XMPP Council - 2011-11-09

:)

step 1: build a robot

Accomplished

Right, I'm here now, let's see if I can remember to be here in 20 minutes. I'd completely forgotten until I sat down at my machine.

heh

Kev, don't worry, I'm here trying to read XEPs and stay awake

I managed to read them earlier.

MattJ: those will put you to sleep!

Indeed they are

It's been a long day

MattJ, I think you'll find the days are actually getting quite short, now.

Longer and longer for me

MattJ, Are you moving close to the speed of light?

Working on it

I feel like I've been fired through a collider if that's what you're wondering

I think the constants of time won't be applying to me for the rest of the year

I misread that as colander.

haha

Assumed that MattJ was feeling drained.

The colander of time?

Or perhaps strained.

It just opens up a whole vein of bad puns, doesn't it?

Yes. Yes it does.

yay, got my inbox under 1500 messages

dwd, for a series of puns based on something I didn't say? :)

did you unsub from 82attendees@ietf.org? (-;

dwd, your standards are slipping

stpeter, Unrecoverable storage failure?

MattJ, These are my standards, and if you don't like them, I have others.

dwd: I wish

I do see Ralph online, shall I ping him?

Can do, I did so a couple of minutes ago.

ok

Hola.

Right, are we sitting comfortably? Then let's begin.

1) Call for assorted baked products.

I'm here.

目前

I'm here, in body at least

linuxwolf: nice!

And assuming Ralph's join indicates he's here, let's continue.

2) Account management.

I'm couch surfing

We said we'd ~vote on accepting this once we'd had community feedback and the author had responded.

I saw a number of comments suggesting this was not the right approach, and I don't remember any in support of it; is that about right?

correct

even a call to move (part of) this to another venue

For my part, I've got two main concerns with it:

1) Using stream features for this is Wrong. 2) The XSF can't be the appropriate place to develop a new security model.

/nod

#2 is rather significant

I might be persuaded to let it through if just 1) was the problem, but when 2) is in the same document, I don't think we can even put it on the vine.

my thoughts exactly

I think that makes sense

I agree

If we're going to rework all the clients and servers to use a new IBR protocol (when the current one is working ok for most purposes) then we should make sure we go about it properly

we can use security models or promote the development of them

but we need Those Guys™

So I think my suggestions for the author are to 1) look at a more appropriate way of managing user accounts than stream features (the suggestion of first binding with ANONYMOUS and then doing something quite like iq:register sounds right to me). 2) Look at getting the new security model standardised through the IETF, so a XEP can work off the base of people who know more about this than the XSF.

I'll write a mail saying as much when I do the minutes.

I'm reading this as everyone being -1, is that correct?

/agreed

+1 2 -1

(I know, I'm approaching dwd's standards)

That's asserting that dwd *has* standards...

True

Moving on then.

3) XEP-0258

Kurt's updated this, and would like us to vote on moving it to Draft

You need to include names :)

Oh, yes

(Although he made it clear he only wanted it voted on if the vote was going to pass it :))

Kev aw!

Kev, oh, that's that then... :)

I guess he doesn't play the lottery much (-;

Well, it all seems fine to me... I don't know the current implementation status of the new version though

Anyway, we've got this implemented in Swift, and the other we have this implemented in M-Link.

linuxwolf, He does, he only plays if he's certain to win, though.

Ok

And it seems to work ok.

Kev, Two implementations, strangely, for Gajim.

Well I'm +1 to draft

(So I'm +1)

linuxwolf / ralphm?

Kev, Also Prosody server-side. RUmour of another XEP-0258 client or two soon, as well.

surprising

.

hmm, major lag for some reason

wfm

I am +1

I'm tempted to vote −1 on the principle of risk assessment (-:

but I am +1

Marvellous.

4) http://xmpp.org/extensions/inbox/correction.html

+1

I'll give Kurt a stern look next time I see him

(I'm in favour, natch)

A XEP by any other name would be the same protocol, FWIW

apparently dwd had such a nightmare on this he couldn't articulate it

MattJ: Right, I think we can quibble about names when it's on the vine.

I am still concerned about the change of any arbitrary past message

linuxwolf: Which is why it says not to do that.

ralphm, Only with changing arbitrary past messages.

I think it is a terrible feature for recognise that some people want it and don't object to the spec being experimental

dwd: I forgot my sarcasmicon again?!

wow…lag and burst

For those not keeping up, this version (unlike the version I put up an age ago) is correcting the most recent message only :)

I've no objections to publishing

noted :)

Excellent.

5) The bas64 stuff in XEP9.

I'm +1

but I don't see text saying a client MUST NOT correct a message that is not it's own

+1

linuxwolf: wait what?

linuxwolf: Ok, I'm happy to tidy that up. It explicitly says that you send this to correct the most recently sent message.

lag

gawdamit

linuxwolf: how would that work anyway?

