Wednesday, May 18, 2011
council@muc.xmpp.org
May
Mon Tue Wed Thu Fri Sat Sun
            1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
         
XMPP Council Room | https://xmpp.org/about/xmpp-standards-foundation#council | Room logs: http://logs.xmpp.org/council/ | https://trello.com/b/ww7zWMlI/xmpp-council-agenda

[03:20:24] *** stpeter has joined the room
[04:08:30] *** stpeter has left the room
[06:43:17] *** Kev shows as "online"
[09:41:52] *** Tobias has joined the room
[10:29:05] *** Tobias has left the room
[11:06:01] *** Tobias has joined the room
[11:06:09] *** Tobias shows as "online"
[11:25:15] *** Kooda shows as "online"
[12:41:11] *** Kev has left the room
[12:49:25] *** Kev has joined the room
[12:49:27] *** Kev shows as "online"
[14:08:08] *** Tobias has left the room
[14:13:47] *** Tobias has joined the room
[14:43:22] *** stpeter has joined the room
[14:47:21] *** Fritzy has joined the room
[14:47:57] <Fritzy> Kev: I have given my feedback and reasons for vetoing Sensors over XMPP
[14:49:15] <Kev> Evening.
[14:49:16] <Kev> Thanks.
[14:51:55] <stpeter> thanks, Fritzy!
[14:52:12] <Fritzy> *nod* -- sorry it's so late.
[14:56:11] *** linuxwolf has joined the room
[14:57:43] <linuxwolf> /sigh … one of my network drops is dead
[14:57:55] <linuxwolf> that's going to make today interesting
[14:58:00] <stpeter> heh
[14:58:17] <linuxwolf> if only I worked for a company that knew something about networking...
[14:58:44] <stpeter> :)
[14:58:59] <linuxwolf> on the plus side, I have no phone, since it's the one plugged into the dead drop (-:
[15:00:40] <Fritzy> You find good in bad. I like!
[15:01:05] <stpeter> every silver lining has a cloud!
[15:01:21] *** ralphm has joined the room
[15:01:36] <linuxwolf> and a nice movie reference, fritzy!
[15:01:43] <ralphm> hi
[15:01:59] *linuxwolf adds Fifth Element to the queue
[15:02:05] <Kev> Right.
[15:02:11] <Kev> Dingding etc.
[15:02:19] *stpeter wonders if http://xmpp.org/about-xmpp/xsf/xmpp-council/tenth-council/ is up to date
[15:02:21] <linuxwolf> !agendaup
[15:02:40] <Kev> 1) Roll Call
[15:02:46] <Kev> MattJ has been poked.
[15:02:58] <Fritzy> here
[15:03:04] <linuxwolf> presente
[15:03:20] *** linuxwolf shows as "dnd" and his status message is "in a meeting!"
[15:03:22] *** linuxwolf shows as "dnd" and his status message is "XSF council meeting"
[15:03:32] *** Dave Cridland has joined the room
[15:03:59] <Kev> Ralph here?
[15:04:03] <stpeter> yes
[15:04:13] <Kev> Oh, sorry, didn't recognise the new nick.
[15:04:16] <Kev> stpeter: Hi Ralph.
[15:04:24] <stpeter> :)
[15:04:25] <Kev> 2) Agenda bashing.
[15:04:36] <Fritzy> None
[15:04:41] <linuxwolf> nay
[15:04:55] <ralphm> new nick?
[15:05:10] <Kev> ralphm: stpeter replied on your behalf.
[15:05:44] <Kev> 3) XEP-0220 (Server Dialback), advance to Draft?
[15:06:15] <Fritzy> +1
[15:06:25] <ralphm>
[15:06:41] <stpeter> ooh nice
[15:06:44] <linuxwolf> there's still some discussion on how to advertise db+errors, correct?
[15:06:52] <stpeter> linuxwolf: yes
[15:07:03] <Kev> I'd rather hold off on Drafting this until there's at least some agreement amongst the authors :)
[15:07:11] <Fritzy> ah, ok
[15:07:16] <linuxwolf> Kev: I don't think that will happen
[15:07:19] *** MattJ has joined the room
[15:07:21] <Kev> The namespace versioning issue needs resolution, too.
