[META] "Free speech," throwaways, and parameter tuning
[META] "Free speech," throwaways, and parameter tuning
I'm leaving the pro-Zionist comment up. Long story short, it's abundantly clear to both me and the bot that it's a real user stating their real opinion. I don't want to create a "free speech safe space" for all the comments that aren't welcome anywhere else because they are horrible, but I also don't want to create a place where there are elaborate boundaries about what you can and can't say. The rules are: No bigotry, no spam, no personal attacks. Not that you can say fuck one person, but not this other person.
I severely don't like the "nanny state" moderation that prevails on a lot of Lemmy. We're all adults here. Hopefully. Even with the tiny amount of content that's been posted so far, I've been batting down reports that are based on the idea that the user base can't make up its own mind about what's trustworthy, what's true, or what's worth reading. This is the first comment I really wanted to delete. To be blunt, my personal opinion is that it's garbage. On the other hand, lemmy.world has political threads that are overflowing with garbage comments, so if we only have one so far, we're beating the curve.
The post about US terrorism should have been removed, only because it came from a throwaway. That part of the bot isn't working yet. I have to change the parameters. I'll get to it soon. However, if it had come from a normal user's account, I wouldn't have a problem with it. The comments look fine. Again, we're adults, or else hopefully can act like it. The problem with posts like that comes when they start dumpster fires of comment sections, or drown out the good content, but that hasn't happened yet! I think that's a really good thing. It means getting rid of all the people who love to start big pointless fights has been working so far. Running across content every once in a while that you think is dead wrong isn't a problem. It's good for you. It builds character.
None of this is set in stone. I'm not trying to make a safe space for horrible content, and if it becomes that way I'll set stricter boundaries. But I think a couple of controversial posts that didn't lead to big bitter shouting matches shows good things about this moderation model, not bad things.
The problem with posts like that comes when they start dumpster fires of comment sections, or drown out the good content, but that hasn’t happened yet!
This usually only happens when threads hit the front page of the all feed and people that are not subscribed to the community see it, vote on it and start commenting in it (which then becomes a self-reinforcing system that pushes it further up the "hot" rating on the all feed).
This community is currently too new and small for that to happen.
As for pro-Zionist comments... if they come from an account that is not only posting such and it isn't outright genocide denial, I agree that it can stay up. But this will likely need human intervention and can't be left to the bot to decide.
And that's why I try to be careful about where my thumb it's when scrolling the all feed lol.
This usually only happens when threads hit the front page of the all feed and people that are not subscribed to the community see it, vote on it and start commenting in it (which then becomes a self-reinforcing system that pushes it further up the “hot” rating on the all feed).
This community is currently too new and small for that to happen.
I'll wait until I can put in place the throwaway account sniping, and more testing, before I try to do much more to promote it. The wider level of attention from [email protected] seems to be a good test which the bot hasn't caught up to be able to handle completely.
As for pro-Zionist comments… if they come from an account that is not only posting such and it isn’t outright genocide denial, I agree that it can stay up.
Yes, that user posts almost all normal content, with a tiny minority of unpopular but still "normal" political views, and a couple of posts that are openly Zionist. They're nowhere near posting a majority of inflammatory content, and the comment wasn't even that bad, it just seemed shocking because it was so pro-Israel, which usually doesn't happen.
But this will likely need human intervention and can’t be left to the bot to decide.
I completely agree. I didn't plan to have the bot replace human moderation, only provide another tool to automate one part of it.
Anyone who is breaking the few rules that do exist, I was planning to ban. I also just edited the sidebar to make it clear that comments must also follow the slrpnk rules.