Give people the benefit of the doubt; a creative thread might look off-topic at first glance.People are allowed to argue; they are not allowed to *troll*.Not everything that gets reported needs to be deleted or banned; don't hesitate to use the clear button if no rule is broken.Janitors cannot readily tell if an anonymous poster is ban evading. Ask a mod to check, or BR for something else. Help make sure nobody gets banned for 30 days for something they didn't do.Don't increase bans beyond 3 days for minor things.Accept appeals for minor things, and be forgiving on /q/ and IRC.Keep in mind: People come to 4chan to have fun, but their fun should not come at the expense of others.
Network: irc.rizon.netChannel: #janiteamKey: VortexCleaningActionNOTE: NEVER GIVE OUT THIS PASSWORD TO ANYONE ON IRC, OR VIA IM OR E-MAIL. If they are a current janitor, they'll be able to access it on /j/ for themselves.Another note: You can only connect with a registered nickname, and must connect to the IRC server using SSL on ports 6697 or 9999.You may need to install OpenSSL yourself on some clients (ex. mIRC).
#janiteam is moving to Discord! Please check your janitor email for details. If you need a new Discord invite, please email RapeApe or contact him on IRC.
this is a backup threadfeel free to use it if, for whatever reason, you ever find yourself unable to get ahold of the mods or janitors (e.g. discord experiences an outage)
>>10263if anyone else sees this post, i have it on good authority that it's since been sorted out (so no need to ping RA about it)
I have the mandate of heaven and discord pooping won't stop me from declaring it.
>>10268>and discord pooping won't stop me from declaring it.frank's pooping could stop a freight train, don't get cocky
>>10288>frank's pooping could stop a freight train, don't get cocky
a toast: to no more discord
you may only post ITT if you've joined the mile high shitposting and BRing club
>>10062don't forget, you're here forever
>>9504On my yearly pilgrimage again, this time internet was included with my ticket.>>9954It was worth it to join the sky janny club. Hot pockets ain't free
My 12-hour flight comes with free wi-fi!
I am now the sky janny, fear me
You’ve heard of sky jannying. Now introducing the long awaited undersea jannying
Stepping away for a bit.
>>10354can i come with you? i are an serious business man also
>>4303still around, just less active while i move house
/vt/A little early but I will be on vacation from the last week of February to the 2nd week of March. Activity will be low to none.
Gonna be away from janny duty the wednesday the 12th until probably the weekend. If /d/ floods during that time, someone please flush.
Gonna be gone for around 2 weeks. Might end up being shorter or longer, not sure yet, but I probably won't be able to clean things.
My list of April Fools ideas is a little short. Why not start collecting really early?
We do it for free so /j/ should be safe.
>>10436was so much fun as it unfolded and eventually /pol/ going down WTF OMG TOPKEK wud do again
So wh4t t1me does th1s end?
>>10440m1dn1ght et
>>10438/j/ gett1ng h1t would be the funn1est th1ng th4t could h4ppen, honestly
Janny fan art thread, because, funnily enough, the last one has reached image limit.AI sloppa welcome
>>10431>I want to make zero dollars, every day, for the rest of my life.
>>10433>you'll make nothing>and you'll be happy
Ask them here.
Speaking of wordfilters, couldn't one be implemented for bikinischizo? I imagine just having a wordfilter that would block any post that mentions both "Kishida" and "bikini" would go a long way. Of course, it would stop legitimate posts, but those are rare enough that it probably wouldn't be a problem.
>>10414I don't know "Kishida" but the bikini poster has been in a number of different threads (so different characters) on /vg/ so I don't think it'd be that simple.
>>10415https://desuarchive.org/a/search/text/kishida%20bikinihttps://desuarchive.org/a/search/text/ldp%20bikinihttps://desuarchive.org/a/search/text/solo%20leveling%20bikiniHere are some of his "highlights".
>>10414>a wordfilter that would block any post that mentions both "Kishida" and "bikini" would go a long way.just BR him, filters are shit against actual people, you would also be hampering the entire userbase over one guy, the bikini evader posts in multiple boards too
>>10417Normally I'd just suggest that, but he's such a problem and reports often take so long to be acted upon (up to several hours, even on busy boards like /a/), that it might just be more efficient to have a filter. If the femanons guy could get a filter, I can't see why bikinischizo can't considering he's an even bigger problem. Besides, yes I know it may inconvenience a few users, but really outside of other anons talking about his hatred for Kishida, the combination of "Kishida" and "bikini" is unlikely to naturally occur anyway.
