For the majority of commenters: UX is not UI.

The poor UX experience is the research a person has to do before they can even participate. You need to have a basic understanding of how the network works, and then you have to shop around for a server.

It’s enough friction to prevent people from on-boarding and that’s not good for a platform that needs people to be valuable.

They didn’t, they signed up for gmail.

They actually sell a service where you can buy a domain name and have it set as your username.

It’s not centralized, it’s also not a federated network like AP. It’s just a different design.

Not quite true. You can host your own personal data store (PDS), run your own labelling service, and you can host your own feeds.

The big relays and app views are the only thing that aren’t really supported by the official app yet.

Earn points through generating content and moderation? Okay, sure, why not?

Use those points to weigh votes in community governance? Suuuure, okay I can see how that could be cool.

The points are on the blockchain? Uh… so what’s stopping rich assholes from buying up points and using their capital to take over communities?

If the points are non-transferable then I can see the merit of a points system… but then why would you need a blockchain at all? It’s all still a closed off walled garden despite what they are pitching.

Wouldn’t it be prudent to build features into Lemmy that make it easy for admins to manage user data though?

I’ve been following BlueSky closely for a while and I’ll just add a few points here:

1. There is currently a federation sandbox for developers, it’s definitely on the way but it is a significantly different model than AP. Severs are really “dumb” and it has an emphasis on using a handful of services to crawl the network and generate a pipeline of all posts.

2. Moderation and custom algorithms are also a part of the decentralized model. Custom algorithms are out now, and custom moderation services are also under development.

Having played with both AP and ATP a fair amount they definitely both have strengths and weaknesses, very different approaches to decentralized social networking.

I asked this in the original thread but I’ll repeat it here:

1. Are there any limitations with the ActivityPub protocol you find limiting? Do you have recommendations for future versions of the protocol?

2. Do you have any thoughts on the AT Protocol (a potential competitor to AP)?

Have you found any limitations or pain points related to the ActivityPub protocol? Are their improvements you would like to see to the spec?

EDIT: On a related note, do you have any thoughts on the AT Protocol?

I don’t know the answer for Lemmy, but I know this is something a number of Mastodon clients have implemented. It should be possible?