or laorra or lora (stands for dick) ghanta (slang for dick) tattay or tatta (stands for testicle(s)) gandu or gaandu or gandoo or gandoo (stands for bottom gay) gaand or gand (stands for ass) bund (stands for ass)
It would be useful for these to be added. Except perhaps "moot" and "bund" which are commonly used in some other languages and therefore might be ordinary username requests. I am not sure if this request is to prevent any username with these words in the name, or just to raise alert for it. If only raising alert then it is fine because there would also be an explanation why. MPS1992 (talk) 20:04, 18 January 2016 (UTC)[reply]
Such words seems to be quite prevalent for Indian accounts created specifically for insulting other editors. --QEDK (T 📖 C)10:24, 19 January 2016 (UTC)[reply]
Not done: It seems that no admins patrolling this category have the necessary skills in regex writing. So I suggest you ask for help at
Some of the blacklist entries that have been added recently are regular expressions that are written in a complicated way and could be optimized to a shorter string that is easier to understand and requires less processing time, while still matching the same bad usernames. Here are some examples of possible improvements:
ga(a|)nd(u|oo|)
Can be simplified to gaa?nd. Using the ? operator to indicate an optional character is more efficient than using | with an empty string. Therefore, the string as it is can be rewritten to gaa?nd(u|oo)?. However, you will notice that since the "u/oo" part at the end is optional, it is actually redundant, since any username containing "gandu" or "gandoo" will match "gand" anyway.
lu(n|l|ll|nd)
Can be simplified to lu[ln]. Using character selectors is more efficient than the | operator. Also, similar to the case above, it is redundant to add extra cases to the regex to check for "lull" or "lund", since these will already match "lul" and "lun" respectively.
l(a|)or(r|)a
Can be simplified to la?orr?a. As noted above, using ? to indicate an optional character is more efficient than using the or operator with an empty string.
b(ai|eh)ncho(a|)d
Similarly, can be simplified to b(ai|eh)nchoa?d.
tatta(y|)
Can be simplified to tatta. Since the y at the end is optional, checking for it is redundant - any username containing "tattay" will match "tatta" anyway.
Done (5 hours prior to 72, but given the mild support above [which would actually have qualified it for addition at 24] and no opposition, this should be fine). Izno (talk) 11:43, 4 May 2016 (UTC)[reply]
Template-protected edit request on 20 September 2016
This
User:DeltaQuad/UAA/Blacklist has been answered. Set the |answered= or |ans= parameter to no to reactivate your request.
A lot of usernames containing the string "bich" seem to be Vietnamese names. I'd like the "bich" filter to have a note saying that Bich is a Vietnamese name (like how the "porn" edit filter has the Thai name note and the "shit" one has the Indian one).
So, I'd like the following added to the string bich:
NOTE(Bich can be a Vietnamese name - make sure this is not the case before blocking.)
According to my count, 6 out of 11 are false positives for referring to "your father". This is a majority, but not a vast one. And users with truely disruptive names are probably less likely than good users to have user pages, either because they were blocked before they could create it or because it was subsequently deleted. 217.132.181.16 (talk) 23:43, 31 October 2018 (UTC)[reply]
Not done: I think there needs to positive evidence that this filter would be desirable. Asking someone else to prove that it won't cause issues is a bit of a fallacy. Izno (talk) 16:02, 10 November 2018 (UTC)[reply]
Template-protected edit request on 11 June 2019
This
User:DeltaQuad/UAA/Blacklist has been answered. Set the |answered= or |ans= parameter to no to reactivate your request.