Hey, you guys know what would be a small but really helpful addition whenever it's next time to update the Janny system and interface? If when a reported post in the queue has the date of when the post itself was made on it, maybe not necessarily in full detail as that's probably not important, but if that post was made today, or if the post was made say, 47 days ago.This would help slightly whenever someone goes and reports posts in very old and even archived threads.Optionally, that posts in archived threads can only be reported for GR1, or anything else where deleting an archived post would still be useful and necessary, like for NSFW images on blue boards, spam links, or dox.Another feature would be for fraudulent reports. Now, if mods got a warn/ban request for every single fake report that would just be swamping you guys with needless shit. Instead, for when a post exhibits no obvious reason as to why it would break any rules, you click a button for that which clears that report and then adds a number to a counter for the guy who reported it (the janny wouldn't know this number).Thus, if a janny suspects someone is making a bunch of fake reports, a mod can go and check and either see a counter reflecting the number of times recent reports by them have been rebuked. Optionally they look in the user's report history and see if a lot of the recent ones are rebuked.Obvious exception for making a fake GR1 report, where perhaps that's sent as a request towards a mod's queue about making a fake illegal report, and it's up to them if there's any action to take.I'm thinking maybe this could save both jannies and mods some time in certain situations.
You know how when you open a BR window, you can see if the IP already has pending BRs?Would it be possible to expand on that and either state how long ago the BR was requested, or at least whether the IP is currently timed out from posting?I janitor a blue board, and the board has a lot of porn spammers. Sometimes these porn spammers wait out the BR timeout and resume spamming, and since there's a BR already up it can trick janitors into thinking they're already timed out and only delete the posts, allowing them to keep doing it.
>>8941does anyone else NOT use the dark mode option when in queue?i don't and it can make posts reported as illegal a little more difficult to see.if this could be made redder so my fucked up eyes can more easily distinguish, i would be very appreciative and stuff
>>10408You can enter custom CSS for the queue by clicking the cog thing at the top.I do not know enough CSS or how the queue tells you what's reported illegal, but you may be able to give illegal reports a different color or a huge glow or a jpg of a lion going >(:3) from your hard drive that way.
>>10408.report-cat-prio { background-color: #ffb5b5;}Try this
>>10385GR9 would ideally mute for longer imo.Its a strict template that needs mod approval for regular use. In the landscape of extremist evasion I've also seen offenders try to be 'economical' by waiting the mute and going full throttle
Hello everyone:I understand that we've all been janitors and moderators for varying amounts of time, but I also want to acknowledge that we are all in this together for the site and for the users. This post is to represent something that I want to put out there for everyone: how is your mental health?I understand that we are exposed to the most abhorrent nonsense users have to offer, and I want to make sure everyone is doing okay. Sometimes the stuff posted on site can get to a person, myself included, with things in their personal life and things that get posted online. Recently in my life, there have been a lot of changes, like moving away from family and friends and adjusting to a new way of life, so what I ask of you today is: How are you doing?
this has been an influence for a net positive?? on my life for almost 2 decades, thank u 4chan for still surviving
haven't been a janny long, but i've been on 4chan for a pretty long time. it's interesting to me that a lot of the crap I used to simply tune out I now have to actually deal with and see if it is breaking rules. I think before I started doing this I simply ignored it without giving it too much thought. that said even though we see a lot of really screwed up shit its definitely offset to me by seeing people posting actual useful advice when they get nothing out of it, and seeing people getting actual useful advice. also without sounding excessively philosophical I think clearing a lot of the reports is just as satisfying. I may not like some of the idiotic posting that happens, but the fact that it has a place to exist where you can post with a reasonable chance of maintaining your anonymity and not suffering reprisals for your opinions is important, in a way it's kinda cool to be in a small part helping to defend that. obviously if something is breaking a site rule we deal with it, but I think it's important that we aren't telling people if their ideas or good or bad. just that they are in fact allowed to have them here and someone elses opinion wont invalidate that.
