XMPP Council - 2022-12-21

I'm not going to make it today.

ack, thx for the notice

hello all

o/

Hi everyone. It's time

1) Roll call

o/

o/

Ge0rG said he won’t make it. is larma here?

I guess not but moving on

2) Agenda bashing

nothing to bash

3) Editors update

the editor published three new XEPs

a) NEW: XEP-0475 (Pubsub Signing) (https://xmpp.org/extensions/xep-0475.html) b) NEW: XEP-0476 (Pubsub Signing: OpenPGP Profile) (https://xmpp.org/extensions/xep-0476.html) c) NEW: XEP-0477 (Pubsub Targeted Encryption) (https://xmpp.org/extensions/xep-0477.html)

and one protoxep

yes

Here, sorry for the delay

4) Items for voting

hi larma :)

a) Proposed XMPP Extension: Stream Limits Advertisement (https://xmpp.org/extensions/inbox/xep-sla.html)

+1 looks straight forward and useful

(larma, could you stay around for a minute after the meeting? I've got a question regarding dino :))

+1 for sla

+1 for xep-sla

+1 though I'm not sure it solves anything if remote limits are lower :'(

5) Pending votes

none

6) Date of Next

I'm going to propose +2w

that sounds appropriate

Works for me

I'm off work in calendarweek 1, so as usual I might be completely out of my rhythm and may require gentle poking

I'm fine either way but makes sense as I assume we aren't going to have a lot of things to vote on next week anyway

MattJ, Zash, re xep-sla, you mistyped XEP-0153 as XEP-0053 fyi.

Oh no

OK. +2w (Jan 4th) it is

7) AOB

none from me

ha good catch Link Mauve

nothing here

RE sla: I'd love if the error message would contain machine readable limit information (to also solve end-to-end issues, even if it may take multiple attempts)

8) Close

Thank you all. Happy holidays. see you next year

thanks daniel :)

Also RE Link Mauve comment: I wasn't sure if it was mistyping 0153 or 0054 :D

happy new year's then!

Thanks all and happy holidays

larma, question: what's the prospect of '392 (consistent color generation) in dino? ✏

thanks all and happy holidays+new year etc etc :D

larma, I also wondered, but went with the most likely candidate. ^^

jonas’: good question. I think it's implemented already but not used for anything. We weren't exactly happy with how it looked, but we might want to revisit that and see what we can do.

I've been using it for months (had rebased your commits) and I'd love to not have to rebase regularly :)

if there's anything which can or has to be done spec wise, you know where to find me :)

Zash, that XEP is very useful but I can't come up with a good solution for when a remote S2S limit is lower :'(

moparisthebest, I don't understand and now is not a good time to process anything (for me) ✏

could you write down what you think the problem is somewhere? mail to standards@ for example

client connects to their server A, the limit is 100 bytes, client sends a 99 byte message to remote server B, but remote server B's limit is 50 bytes

I guess server A doesn't bother trying to send it to B and instead sends an error to the client, but that's pretty rough

better than the s2s connection being closed and anything that was in transit being lost

I agree

which is what happens regularly right now, any time anyone requests the vcard of someone with huge avatars

The only "solution" would be fragmentation and that would get a big NO from me...

but the point is to make the limits clear to both parties, so then they can abort early if they find they're about to send something too big. doesn't the protoxep text say this already?

you could have a way to query your server for the limit to remote server B, but that gets pretty rough too

The limits could be different depending on the stream though.

The working name for this was "pathmtu". :)

I typed "query your server for the mtu" but then backspaced lol

Does Council want us to announce new XEPs or something like this? It could look like this: https://fosstodon.org/@xmpp/108058245120853596 The only thing I'd require is what info should go in etc. and maybe a ping if something new is out

I observe that this SLA still waits for Ge0rG

emus, yea I think it at least needs waiting until editor publishes them, but you could announce the 3 new ones

But is it correct to wait or not? Lets go only the official way

In genereal such posts get quite some attention, at least the last times there were way more shares than for other posts

larma, re machine readable limits in errors, something like https://xmpp.org/extensions/xep-0363.html#errors ?

I have another half-finished protoxep full of various s2s related error conditions

Zash: yes, like that

I made a note

thanks