[15:07:34] *MattJ enters quietly
[15:07:39] <stpeter> Kev: I think that's the same issue, no?
[15:08:05] <Kev> stpeter: Is it? I thought he was complaining about a bug in Example 7.
[15:08:12] <Kev> Ah.
[15:08:19] <Kev> Which is, indeed, error handling.
[15:08:37] <Fritzy> sounds like I should retract my vote.
[15:08:49] <Fritzy> If things are still up in the air
[15:09:04] <linuxwolf> s/to=sender.tld/to=target.tld/
[15:09:05] <MattJ> I think they are
[15:10:02] <Kev> stpeter: Do you have comment on this?
[15:10:25] <Dave Cridland> FWIW, i think the existing errors feature advertising has been about long enough that even if the implementations *could* be changed, I suspect there'll be confusion if it were changed in the spec.
[15:10:59] <MattJ> Prosody doesn't implement it, I'm not sure ejabberd does - M-Link and PSYC? :)
[15:11:05] <ralphm> I'm not sure why this would impede advancement, as this was already in the spec and people apparently have implemented and it is seems backwards compatible
[15:11:14] <linuxwolf> also, technically, XEP-0220 is experimental
[15:11:18] <Dave Cridland> MattJ, I thought someone else mentioned doing so, as well.
[15:11:45] <Dave Cridland> linuxwolf, That's not an argument *for* changing it.
[15:12:00] <Kev> ralphm: No-one has implemented the spec as it stands, because it's just had the namespace changed.
[15:12:06] <stpeter> do we agree that this is a poor design?

<stream:features>
<dialback xmlns='urn:xmpp:features:dialback'>
<errors/>
</dialback>
</stream:features>

[15:12:06] <linuxwolf> heh
[15:12:11] <Kev> That is - all existing implementations are incompatible with the current revision.
[15:12:13] <linuxwolf> I think it is
[15:12:13] <MattJ> stpeter, I think so, yes
[15:12:22] <linuxwolf> (a bad design)
[15:12:33] <Kev> stpeter: Poor design? Yes. What's already in the wild? Probably.
[15:12:34] <ralphm> Kev: oh, right
[15:12:37] <stpeter> because if we keep that, then we need to add a note to the spec saying "this is a bad design, if you design your own stream features in the future then don't do it this way"
[15:12:37] <linuxwolf> dwd: a bad design is a good reason to change it
[15:12:55] <Dave Cridland> linuxwolf, Cool. Can we change self-framing XML, too, then?
[15:13:06] <Fritzy> haha
[15:13:11] *ralphm gets popcorn
[15:13:14] <linuxwolf> is self-framing XML an experimental XEP?
[15:13:16] <stpeter> and we need to say "if we add new dialback-related features in the future, then they MUST NOT be added as new child elements like <errors/>"
[15:13:26] <Kev> stpeter: That is one of the two options, yes.
[15:13:29] <Dave Cridland> linuxwolf, Proposed Standard, so technically still subject to change.
[15:13:52] <stpeter> how many beers do I need to buy for the M-Links developers to change their minds?
[15:14:08] <stpeter> s/Links/Link/
[15:14:09] <Dave Cridland> (I'm not even sure it is bad design - at best it's inconsistent with Disco)
[15:14:26] <Kev> stpeter: I'm not firmly in one camp or the other. I'd just like to see something approaching agreement.
[15:14:47] <Kev> (Well, I'm firmly in the 'we should not go with the current XEP approach' camp, but not firmly in either of the other two)
[15:14:47] <linuxwolf> I think it's a bad design…and would require the spec to be updated (and the namespace to be changed) if we added more later
[15:15:01] <Kev> linuxwolf: No, that's not true.
[15:15:07] <linuxwolf> it's a schema change
[15:15:11] <Dave Cridland> linuxwolf, I don't see how you arrive at that conclusion.
