UI differences are a big factor in the success/failure of decentralised federation of diverse platforms and content
And this seems a good example: bridged #mastodon posts onto #BlueSky which has a low
from maegul@hachyderm.io to fediverse@lemmy.ml on 22 May 2024 00:43
https://hachyderm.io/users/maegul/statuses/112481968174551358
from maegul@hachyderm.io to fediverse@lemmy.ml on 22 May 2024 00:43
https://hachyderm.io/users/maegul/statuses/112481968174551358
UI differences are a big factor in the success/failure of decentralised federation of diverse platforms and content
And this seems a good example: bridged #mastodon posts onto #BlueSky which has a lower character limit than Mastodon.
So, just like #lemmy posts on mastodon, you don't get the full content of the post (which ends with an abrupt ellipsis here) and have to take a link to the original platform.
However powerful the underlying protocols, this isn't far from screenshots.
threaded - newest
@maegul @fediverse The bridge fails to detect line break in Chinese contents, at least from Bluesky to Mastodon
If this is the post on Bluesky in Chinese, it will be displayed in only one paragraph under Fediverse
How do you see Lemmy posts on Mastadon anyway? This comment I understand but what about my posts?
You can follow a lemmy community on Mastodon, and even comment from Mastodon.
It’s not optimum, you see it like the community retooting each post and then the comments as reply to post but it works
… but it’s pretty bad on “both sides” in my opinion.
Especially when all the workarounds to make it barely usable from Mastodon spill over into Lemmy in the form of bots posting weird hashtags or headlines being full of hashtags,…
I wouldn’t really count Mastodon/Bluesky bridging as federation. They’re incompatible protocols that were never intended to work together (arguably Bluesky was explicitly designed to avoid using AP).
Eh. AP isn’t magic. Platforms can be pretty incompatible because of their differing use or implementation of AP. I feel like at some point there’s a blurry line between a bridge being something different and just an extension of 2 protocols.
Definitely, AP is not magic. But if even within one protocol round-tripping and full-fidelity is impossible or very difficult, that makes it only harder and less likely through a bridge.
Yea a complete protocol to protocol bridge seems far off for sure. But a platform to platform bridge seems to be working fine for the moment (masto-bsky), and that seems a reasonable expectation.
I would say federation doesn’t necessitate the use of a single protocol across the network.
Imagine the scenario where very slowly each protocol aligns on features and just requires simple translation? What if one or both start to implement parts of the other spec?
Would you only call it federation if they used identical protocols?
I’d say federation is about the actual content being shared across decentralised services and not the technical means by which it’s shared
IMO bridging or translation isn’t federation per se. Also it seems unlikely that protocols would converge to that extent. In fact AP implementations are already different enough that even within the same protocol it’s hard to represent all the different activities instances can present.