XSF Discussion - 2019-09-16

Zash, rion, the people doing Renegade X started writing a PRECIS implementation suitable for use from C, or at least are planning on that.

I don’t think any code has been written yet.

Join chat@xmpp.rs (xmpp:chat@xmpp.rs?join) to get in touch with them, if you want to help.

Link Mauve, that C&C Renegage remake?

I like the idea of MIX, but whenever I look at the specs I realize how much cleanup is still needed

I like the idea of XMPP, but whenever I look at the specs I realize how much cleanup is still needed

hah

jonas’: you could run for Council.

are the elections open again?

ah, they are

oh, Daniel’s coming back

I need to find all the banners I made the last time.

https://upload.yax.im/upload/L9QpGat3GD9S3nMF/make-xmpp-great-again.jpg = 404

Who made this shit?

sounds trumpy Ge0rG

that’s from back when Trump hadn’t won yet and it was still funny

it is still funny, just like telling people that wa is perfect for pupils and at school. that's also funny 🙄

ladies and gentlemen, we have a MIX create/join in aioxmpp! https://paste.debian.net/hidden/2b9100eb/

gg

flow: would you accept a PR containing the Oob.java file in the appropriate Smack namespace (which one?): https://github.com/yaxim-org/yaxim/blob/master/src/org/yaxim/androidclient/packet/Oob.java

I'm currently writing slides for a developers introduction to xmpp

Have we ever documented the session optional child thing?

Or is that just something people need to know? Lol

Hm, hold on

Daniel: https://datatracker.ietf.org/doc/draft-cridland-xmpp-session/

oh nice

how would people usually discover that?

that draft i mean

Even nicer would be to get that finished and RFC'd

Ie an RFC that updates RFC 6121

Daniel: what do you mean discover? If a server presents that child element it supports it?

no how a developer would know about it

Ah, well, it is an expired draft, so I assume the answer is they shouldn't?

(casually mentions dwd)

It documents something that solves a problem and that is implemented.

Well if that's the case, indeed it should progress, because it adds a new element to an existing namespace.

I didn't know it was actually implemented.

I think both ejabberd and prosody do

At least

Yes

Daniel: so the normal way would be this document progressing to RFC and being recorded as updating 6121. It would then show up at the top of the latter's document.

good to know thanks.

Last discussion on this seems to be the same day the -01 was published, on the xmppwg list, by Kev and dwd.

Discussion from then suggests M-Link also does it

so does openfire

Are there clients that still fail if you stop advertising <session/>?

dunno. are there servers that still fail if you don’t send session?

dunno. haven't looked at this since that draft

Aw, MUC logs from that time lost in the crash? :(

aioxmpp also supports that optional thing

Daniel, jabber.ccc.de failed back when I started aioxmpp, not sure if it has been updated since then

The optional thing is fairly widely supported I believe. Gajim most likely as well. However I didn't know about that draft and thought it was just collective consciousness

It existed before the draft, IIRC waqas may have talked about it before that

https://hg.prosody.im/trunk/rev/0bbbc9042361

Prosody 0.6

I think I remember that ejabberd only got that relatively recently. Circa 2016 or so.

And as jonas’ pointed out ejabberd in 2015 even required it

2016 indeed, if this is it: https://github.com/processone/ejabberd/commit/a150bf8fdc8061bc60e5cb1a64a1e06f962d6a4e

Ge0rG, sure. BTW, I am not sure if discussing Smack related stuff is really on-topic in xsf@, consider using smack@conference.igniterealtime.org, where we should continue discussing the details

flow: or you might join jdev@, where it's also on topic.

Zash, oh wow, I was just assuming jabber.ccc.de was running old software

jonas’: I fell for that as well, recently