[15:15:15] <stpeter> I note that http://xmpp.org/extensions/xep-0288.html has:

<stream:features>
<starttls xmlns='urn:ietf:params:xml:ns:xmpp-tls'/>
<bidi xmlns='urn:xmpp:features:bidi'/>
</stream:features>
[15:15:29] <stpeter> not:

<stream:features>
<dialback xmlns='urn:xmpp:features:dialback'>
<bidi/>
</dialback>
</stream:features>
[15:15:32] <Kev> linuxwolf: Adding more *the wrong way* later would lead to a schema change.
[15:15:37] <stpeter> although bidi is not strictly a dialback thing
[15:15:39] <linuxwolf> Kev: that's what I'm saying
[15:15:40] <Kev> I'm assuming if we add more features in the future, we'll do it right.
[15:15:45] <linuxwolf> So #1 is out
[15:15:49] <Kev> No.
[15:16:06] <Kev> We've done it the wrong way once, they question is whether to live with it or not.
[15:16:31] <Kev> There's no cause and effect between letting the (previous version) current way stand and having to do it wrong again in the future.
[15:16:35] <linuxwolf> ok, then my vote is to not live with it
[15:16:47] <Dave Cridland> I'd be fine with <errors/> being the last thing added to dialback's stream feature.
[15:16:51] <linuxwolf> but, IMO, we have time to change this
[15:17:00] <MattJ> So is mine, but only if we can track down existing implementations (I'm not sure there are many in reality)
[15:17:22] <linuxwolf> I'm −1 on keeping it, unless there's a plethora of implementation that will break
[15:17:53] <Kev> My preference is that we should only break namespaces when there's an incompatible change being made, really.
[15:17:57] <linuxwolf> then I'll re-evaluate (-:
[15:17:58] <ralphm> we /could/ define a real stream feature and deprecate <errors/> but I'm unsure about the actual impact
[15:18:07] <Dave Cridland> ralphm, Same issue.
[15:18:14] <ralphm> of course
[15:18:37] <MattJ> Dave Cridland, are you saying you're absolutely against switching to a <dialback-errors> feature?
[15:18:38] <stpeter> it's not a hill for me to die on, but it's ugly and the wrong way to do it and sets a bad precedent -- people look at the XEPs for examples of what to do in their own extensions and someone will say "oh they did it in XEP-0220 so it's fine in this extension" (and realize that people might never even look at the spec, they'll just see the XML go across the wire)
[15:18:38] <linuxwolf> we're still advertising support of dialback (sans error-handling) via a namespace declaration, no?
[15:19:04] <Kev> I am, in any case, -1 on advancing 220 as it currently stands, so this discussion can take place on list :)
[15:19:05] <stpeter> linuxwolf: no
[15:19:12] <MattJ> stpeter, I agree it seems bad to have backwards-compatibility cruft in such a "recent" feature
[15:19:24] <stpeter> linuxwolf: advertisment is just via namespace on the stream header
[15:19:29] <Kev> (Which, given that we've not changed 0001 yet means that -0220 has just died. Yay!).
[15:19:38] <ralphm> so discussion ensues and we -1 it for now
[15:19:49] <ralphm> Kev: hehehe
[15:19:52] <Kev> 4) XEP-0262 (Use of ZRTP in Jingle RTP Sessions), advance to Draft?
[15:19:58] <ralphm>
[15:19:58] <Dave Cridland> MattJ, I am generally against changing deployed specs incompatibly for aesthetic reasons, yes.
[15:20:04] <stpeter> I suppose we could say that <dialback xmlns='urn:xmpp:features:dialback'/> means you support dialback with errors and get rid of the child element
[15:20:19] <stpeter> I'd be fine with that
[15:20:21] <linuxwolf> stpeter: I would be ok with that
[15:20:25] <ralphm> stpeter: and if you don't actually do it it doesn't break stuff
[15:20:35] <stpeter> ralphm: right
[15:20:37] <ralphm> and the <errors/> is merely informational no-op
[15:20:41] <Dave Cridland> stpeter, I've a nagging feeling that there exist implementation advertising dialback like that but without supporting errors.
[15:20:42] <MattJ> Is it?
[15:20:46] <ralphm> that'd work for me
[15:20:58] <Kev> Folks, this discussion doesn't need to happen now.
[15:21:03] <ralphm> right
[15:21:06] <Kev> Can we move onto (4) please.
