The problem is that a significant chunk of the community believes that it's not hate speech but simple hyperbole in response to uni-directional injustice.

It's easy to string match, bans/deletions can be instant and abrupt. They see the quick removal of their post and compare it to what they perceive as genuinely hateful posts that stay up because the post cleverly avoids violations that are easy to include in content policies.

Check out the "men are trash" question here https://www.theverge.com/interface/2019/10/3/20895119/facebo...

And I'm not saying you're wrong. Lots of people do think their hate speech is simple hyperbole, or that it's justified by misdeeds of the group they hate. But what's the point of a rule against hate speech if you're not willing to tell those people no?

I don't read the parent as advocating the opinion that the people subjected to the "less dark" form shouldn't consider it a kind of hate speech or shouldn't feel offended by it. The context is moderation, and why it makes sense to treat the cases differently - that doesn't mean simply always allowing one and always disallowing the other. The point is that prudent and pragmatic judgement is necessary.

That doesn't really change my perspective. I'd be very frustated if a moderator told me "well, we generally don't allow hate speech, but let's be prudent and allow it when it's targeted at you".

My point is more that prudent and pragmatic judgment is nearly impossible, if your goal is to be seen as a neutral platform for a large collection of communities.