real

And your takeaway should be to ask why every programming environment you've ever used been a hemorrhaging regression on the mother of all demos?

The people who create the protocol, and the people who can write the software that interacts with it ultimately have power over everybody else. That's not great, and we should consider alternatives. It is still better than one alternative, one party with total control.

That protocol could be the basis for a meta medium that itself could be implemented once and all the new unanticipated, unimagined features that people would want to create could be expressed within it.

Protocols are just how you do cooperation, you have to agree on enough to do communication. In the blog post they advocate for implementing several in one client. And the client somehow unify their UX, possibly with a protocol.

The people who create the protocol, and the people who can write the software that interacts with it ultimately have power over everybody else. That's not great, and we should consider alternatives. It is still better than one alternative, one party with total control.

I don't think it puts the onus on people running homeservers. The onus is put protocol design.

>In other words: The problem about using a protocol is that you cede control to it.

Can I please have personal dynamic media, can I have networking meta media please 🥹

Lack of introspection about decisions (context, appealing, reversing), having to choose between different walled off communities, the different admins vying for power basically is the problem I have with fediverse. Those are genuine moderation issues though, and isn't the same problem as abuse+spam

Is it less relevant? You might have a choice over who you can trust but that alone doesn't make any option trustworthy.

Also, being able to move homeserver is great but there will still be situations where you have to make a mutually exclusive choice about who you can associate with from where

Distribute the tools of moderation to users, get as much moderation done at the user level so there is less to do at the PDS level.

Though no doubt that isn't the full story and there could have been some other political motivation behind that.

And that's whether GDPR is going to let you just label arbitrarily in the first place. The browser extension Shinigami Eyes for annotating transphobic web content got spat out by the Norwegian GDPR regulator: "subjective assessments will have unfortunate consequences for the freedom of expression."

On the other hand you also can't really do this without also defending against what happens when someone you want to associate with gets hit with a label you disagree with. Whether the label is is intentional or not won't matter because for any naive consumer the consequence is the same.

hammering out some skeets with the team

Ok, but is it going to be necessary to do that again for growth?

But if this has to be, I'd recommend having this be either a distinct invitation or an annotation on the invitation rather than a public annotation of their profile. Because it needs thinking through whether "Trust" via a vouch should just be a unary thing in one context alone.

Wasn't the invitation a signal of your trust that they should be able to have an account? What exactly are you vouching they are trustworthy of if not that? I have to ask why are you inviting them?

Distributed moderation is an ideal that is more aligned with what people really need when they look to bullshit AI solutions (not saying AI can't be a component of that but shhh, it's not nearly a silver bullet). And you actually end up "automating" most of the role of a moderator.