For the first part - lets not bring a debate about transsexualism vs transgenderism in to this bot config - just say "is offensive to some" (I'll assume that you don't have to be in that class to be offended by it right?) For the second, that seems like a way to common word that is more likely to be used in its traditional context. — xaosfluxTalk04:34, 22 June 2019 (UTC)[reply]
I will obviously have an opinion on the first part, but either way, I removed the note as a botop action. I'll leave the second one as it's not a term I am familiar with,
Note: I just had to file this UAA report, so I'm starting to really think that I was onto something by maybe suggesting ;trap:LOW_CONFIDENCE because I'd rather these accounts get reviewed rather than not. –MJL‐Talk‐☖16:18, 30 June 2019 (UTC)[reply]
Not done for "trap" too many FP's, and also too many suggestions above about what parameter to use. Certainly without prejudice for further discussion or being overrided by bot owner. — xaosfluxTalk23:57, 2 July 2019 (UTC)[reply]
Template-protected edit request on 4 November 2019
This
User:DeltaQuad/UAA/Blacklist has been answered. Set the |answered= or |ans= parameter to no to reactivate your request.
Apologies if I didn't format this right, first time making a request here. Intent, as the section header suggests, is to catch lyrics website spammers, the websites almost always contain the word "lyric" or "lyrics", the usernames are often the name of the website but sometimes have numbers thrown in as well. "lyrics" would work as well, if you think that "lyric" might have false positives. NO_SIM_MATCH may be appropriate here (these are website names, so obfuscation is unlikely), but since the note says "rarely needed" I didn't add it above since I wasn't certain.
Not done: it's not clear what changes you want to be made. Please mention the specific changes in a "change X to Y" format and provide a reliable source if appropriate. Izno (talk) 22:28, 20 June 2020 (UTC)[reply]
Template-protected edit request on 9 July 2020
This edit request has been answered. Set the |answered= or |ans= parameter to no to reactivate your request.
Note: "team" and "group" are already present (search for ;\bteam$|\bgroup$:LABEL(role account?)). If we want to do this, code to add would be ;\bteam$|\bgroup$|\social$|\media$:LABEL(role account?),NOTE(Usernames containing this string are often role accounts - check if this is the case). Best, --Mdaniels5757 (talk) 20:23, 13 July 2020 (UTC)[reply]
A possible regex: ;(75,?.?000 ?cases?|Sievert ?81):SOCK_PUPPET(Sievert 81). For conservapedia, ;conservapedia:NOTE(Possible sockpuppet of Sievert 81),LOW_CONFIDENCE could work, but I'm not so sure that's a good idea. —Mdaniels5757 (talk • contribs) 18:54, 26 December 2020 (UTC)[reply]
Deactivated TPER since pending for over 2 wks, and nobody seems to want to implement it. Amanda can do so if she wants, I guess. I’m not entirely sure there’s a point to doing this for someone with only one set of socks at SPI, though? ProcrastinatingReader (talk) 01:36, 28 December 2020 (UTC)[reply]
Template-protected edit request on 27 December 2020
Could you please Add the following below in exact wording:
Hello @AmandaNP. I have a suggestion regarding to the username blacklist.
1. The current surge of so-called 'lyric video' creators. How about considering the "lyric video or LV" in the blacklist as ot might appear as a company/organization/any group?
The transcript for this is/are:
lyric[ ]video
NOTE refers to the lyric video creators on social media outlets.
2. As the suppression of vandalism in pages with alleged Among Us game references like impostor/imposter/sus/amogus and other related terms. Users with those names or substring tends to vandalise pages with Impostor or SUS and alike in the mainspace.
The transcript will be like:
impostor/imposter
NOTE Among Us-related vandalism and disruptive editing.
we request that the NOTEs be respectively clarified as "ethnic slur for Indigenous North American women" and "ethnic slur for South Asians". –LaundryPizza03 (dc̄) 22:55, 26 August 2021 (UTC)[reply]
I recently encountered a username containing ching chong that turned up at AIV, and noticed that imitation Chinese is not currently covered by the blacklist. Thus, I recommend the following addition:
:ching[ -]*ch[ao]ng:NOTE(imitation of the Chinese language)
This won't capture all kinds of imitation Chinese, but should cover the most common variety, ching chong or ching chang chong. Ching, chang, and chong by themselves should not trip the filter since they are common Chinese names. –LaundryPizza03 (dc̄) 09:02, 11 December 2021 (UTC)[reply]
I'm not sure how what I did is wrong, as it seems to match the format of the other entries. Anyway, I have to wait 72 hours before fixing this due to the rules given in the page edit notice. * Pppery *it has begun...14:55, 8 January 2022 (UTC)[reply]
Partly done: No objections in >72h - most of these are already there but I added "partnership" and "associate" singular. firefly ( t · c ) 12:20, 28 January 2022 (UTC)[reply]
Duplicated comments
Hi, I have noticed there are duplicated comments in the list and some of them seems to be mistakes. Please see if they need some fix.
