I am wondering if there is a discord (ironic i know if there is something equvialent let me know) where I can get into contact with people developing in this space.

I am hoping to get some questions answered such as:

How does a different architecture like kbin@social ui (and possible backend) communicate with posts from lemmy? Is it through ActivityPub?

Could someone from lemmy login to mastodon with their same account? Why or why not?

Could two different fediverse apps technically do all the same things if they used the exact protocol to do the same things with a different ui?

If someone wanted to go about making their own fediverse app where is a good first learning step?

you are viewing a single comment's thread
view the rest of the comments

[–] nosurf 2 points 1 year ago* (last edited 1 year ago) (1 children)

If you want to learn more about this, I’d recommend setting up a private Lemmy, a private kbin, a private mastodon, and maybe others you’re interested in. Try to federate them and see how it works and what it looks like from each app.

This is a great idea, had not thought of this.

Thanks for all the answers, it fit together exactly what I was missing.

I will say this, it from my small exploring of this it feels weird to me that all these fediverse apps dont share a sso or openid auth for all of them. Like since mastodon is the big player in this all other apps should support using them for auth to get easier entry into all of this as a new comer. Though this is kind of all about decentralization so maybe not?

[–] PriorProject 2 points 1 year ago* (last edited 1 year ago)

I will say this, it from my small exploring of this it feels weird to me that all these fediverse apps dont share a sso or openid auth for all of them.

I equally found this weird, and it kept me out of the fediverse for a long time even though I was following the tech through non-federated news sources. "I don't want to choose a server" is also a very common UX critique of the fediverse. I don't really have the federated app design chops to critique this choice, but I speculate it went something like this:

Your "account" is more than your identity. It's also your data. So I have a post and comment history on Lemmy.world, and they would have to store that and federated instances would need to be able to find it even if both supported oath to determine my identity.
And data is "heavy". You don't want to make the choice of home-server transparent and then have a heavy user with thousands of posts or followers hopping around from instance to instance dynamically all day. And most volunteer fediverse instance admins want to approve account sign up anyway because they don't want to host child porn or some wildly aggressive idiot so nobody is going to enable highly dynamic account migration anyway. So even though you have federated identity, you STILL need a home instance and a manually initiated account migration process to handle all the expensive data movement.
So supporting federated identity doesn't actually elide the need to make a server choice and doesn't make your account all that much more portable. In fact, it adds an ADDITIONAL choice for your identity server. And lots of fedi people get persnickety about the major oath providers. I could imagine tons of fedi instances saying "fuck google, we're not accepting them as an oath provider", and now the choice of identity provider becomes every bit as confusing and fraught as the choice of home instance, and you need to pick both.

And of course, federated identity is super complicated. I suspect people decided it just didn't pull it's weight in complexity. Anyy, that's a lot of wild speculation... but it's my theory about why federated identity never quite took off in the Fediverse.