[15:21:08] <stpeter> Dave Cridland: I don't know of any implementations that do it that way
[15:21:20] <linuxwolf> ok, let's focus!
[15:21:22] <linuxwolf> #4
[15:21:22] <stpeter> Kev: trying to use the high bandwidth connection to reach consensus :)
[15:21:27] <stpeter> bt I'll propose it on the list
[15:21:42] <Kev> Discuss in 10mins once Council's over and I can start ignoring the chat :)
[15:21:47] <Kev> 4) XEP-0262 (Use of ZRTP in Jingle RTP Sessions), advance to Draft?
[15:22:10] <Kev> I haven't looked up the feedback thread yet. Will vote on-list.
[15:22:40] <linuxwolf> was there feedback? I don't remember seeing any
[15:22:48] <MattJ> There wasn't feedback iirc
[15:22:54] <Kev> I don't remember any, but I need to find the thread to be sure :)
[15:23:15] <stpeter> heh
[15:23:16] <MattJ> No feedback and I'm not aware of any implementations
[15:23:24] <stpeter> it is implemented in Jitsi
[15:23:28] <MattJ> Ah good
[15:23:31] <Kev> One reply, from Peter, saying that the date was wrong.
[15:23:32] <stpeter> for voice and video
[15:23:45] <linuxwolf> stpeter: can you ping them to send an "a ok" or something about it, then?
[15:23:47] <stpeter> I've seen it displayed at FOSDEM so it must be true
[15:23:58] <Fritzy> well, that's something!
[15:24:05] <Fritzy> Should we ask them for feedback specifically?
[15:24:20] <Kev> If there are known implementations, and no feedback to the LC, I think it's worth poking, yes.
[15:24:27] <MattJ> +1
[15:24:28] <linuxwolf> definitely
[15:24:30] <Dave Cridland> Is Jitsi SIP Communicator as-was?
[15:24:31] <stpeter> Kev: pokage in progress
[15:24:32] <ralphm> +1
[15:24:32] <MattJ> (to poking)
[15:24:35] <MattJ> Dave Cridland, yes
[15:24:35] <Kev> Dave Cridland: Yes.
[15:24:40] <linuxwolf> +1 to POKE, DNV for now
[15:24:43] <Dave Cridland> Anyone else want to say yes?
[15:24:48] <MattJ> Dave Cridland, no
[15:24:48] <Kev> Dave Cridland: No.
[15:25:06] <Kev> Ok, so who's volunteering to poke them for feedback?
[15:25:12] <MattJ> I just have
[15:25:15] <Kev> Ta.
[15:25:17] <linuxwolf> thanks
[15:25:20] <Kev> 5) Date of next.
[15:25:46] <MattJ> Next week is ok for me as far as I am aware
[15:25:50] <Kev> Jolly good.
[15:25:52] <Fritzy> same here
[15:25:53] <linuxwolf> works for me
[15:26:03] <Kev> ralphm?
[15:26:14] <stpeter> any further votes on 178?
[15:26:30] <ralphm> +1
[15:26:31] <Kev> I think the voting period's expired for that hasn't it? :)
[15:26:42] <Fritzy> very much so
[15:26:43] <Kev> 6) AOB.
[15:26:56] <linuxwolf> +1 on −178
[15:27:06] <linuxwolf> thought I did that on-list
[15:27:26] <ralphm> any news on the summit?
[15:27:36] <Kev> ralphm: I think I saw a mail fly by saying it was cancelled.
[15:27:38] <stpeter> linuxwolf: I missed that
[15:27:39] <linuxwolf> but I'm getting ~200 emails/day, so it's getting lost (-:
[15:27:45] <Kev> Although I don't think I was paying much attention at the time.
[15:27:45] <stpeter> linuxwolf: man up :P
[15:28:11] <stpeter> right, we haven't organized the summer summit so we'll see you all in Brussels again someday
[15:28:15] <MattJ> Yay
[15:28:44] <stpeter> so meeting next Wed?
[15:28:47] <Kev> Yep.
[15:28:52] <stpeter> adding to calendar
[15:28:53] <Kev> And we're done with 3mins to spare.
[15:28:58] <Kev> Thanks Peter.