The first duplication I found:
nazi
NOTE(Usernames that promote a controversial or potentially inflammatory point of view may be in violation of the username policy due to creating a hostile editing environment- but some real names contain the string "nazi" - especially names from the Middle East - be careful that this is not the case before blocking.)
national ?social(ist|ism)
NOTE(Usernames that promote a controversial or potentially inflammatory point of view may be in violation of the username policy due to creating a hostile editing environment- but some real names contain the string "nazi" - especially names from the Middle East - be careful that this is not the case before blocking.)
– the reference to 'nazi' as a part of some real names is not necessary in the comment to the 'national socialism' item.
The other one is here:
rac(ial)?(ism|ist)
NOTE(Accusations of racism may constitute a personal attack)
antisemit
NOTE(Accusations of racism may constitute a personal attack)
– antisemitism may be considered a special case of racism, but I think it would be better to use the specific term in the comment.
I have been seeing accounts containing the numbers 69 and 420, and sometimes, the numbers are mashed together. They have been used to vandalize Wikipedia (examples: [2], [3], [4]). I know that 69 has a
Probably meant to do parentheses instead of square brackets. I saw one of those "test account" pages as a vandalism-account, but I was probably overreacting. Painting17 (talk) 17:13, 3 April 2022 (UTC)[reply]
Partly done. I added Shko, with some tweaks to avoid false positives on something like "JoshKorman", since I'm familiar with that case. As to the rest, though, there's insufficient explanation of why they're needed, and I'm uncomfortable actioning something where I'd have to fix syntax errors and guess at what was intended. @Painting17: Feel free to resubmit with syntax errors corrected and with an explanation of why each addition is necessary, preferably linking to examples of disruptive accounts that match the patterns in question. If you need help with regular expressions, see https://www.regular-expressions.info/ and https://regex101.com/. -- Tamzin[cetacean needed] (she/they)19:26, 18 April 2022 (UTC)[reply]
Template-protected edit request on 3 May 2022 - SPaG
This edit request has been answered. Set the |answered= or |ans= parameter to no to reactivate your request.
under lmfao, the note currently reads NOTE(stands for "laughing my f-ing ass off|). I assume its meant to read NOTE(stands for "laughing my f-ing ass off")Aidan9382(talk)12:51, 3 May 2022 (UTC)[reply]
NOTE(Usernames that promote a controversial or potentially inflammatory point of view may be in violation of the username policy due to creating a hostile editing environment- but some real names contain the string "nazi" - especially names from the Middle East - be careful that this is not the case before blocking.)
to
national ?social(ist|ism)
NOTE(Usernames that promote a controversial or potentially inflammatory point of view may be in violation of the username policy due to creating a hostile editing environment)
Change
antisemit
NOTE(Accusations of racism may constitute a personal attack)
to
anti[\- ]?semit
NOTE(Accusations of racism or antisemitism may constitute a personal attack)
Not done for now: This is actually a controversial edit, so you'll need to discuss first with other editors. Please open a new section here and start a discussion. A new edit request may be opened only when there is agreement among editors for the proposed change(s).
Yes, got that. In the first case, I think the originator had posterity in mind where younger editors might not readily associate "the n-word" with National Socialism. So let me take my TE hat off and object to that elimination of a plausible explanation. Nor do I see any need to use the antisemitism term in the Note(. The note defines the term, so "racism" should be enough.
Since I expressed opinions, I'll reopen this request so another TE can objectively disposition it.
The role accounts lins should add the string commission for several entries at Commission, particularly the one that is a type of government agency. The username the prompted this request was Humanrightscomissionindia (talk·contribs), which implies a role account for the National Human Rights Commission of India. Actually, since it's misspelled, I would add com+is+ion and change committee to com+it+ee. Also, department is listed twice for some reason.