>>10319>>10321[s4s] is super fun but not really safe for work :3
>>10388duck yu
>>8828>mental health thread gets bumped>4chan goes downthe lesson is clear: internalize your misery and the site persists, janiteam is the kid in Omelas
The idea is if you need to provide screenshots or logs related to janitoring that you upload it here instead of hosting it somewhere else.
>>2675>http://img.4chan.netshould we reconsider hosting this banner?it resolves on some shitty phishing "discover small personal loan quick cash" page
>>10359to clarify, the banner itself does notthe url in it does
>>10360also maybe time to updatehttps://s.4cdn.org/image/banners/5.png
>>103951000 hours in mspaint, who do i send the bill to
>>10399nobody, you do it for free
What possibilities actually exist for 4Chan to utilize Large Language Model technology for the purposes of assisting with moderating the website? My idea here isn't so much a plain autojanny doing all the work, as much as a special tool for assisting us with doing some of the most important work, which falls under GR1.Say that someone like the Babyshitter is subtly altering the image so that it doesn't get caught by the autoblock thing, like he's currently doing, what if we had an AI bot which had used the content put into the autoblock as its training data for the purposes of identifying subtly altered versions of it?Then that bot patrols a board and then scans through every currently active thread for if someone posts something like that, and if someone does, it immediately makes a global GR1 report and maybe also pings the moderation channel on Discord. A crucial question here is if the bot should be given any operational autonomy for this task, and thus be able to request/issue bans and delete posts. AI can be very fallible, so I'm inclined to say probably not.What would be the practical and logistical constraints in developing this kind of bot, and then operating it? My primary idea with this is to deal with those posting illegal content like CSAM, and possibly certain people operating their own spam bots (ergo fighting fire with fire).Finally, if this is at all possible to implement, what do we name the thing? Murphy? Fagballs?
>>10390>>10391>>10392Well, best I understand, the training data for LLMs would not directly contain the content, thus it shouldn't by itself contain anything illegal. Someone who understands this technology better than me is free to chime in, however. I also don't think that it would even need to run on every board, at most it might run kind of regularly on places like /b/ and /v/, high traffic boards where there's known troublemakers of higher consequence, and only ever get rolled out termporarily for if someone starts shitting up some other board with botted spam.Combative shitposters with a grudge are already weaponizing things like bots and altering image hashes, and I think that it would be in our own interest to try to see if there's any potential use in LLMs assisting with jannying.
no
>>10389c'mon, dude, you're going to put me out of a job
>>10390>I think intentionally training an AI with GR1 material would be a legal grey area that we wouldn't wanna wade intoDoesn't literally every other social media company do this already? This doesn't sound like something unprecedented to me at all.
>>10397It's probably one of those things where you have to be REALLY sure you have all your ducks in a row, and potentially have lawyers and stuff involved to make sure that remains the case, so I imagine it's a big hassle
This thread is for the unofficial extension Hotpocket Helper. Whether you're a janitor of 1 board, a global, or a mod, this extension adds a plethora of features to streamline keeping the site clean!It's also compatible with 4chanX! If you can load your janitor/mod tools, it just werks™.Instructions for installing HPH scripts and the most recent version of HPH can be found in the first few replies. Please use this thread for feedback, suggestions, and any questions.
>>10376>>10377He said Developed Mode not developer mode, that's something else
>>10379No I'm just retarded, it's Developer Mode. It was a typo on my end
>>10380SOCKS. NOW
>>10274>>10291please mod-sama, I had grouped reports for a few refreshes and then it went away. imagine seeing in black and white all your life and then you get three glorious minutes of in-color viewing and you're back to black-and-white. would you ever be able live your life the same way again? I wouldn't. I wouldn't......
>>10386I did go to do it back then, but after looking into it, the reason it broke was related to browsers changing, or maybe tampermonkey itself. I'm unsure. An actual webdev could probably answer that.I could probably do it with some alternative approach, but I have just not yet got around to it.