[15:29:01] <Kev> Thanks all.
[15:29:02] <linuxwolf> and now to the flame war
[15:29:04] *Kev bangs teh gavel.
[15:29:09] <MattJ> Heh
[15:29:10] <stpeter> BTW if folks want to look at my i18n stuff it'd be cool
[15:29:16] <MattJ> stpeter, where is that?
[15:29:22] <stpeter> xmpp@ietf.org
[15:29:26] <ralphm> the precis draft
[15:29:33] <stpeter> or http://xmpp.org/2011/05/progress-on-internationalization/
[15:29:40] <MattJ> Ah yes, thanks
[15:29:41] <linuxwolf> stpeter: on my /todo
[15:29:45] <MattJ> On mine too
[15:29:59] *linuxwolf plans on reading it while at Casa Bonita
[15:30:01] <stpeter> thank
[15:30:09] <stpeter> thanks even
[15:30:14] <Fritzy> ciao
[15:30:17] *** Fritzy has left the room
[15:30:17] <stpeter> we can see that Ralph likes Unicode ;-)
[15:30:41] <stpeter> Unicode Rocks™ and all that
[15:30:45] <linuxwolf> Ralphm: what happened to "if you don't like ASCII…f**k you"? (-:
[15:30:50] *** Tobias has left the room
[15:30:51] <stpeter> heh
[15:31:05] <ralphm> linuxwolf: did I ever say that?
[15:31:30] <ralphm> linuxwolf: also, I don't see how that's conflicting with liking unicode
[15:31:52] <stpeter> ralphm is a man of contradictions :)
[15:32:17] <linuxwolf> haha
[15:32:17] *stpeter updates http://xmpp.org/about-xmpp/xsf/xmpp-council/tenth-council/ with linuxwolf's vote
[15:32:29] <ralphm> stpeter: pft! Al
[15:32:32] <linuxwolf> ralphm: well, I said the first part…and beers were involved (-:
[15:32:49] <ralphm> linuxwolf: agreed
[15:32:54] <MattJ> stpeter, hmm, my vote isn't there... I'm +1
[15:33:12] <stpeter> ok
[15:33:13] <MattJ> But I'm working more on Prosody's implementation soon so I may have some feedback
[15:33:18] <ralphm> as ASCII is a strict subset of unicode, I don't see how this is a contradication
[15:33:29] <stpeter> not sure if Fritzy voted
[15:33:30] <MattJ> ralphm, you win :)
[15:33:36] <linuxwolf> ASCII is a strict subset of UTF8, actually
[15:33:38] <linuxwolf> (-:
[15:33:38] <stpeter> I think I missed a meeting while travelling of late
[15:33:53] <ralphm> linuxwolf: hehe
[15:34:04] <linuxwolf> and UTF8 is a particular encoding of Unicode, afterall (-:
[15:34:05] <ralphm> nevertheless a thing to like
[15:34:13] <linuxwolf> I've got a slidedeck I can show you about it (-:
[15:34:45] <linuxwolf> stpeter: I don't think my vote on 0220 is accurate
[15:35:08] <linuxwolf> given the revote and updates and whatnot
[15:35:36] <Kev> Indeed. I'm currently -1.
[15:36:01] *stpeter pings Fritzy about 178
[15:36:58] <stpeter> Kev: noted
[15:38:21] <stpeter> heh, now I'm receiving password requests for chat.facebook.com at jabber.org
[15:38:52] <linuxwolf> stpeter: I'm "" on XEP-0220 right now, pending further discussion about advertisement
[15:39:03] <linuxwolf> ugh
[15:39:27] <linuxwolf> and I'm 10 minutes late to a 15-minute meeting (-:
[15:39:29] <linuxwolf> adios
[15:39:31] *** linuxwolf has left the room
[15:40:06] <stpeter> linuxwolf: noted
[15:40:12] <stpeter> I changed ralphm to "" as well
[15:40:18] <stpeter> so the only vote is Kev's -1 :)
[15:40:31] <Kev> Which I'll no doubt be changing shortly.
[15:40:34] <Dave Cridland> stpeter, ralphm may well be “”, in fact.