I'll save any further comments until I see xep-correct on the list (-:

re XEP-009 … +1

+1

I somehow missed Kev's two XEP-0009 messages

I thought linuxwolf's lag was going into reverse

6) XEP-0068 v1.2 http://xmpp.org/extensions/tmp/xep-0068-1.2.html http://xmpp.org/extensions/diff/api/xep/0068/diff/1.1/vs/1.2pre1

mindlag (-:

Wait, where did this come from? :)

So, I missed the RFC leading up to this, and I don't really like it, so I need to go do some list reading, I think.

heh

no hurry

I did read it, and this would be a possible conclusion

see http://mail.jabber.org/pipermail/standards/2011-October/025341.html

It's an RFC already? Or still a draft?

etc.

I'd like to just pass this off to the next Council instead.

I'm still wondering about suggesting a way to avoid conflict (in case no registration takes place), in the distributed extensibility sense

Kev: sure

MattJ: I use the common term, not the right term :)

Ok :)

I thought stpeter might be fast-tracking his own documents for a moment :)

MattJ: should go to Working Group Last Call at the IETF a few weeks from now, I'd expect

7 http://xmpp.org/extensions/inbox/muc-unique.html

as I said, no hurry

Sigh, I'm still torn on this...

I'm ambivalent

I guess I think overall it should be published, mainly because it already has implementations

I did like the suggestion of something like clark's notation, not the one form to rule them all

A kind of non-historical historical

right

So, are we publishing?

Nobody has to use or implement it if they don't want to

But people have

MattJ: are those implementation using that unauthorized namespace?

So yes, publish

The what what?

http://jabber.org/protocol/muc#unique

no objections to publishing

ralphm, why is it unauthorized?

Prosody uses 'http://jabber.org/protocol/muc#unique' it seems, yes

because, you know, it isn't a XEP yet, it is a new namespace and we use different ones now?

ralphm: It is a XEP, it's in XEP-0045.

just asking. If there already is deployment, well, yeah

This is splitting it out.

Well, that's my point about it being semi-historical

we're just moving things around

http://xmpp.org/extensions/xep-0045.html#schemas-unique

Five minutes to go and lots still to do. I don't want the last meeting of term running over!

erm, I missed that, sorry

ralphm: Are you ok on publishing?

+1 then

Excellent.

pubit!

8) http://xmpp.org/extensions/inbox/dmuc3.html

we need more of those!

distribute them all

I haven't submitted mine yet!

DISTRIBUTE ALL THE THINGS!

This is an odd one. It seems to be taking the approach from FMUC, including copy/pasting blocks of the text from FMUC, but trying to publish it under a new author, with slightly different syntax.

mattj: we'll reserve dmuc4 for you!

Thank you

I heard my next door neighbour wants to submit one too

Kev: yeah, I think the author might want to simply post to the list -- perhaps it didn't even belong in the inbox

Can he have dmuc5?

focus please

linuxwolf, I'm focusing hard

Kev: can't we suggest he works with the fmuc people?

ralphm: We can, that's me.

(-:

Kev: I know, this is you other hat, you know?

But I don't mind this going onto the vine, we've got 3 other variants up there already :)

well, if it is mostly a copy I'm -1

ralphm: I think you should decide for yourself rather than using my biased summary :)

Same, but I haven't read it, so if it's any more complicated I'm voting on list

I'll vote on list

Ok, let's leave this clean and just leave it for the next Council, then.

Kev: that was my plan

Rather than having the confusion of voting crossing the term end.

Good point

I think that's everything, so:

9) Thanks folks.

And thank you, Kev for chairing

Thanks all for the hard work.

Thank you :)

hear hear!

20s left

10) Any other business

beers, fireworks

(Lasting less than 10 seconds)

None

nay

Oh, one more thing

:-D

Why do these meetings take so long?

11) Fini.

再见

Thanks folks, see you on the lists, good luck to the next Council.

+1

thanks indeed

I will sort out minutes, but not tonight.

谢谢大家

Oh yes, thanks for humouring me with the late meeting :)

linuxwolf: care to translate?

thank you to everyone?

practicing my copy/paste Chinese (-:

MattJ: It's ok! I will now go sleep

linuxwolf, with a question mark because you're not quite sure? :)

'night ralphm :)

MattJ: precisely

stpeter, some people get all the fun

IRItating?

enjoy :)

I kinda wish it was called JOES

(-:<

linuxwolf: just buy some stickers

use funky fonts

...

profit!

heh

maybe I can get ekr to raise a "point of order" on the WG name (-:

I'm sure you could

should be fun

/nod