Lemmy has hard coded censorship. It can be disabled via the code, but irrc not via config. It was a large source of criticism, even 3 years ago: https://linuxreviews.org/Lemmy
It is a server config option. Lemmy.ml uses a banned word list, while other instances don't. The words are still there and visible from instances that don't censor.
Are there any instances that you know of that disable the censorship? I already got my fill of cybernannies on Reddit.
Edit: My alt on sh.itjust.works can see the censored words, and it defederates with much fewer instances than lemmy.ml. Seems like migrating over there is the best choice.
Edit 2: I think I've finally figured it out based on what my sh.itjust.works account shows:
If you're on a censored instance, all words will be blocked no matter what
If you're on a non-censored instance looking at a comment from a censored instance, that comment will still show up as removed
If you're on a non-censored instance looking at a comment from a non-censored instance, the comment won't be censored
When I'm on my sh.itjust.works profile, the comment by @[email protected] isn't censored, but mine still is. So, the censorship gets you both coming and going - no censored words can make it into a censored instance, and no censored words can make it out either.
I've just edited all of my scripts to point to my sh.itjust.works profile instead of lemmy.ml. Wish there was a way to migrate comments and posts over, but no biggie.
Has anyone made a multi-Lemmy yet? Where all of your posts and comments from all of your profiles can show up in one place?
(also, what's the formatting for double-spacing after a list?)
I've noticed that a few of my comments wouldn't actually post. The word for when you commit Sudoku (suicide) seems to be on the no-no list. I also wanted to comment about a weapon known by a two letter designation (AK) and that wasn't allowed either.