Edit: I’ve been schooled in the comments, this is no longer a thing, they’ve relented and allowed more control over the censor list. But hey, it’s kinda funny still, so, feel free to keep reading if you like…

Original post follows.

Coming here from the Great #RedditMigration. When all of the alternatives were being discussed, one thing I noticed was that Lemmy seems to have one single global slur list that’s literally hard-coded into the software. Your only option (as a server admin) is to either enable or disable it, but the words it’s going to block are permanent and unchangeable unless you can talk the devs into updating the regex for the next release.

And I checked out some of their GitHub issues, and the devs seem almost militantly defensive of this. Like, they are not entertaining any suggestions at all to make it editable. Their actual solution is, if you don’t like it, fork the repo and change it your own damn self, but we’re not touching it, go piss up a rope. Don’t mind one of the words they don’t like? Too bad. English word is perfectly innocuous in your language? Piss off. Want to block the equivalent Djiboutian words? Believe it or not, go to hell.

Seems like such a weird hill to die on. And this is concerning as the lemmyverse is just taking off. This just seems like such an easy and obvious thing to make configurable for all kinds of reasons, and they’re just closing down tickets. Not even a “don’t have time right now but we’ll put it in the backlog and get there soon”. Just, nope.

So anyways. I dunno where I’m going with this really. I’m not in a position to really do anything about it. So this is just a rant I guess. But if the main devs of this thing we’re all migrating to are this… I dunno, arrogant? Opinionated?.. already, then what’s gonna happen later when there are actual real issues that need addressed? We’re leaving one spezhole and going straight to another?

    • ilovededyoupiggy@sh.itjust.worksOPEnglish
      283·
      1 year ago

      I’m glad they’ve relented and made it editable now! The threads I’d seen mastodon about it, and the links to their GitHub tickets, were pretty disappointing to me. Glad it’s not the case anymore.

      • nivenkos@lemmy.worldEnglish
        17·
        1 year ago

        FWIW it’s been this way for well over a year IIRC.

        But the instance you post on needs to have it configured reasonably too.

  • nivenkos@lemmy.worldEnglish
    14·
    1 year ago

    It has been configurable already, for at least a year if not more.

    I wish it weren’t enabled by default too though.

  • thirdorbital@sh.itjust.worksEnglish
    121·
    1 year ago

    I naively want to believe that this won’t be an issue for most users having most conversations. But out of curiosity can you link what the ban list actually contains?

    • wildbus8979@sh.itjust.worksEnglish
      9·
      1 year ago

      I ran into this earlier today, I couldn’t even figure out what the word was. Rewrote my message three times, and then gave up.

      • New_account@lemmy.worldEnglish
        11·
        1 year ago

        Gotta love when classic cl***ic profanity filters are implemented without learning the mistakes that have been made again and again over the last 30 years.

        • TheTedJamesExperiment@kbin.social
          81·
          1 year ago

          From Software games have filters for player names and it often leads to stuff like this. Thus if you put “knight” in your character’s name, a common thing that people do given that they’re medieval fantasy games, it appears as “k***ht.” Likewise “dead” becomes “d***” and “Thomas” becomes “T**mas.” Even more embarrassingly, the filter is case sensitive and all the filtered words are lowercase, so you could name your character a slur and it would be fine as long as you capitalize the first letter.

        • taladar@sh.itjust.worksEnglish
          2·
          1 year ago

          It is known as the Scunthorpe problem after a town in, I think England and a certain word it contains that is often blocked.

      • MeowdyPardner@kbin.social
        7·
        1 year ago

        I think you have to ask the admin of sh.itjust.works what they set the filer to because it seems customizable now.

  • PriorProject@lemmy.worldEnglish
    10·
    1 year ago

    Have you confirmed that the word list is still in play? I thought after much debate they killed it a decent while ago.

  • ReaealEnglish
    112·
    1 year ago

    Romanian speakers: Is that a beech tree?

    Lemmy:

  • AnonymousLlama@kbin.social
    6·
    1 year ago

    I think from memory that the kbin dev was talking about getting “AutoMod” enabled on the site to help future moderation so having something like that would give you more options over how content is held in moderation.

    I assume once AutoMod gets integrated there won’t be a need for a global ban list anymore.

  • Protahgonist@kbin.social
    94·
    1 year ago

    Reduced because this is already no longer an issue, making this an amusing rant but also disinformation. In any case, come to kbin! It’s very nice.

  • Googleproof@sh.itjust.worksEnglish
    51·
    1 year ago

    Spez is a symptom of reddit’s problem, not the problem itself. Put someone in who wants to help the users more than the investors, and the investors will just kick them out. Which is why I’m more sympathetic to the Marxists around here.

    So far as a ban list goes, it’s better to be hard-coded than a database call for performance reasons, and if it stops 95% of bad actors, that’d be a good thing. Problem is that it won’t, they’ll find a way around the regex. So yeah, this is something that should be handled by mods or communities.

    It provides a minor inconvenience to people wanting to set up alt-right instances, and makes them feel unwelcome, so maybe that’s enough to justify it’s existence.

  • lohrun@fediverse.boo
    31·
    1 year ago

    You could always join us in kbin land ;) @ernest has been a really supportive dev and there seems to be more devs over here willing to help him work on it

  • 🐱TheCat@sh.itjust.worksEnglish
    1·
    1 year ago

    ‘believe it or not, go to hell’ is my new catchphrase and you can’t stop me

    but also maybe throw an edit at the top of your post to say this is a solved issue if you want to prevent spreading disinformation.

  • BuddhaBeettle@kbin.social
    21·
    1 year ago

    Since that issue seems to be resolved, just came here to say, I did experiment this issue on a smaller scale on reddit.
    Lets just say a very common color and sometimes even word of endearment for your loved ones in my native language is a common english slur (and a very bad one at that). I would be telling my friends of the x colored dress I bought via chat and the message would be flagged down and would not send.
    I had to be very careful cause I was afraid my account would get banned one day.