[Warning: I didn't read her full report yet.]
I agree that it's flawed. As of now, to my knowledge:
- users can report content and state the reason (e.g. "kgshkewq");
- moderators can read the reports;
- moderators can remove content, and ban users;
- admins can set up the registration method, to either "free for all" or manual approval.
It works in a small scale, but once it gets big enough, it'll become a mess.
However I don't agree with her IMHO dichotomic conclusion that "if you care about user safety, do not deploy either of these" (implication: "if you deploy either Lemmy or Kbin you don't care about user safety"). It's a more weighted decision that depends on:
- scale (how many users),
- how much do you think that people would be willing to moderate it (brute-forcing lack of mod tools)
- the nature of the group that you're trying to migrate (is it a marginalised group often attacked by society? or just a bunch of randoms sharing meme?)
Plus discouraging people from migrating creates a chicken-and-egg problem. Open source projects rely a lot on user contributions; specially for code. If both Kbin and Lemmy are used enough then someone will eventually code the missing tools.