Obviously I cannot quote here because apparently meta discussion is against the rules but as you can see there are a lot of words that are far from being slurs on that list.
Language is highly contextual and dynamic. I thought we already figured out in the 90s that “badword” filters don’t work and are absurd; yet here we are!
Finally it’s a hardcoded list of words which is just silly and amateur no matter how you look at it. For example you are not able to start a non-english instance of lemmy without hard fork of it because of hard-coded slur list that will capture non-english words.
The slur detection is very amateur as well that can be avoided by a single extra character (like empty unicode characters) so it requires manual moderation anyway — why is the filter even there if the only thing it’ll catch is false positives? It’s just such a perplexing UX design that is nothing but a virtue signaling.
deleted by creator
You should take a look at the slur list in lemmy’s codebase
Obviously I cannot quote here because apparently meta discussion is against the rules but as you can see there are a lot of words that are far from being slurs on that list.
Language is highly contextual and dynamic. I thought we already figured out in the 90s that “badword” filters don’t work and are absurd; yet here we are!
Finally it’s a hardcoded list of words which is just silly and amateur no matter how you look at it. For example you are not able to start a non-english instance of lemmy without hard fork of it because of hard-coded slur list that will capture non-english words.
The slur detection is very amateur as well that can be avoided by a single extra character (like empty unicode characters) so it requires manual moderation anyway — why is the filter even there if the only thing it’ll catch is false positives? It’s just such a perplexing UX design that is nothing but a virtue signaling.
deleted by creator