jdev - 2022-08-17

sam, what’s the difference with Lemmy for instance?

Or when you say room, what’s the difference with XMPP rooms?

Link Mauve: scuttlebutt is more like Facebook (you follow people and get their long form posts). Room servers are just how you connect with new people and share posts they're more like servers in irc, you can connect to multiple and get posts from the people in them

So pretty much like Mastodon?

Link Mauve: except more long form and the network architecture is different (servers don't store posts, they just connect users, things are more or less p2p). I haven't really decided if it's good or not yet, still experimenting so it would be good to have a few more contacts

I’m not the right person to use such social stuff, even web forums were too much for me twenty years ago.

I'm confused by the fallback part of the message replies spec: https://xmpp.org/extensions/xep-0461.html#compat The given NS is 'urn:xmpp:feature-fallback:0' but I don't understand which xep it refers to. is it https://xmpp.org/extensions/xep-0428.html or a brand new one https://xmpp.org/extensions/inbox/compatibility-fallback.html ? Looks more like option 2, since <fallback> has a "for" attribute in the example. Any thoughts?

nicoco_, thought: search for clues in the council room logs

will try that, thanks 🙂

https://mail.jabber.org/pipermail/standards/2022-June/038860.html / https://logs.xmpp.org/council/2022-01-19?p=h

Hm, does that mean Movim implements a ProtoXEP? :/

That's ... uncomfortable

nicoco_, anyway, the examples are not normative, don't stare yourself blind at them

and exercise caution when implementing Experimental XEPs, they're subject to change at a whim and then you will not be having a good time if you deployed it already

thanks for the mailing list and council logs pointers, that's very useful. i thought message replies were a bit more mature than they are… but experimental xeps are "encouraged in exploratory implementations" and this is exactly where I'm at :)

Not even a year old yet

It was submitted at the same time and by the same authors as Compatibility Fallback so I imagine they expected them to be accepted together.

But now we're in the awkward situation that these things are already deployed and used in the wild, which will be messy if any breaking changes are done, as authors are basically free to do during Experimental