XSF Discussion - 2021-12-26

>they also disabled the connection with public servers Reminded me of this speech https://www.youtube.com/watch?v=FEU632_Em3g Iirc he even mentioned xmpp as one of the technologies that have "already failed" to stand against "antidisintermediation"

XEP-0049 doesn’t define any discovery mechanism, yet Prosody exposes it on the domain JID, poezio checks for that, possibly other implementations also do.

In the process of phasing it out, I’d like to keep support for the one specification still in wide use making use of it: XEP-0048 version 1.0.

TIL http://archive.jabber.org/docs/jpg/

It feels wrong to advertise support for jabber:iq:private on the domain JID when bookmarks payloads are the only allowed ones, but clients (like poezio currently) might not even try to use bookmarks when the jabber:iq:private feature is not set.

TIL!

So, what would be a good way forward? Advertising the feature but erroring out on any write to a different store than legacy bookmarks?

That would work.

Not that many other uses of 49 remains in active use AFAIK

Here are some statistics, on jabber.fr only: https://linkmauve.fr/files/private.txt

But which have been used this decade?

I have no data on this.

Perhaps we should convert all this to '223 with a '49→'223 conversion layer.

Would this bring anything?

It’s not like JWChat or Jappix is going to be updated to use 0223.

I mean move all the data into PEP and get rid of the old '49 code

That would make all of these clients not work any longer.

It would if there's yet another compat layer

Move *all* the data into PEP!

porn spam

spam

spam

/ban fdsa

spam

spam

spam

spam

spam

spam

spam

spam

spam

porn spam

spam

spam

spam

spam

spam

spam

spam

spam

spam

spam

spam