Unsure how much information on this can be shared, if any, but I'm curious what criteria is considered when deciding to open a new board. Even as a janitor, it's a very opaque process that doesn't seem to follow any consistent logic (observable).I can offer a few examples from the newer boards./vt/ was obviously created due to immense demand for it, and it has quickly grown to becoming one of the most active boards on the site. In contrast, it feels like it's been well over a decade of users begging for some kind of internet celebrity board and those requests have gone unanswered, despite the fact that there is clearly interest in it (/v/ janitors will have firsthand experience seeing this demand from deleting threads on this subject). What's the difference between these two things? Is it just a subjective thing, that the 4chan team does not want standalone boards dedicated to certain topics?Then at the same time, we get boards like /qst/ or /vmg/ that don't even get 1000 posts in a day and nobody was really asking for, so clearly demand isn't the only thing considered, or even necessary. For my part I'm asking this because I think it may be worth considering some kind of /ai/ board for posting AI generated content, whether it be audio, images, video, text, etc. /g/ has several ai related generals and most don't discuss the technology but are instead just dumping grounds for content, and I think there's an opportunity here to separate that space and allow it to scale. It's just hard to argue for this when I don't really understand what the common thread between newly created boards is.
>>9935I don't think there's really much reason to get rid of boards that have already been made unless they have become problematic to keep moderated or they are otherwise inconveniencing the functioning of the rest of the site in some way. Slow and unpopular boards should by their nature be relatively easy to manage anyway so might as well let the few people who like them have their fun.
>>9928>what combined boards that share rough overlap would be the least usable?Merge /a/, /co/, all the /v/s, /tg/, /mu/, /m/, /lit/, /pw/, /sp/, /tg/ and /tv/ into >/ent/: Entertainment Mediumsand watch the world burn
>>9939>merging /a/ and /tv/ would be like when segregation ended in the south and i'm not even sure which one would be Ruby Bridges
>>9935i don't remember if this was my own post but frankly i agree wholeheartedly so yeah it probably was
>>9933/mlp/ has been in steady posting decline since the show ended. If I had to pick 1 current board that will be deleted next, it would be /mlp/.
has the possibility of using an LLM to screen posts been considered yet?taking the global rules and providing them as context for either a cloud based (expensive? prices are coming down quickly though) or locally based (llama or mixtral8b? cheaper, buy a 5090 or something lol and run it there) LLM, then providing it instructions to rate posts based on rulebreaking probability could add another method to get posts flagged to moderation staff without relying solely on user reports. posts above a certain rule breaking threshold as determined by the model could be added to either the report queue or a new queue. thoughts? apologies if this is a dupe of topics already discussed
I don't know anything about the actual numbers involved but I have to assume it would be more costly than you'd think considering the volume of posts here
>>10339no, fuck AI.
>>10345I assume this too. Basic screening via bad word and bad phrase lists is cheap, but throwing EVERY post at an LLM is the kind of "burn money to get an audience" loss leader that people do when they've got VC money and then panic when it doesn't make them dominate the whole market within 18 months.Especially if all it's doing is reporting things that would otherwise go unreported. That limits it to>Things that no user is bothered about>Things that users are specifically trying to keep hiddenJannies and users do a good job of detecting and reporting GR1-related discussion even when users try to keep it subtle.And if an autojanny is flooding the flesh-jannies' queues with things that might be breaking less important rules and that no user is bothered enough to report, that's going to make more work for the janiteam, which means hiring more of us to cope with the output of the LLM janny.Yes, there are some threads where users see rulebreaking shit, don't report it, and then post "OMG why is janny not removing this, the porn has been up for 5 hours", but it is still cheaper and easier for us to try to convince users to report blatant rulebreaking than it would be for 4chan to implement a full blown AI report queue.Not to mention, things that aren't being reported and could in theory be removed by stricter moderation and aren't bothering users - that's the kind of "freedom to argue, shitpost, and be merry" that 4chan has as a unique selling point.
The only problem that I think that ML (and not a LLM btw) would be good at solving is false reports:>user makes a report on a high clear rate board>make a copy of said report with the user's IP and add it to a list>look in the list for users that have more than X reports>feed the list to an ML model written in Python to categorize individual reports as true or false>use Python analyze the results>send the worthwhile users' reports to a false report queue sorted by IPs>a Mods can then look at the false report queue when they feel like it to warn/ban the actual false reportersThis can:>cost as little or as much as you want (mostly electricity costs from running a GPU)>work on as many or as few boards as you need it to (the more reports you have to go through and the larger the training data the longer it will take to run)You get a system:>that can find false reporters without human input>which isn't making any decisions for humans>that streamlines the handling of false reportsComment too long. Click here to view the full text.
>>10339ur tryna take my fuckin job away no fuck u