This change would yield:
While this is, strictly speaking, correct, I don't know if this is really a worthwhile change - the current message communicates that the rule might FP on surnames, and I don't think we need encyclopedia-level accuracy here. GeneralNotability (talk) 19:11, 5 September 2022 (UTC)[reply]
Not done for now: This appears to be a controversial edit, especially when we consider that the originators could have gone even further, since the surname page allows for the inclusion of "given name" as well...
NOTE("SEO" stands for the promotional technique "search engine optimization" - but Seo is also a Japanese surname as well as a Korean surname and given name)
Done Ku Klux Klan. Not done for the admin one since I think it is helpful to specify what things are recognized as positions of authority rather than be vague. Not done for lawsuit because I don't see the point. * Pppery *it has begun...15:42, 8 November 2022 (UTC)[reply]
Edit request 21 November 2022
This edit request has been answered. Set the |answered= or |ans= parameter to no to reactivate your request.
@AmandaNP thanks for the note, it doesn't matter to me, I'm not personally invested in this request from AKK-700; was just patrolling outstanding edit requests and came across this in your personal userspace. — xaosfluxTalk21:34, 17 December 2022 (UTC)[reply]
Done most. Not done Ass size (this seems to remove information from the report for no clear value), sodomy (the article sodomy suggests the term is broader than simply anal sex), death threat ones (this seems obvious, I'm not sure why you think it should be included), sock/meatpuppet (admins should be able to figure this out for themselves without a note). However, your rate of requests here is a bit high and they're starting to feel like makework. * Pppery *it has begun...21:18, 20 December 2022 (UTC)[reply]
Edit request 4 March 2023
This edit request has been answered. Set the |answered= or |ans= parameter to no to reactivate your request.
:(is|was) ?a ?fag/:are ?fags (the entry these are duplicating is flagged as low-confidence but these aren't. Higher confidence for more specific matches seems reasonable)
;is ?a? ?(stupid|dumb): (I don't see what the dumb part of this regex duplicates)
queers (again the duplicate is low-confidence)
porno (same)
pissing/pisser (same)
hitler (admins at UAA should already know who Hitler is and not need a reminder IMO)
pussy (broadening a regex and then making it low-confidence at the same time seems counterproductive)
fanny (not convinced this is a good idea given how many uses fanny has)
;apk:NOTE(Sometimes used by spammers advertising an APK downloader. Check user's contributions to see if this is the case before blocking.),LOW_CONFIDENCE
I've noticed the string 'apk' appear in the usernames of spammers advertising apk downloaders. Examples: [5], [6], [7]. Feel free to make corrections as needed, as coding is not my area of expertise. ◇HelenDegenerate◆ 19:53, 12 May 2023 (UTC)[reply]
Are the regexes evaluated by Python, JavaScript, or something else? More specifically, I would like to know which look-ahead and look-behind assertions are available to blacklist some additional patterns without causing false positives. Thanks. Daniel Quinlan (talk) 02:21, 5 March 2024 (UTC)[reply]
@Pppery, One more possible addition would be 'syndicate', another role account. There was someone registered today but I don't feel like digging thru ListUsers. 🇺🇲JayCubby✡ please edit my user page! Talk02:56, 2 April 2024 (UTC)[reply]
I've noticed that new usernames which contain "Skibidi" in them often are used only for disruption/vandalism/trolling. Is there any way we could add a filter which blocks all usernames with "Skibidi" and/or sends them to UAA? Same with "Erm, what the sigma", for the same reason. Originally requested at Wikipedia:Edit filter/Requested If you reply here, please ping me. — thetechie@enwiki: ~/talk/$03:33, 29 May 2024 (UTC)[reply]
Not done I'm not seeing sufficient cause to do this - it seems like the existing systems are working as they should (especially since the second edit was reverted by ClueBot). * Pppery *it has begun...05:10, 6 June 2024 (UTC)[reply]