How Lemmy could interop with Mastodon, as imagined in Frontpage + Bluesky (unravel.fyi)

submitted 2 months ago by erlend_sh@lemmy.world to c/fediverse@lemmy.world

63 comments fedilink hide all child comments

Hey 👋 if you don't know us already, we're building Frontpage; an AT Procol based federated link aggregator. We shipped an initial MVP in closed beta recently and have since been thinking about the road to general availability.

This post is an RFC (Request for Comments) targeted at technically minded folks who are interested in seeing the progression of atproto for non-Bluesky/microblogging use cases. All that's to say the language that follows assumes some knowledge about how Bluesky and atproto work! I've tried to include links to explain what all of the jargon means though, so hopefully it's not entirely nonsense for folks a little less familiar!

When you post on Frontpage, we propose that a mirror post will also be created in your Bluesky account. When you comment on Frontpage, we propose that a mirror reply will be created in your Bluesky account.

Conversely, when you reply to one of these mirrored posts in Bluesky - we will show it as a reply in Frontpage.

Additionally, Bluesky likes will be translated to Frontpage votes and vice versa.

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

[-] aasatru@kbin.earth 2 points 2 months ago

If I understand correctly, there's a central pump running behind the scenes in any AT implementation. You feed content into the central hub, and it pumps it out to everyone connected to it. Bluesky itself provides the one major pump that feeds its network right now.

So in that sense, Bluesky is a centralised network with decentralized users.

Frontpage is building a different pump, spreading different kind of content to a different type of platform. So there's no obvious connection between the Bluesky pump and the Frontpage pump - that's why they're talking about bridging in the post.

It almost seems a bit silly - in order for two AT hubs to talk, you need to build a bridge for them. At that point, you could might as well have built an AP protocol and made it work with Bridgy.fed.

Furthermore, all "instances" running Frontpage would process data through the same central hub. If that goes down or they run out of funding, it's all over.

I'm applauding the Frontpage crowd for trying something new. But I'm not entirely convinced I see the benefit compared to what we're doing over here.