[15:40:40] <stpeter> heh
[15:41:23] *** Kooda shows as "away"
[15:46:42] <ralphm>
[15:48:37] <stpeter> :)
[15:50:57] <Dave Cridland>
[15:51:34] <Dave Cridland> I tried a bottom bracket with combining umlaut once, and it actually worked. I was particularly happy.
[15:54:56] <ralphm> ̈̈̈̈‿̈
[15:55:01] <ralphm> ooh
[15:55:28] <Dave Cridland> Unicode Art is so much more experssive than the ASCII kind.
[15:55:31] <ralphm> ‿̈
[15:55:41] <ralphm> I had another combining one in there
[15:55:45] <Dave Cridland> There's actually a smile codepoint that might be better.
[15:57:22] <ralphm> ⁀̤
[15:57:50] *** Kooda shows as "online"
[15:58:17] <ralphm> ̤⁀⃘̤
[16:01:58] *** Dave Cridland shows as "away" and his status message is " (Away as a result of being idle more than 5 min)"
[16:07:04] *** Dave Cridland shows as "online"
[16:07:54] <Dave Cridland> ◡̈
[16:08:02] <Dave Cridland> Many options.
[16:09:01] <ralphm> I just tried a combination that kicked out my connection
[16:09:15] <ralphm> with combining musical symbols
[16:09:39] <ralphm> probably because they are in another plane
[16:16:16] <Dave Cridland> Could be anti-codepoints.
[16:16:33] <Dave Cridland> You can't put those in contact with ordinary codepoints, they'll explode.
[16:19:03] <ralphm> I had a cross-note-head with a combining 1/64 stem
[16:24:04] *** ralphm shows as "away" and his status message is "Away as a result of being idle"
[16:24:09] <Dave Cridland> Yeah, you need to have a special screwdriver for those, though.
[16:24:26] *** Neustradamus shows as "away"
[16:25:27] *** ralphm has left the room
[16:26:43] *** Neustradamus has left the room
[16:26:45] *** Tobias has joined the room
[16:26:52] *** Neustradamus shows as "away"
[16:31:13] *** Neustradamus has left the room
[16:57:34] *** Tobias shows as "away" and his status message is "Auto Status (idle)"
[17:17:34] *** Tobias shows as "xa" and his status message is "Auto Status (idle)"
[17:25:13] *** Dave Cridland shows as "away" and his status message is " (Away as a result of being idle more than 5 min)"
[17:25:18] *** Dave Cridland shows as "online"
[17:32:03] *** Tobias shows as "online"
[18:02:32] *** Dave Cridland shows as "away" and his status message is " (Away as a result of being idle more than 5 min)"
[18:12:32] *** Dave Cridland shows as "xa" and his status message is " (Not available as a result of being idle more than 15 min)"
[18:16:39] *** stpeter has left the room
[18:25:10] *** Kev shows as "away"
[18:36:54] *** MattJ shows as "away" and his status message is "Away as a result of being too idle"
[18:39:54] *** MattJ shows as "online"
[19:17:27] *** MattJ shows as "away" and his status message is "Away as a result of being too idle"
[19:24:15] *** MattJ shows as "online"
[20:20:46] *** Kev shows as "online"
[20:31:51] *** Kooda shows as "online"
[20:35:43] *** Kooda shows as "online" and his status message is "\-"
[20:35:47] *** Kooda shows as "online"
[20:35:55] *** Kooda shows as "online" and his status message is ""-""
[20:35:59] *** Kooda shows as "online"
[20:54:09] *** Kooda shows as "xa" and his status message is "détaché"
[21:16:27] *** Neustradamus shows as "away"
[21:23:05] *** Neustradamus shows as "away"
[21:23:15] *** Neustradamus has left the room
[21:23:35] *** Neustradamus shows as "away"
[22:04:35] *** Kev shows as "away"
[22:10:18] *** Tobias shows as "away" and his status message is "Auto Status (idle)"
[22:10:45] *** Tobias shows as "online"
[22:19:56] *** MattJ shows as "away" and his status message is "Away as a result of being too idle"
[22:23:39] *** Tobias has left the room
[22:25:38] *** MattJ shows as "online"