Wednesday, August 03, 2011
council@muc.xmpp.org
August
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

[00:55:07] *** Neustradamus has left the room
[01:57:19] *** Kooda shows as "xa" and his status message is "détaché"
[06:16:38] *** Kev has joined the room
[06:16:39] *** Kev shows as "online"
[10:02:33] *** Kooda shows as "online"
[10:39:29] *** Kooda shows as "xa" and his status message is "mange"
[11:01:26] *** Kooda shows as "online"
[13:31:10] *** linuxwolf has joined the room
[13:49:52] *** linuxwolf shows as "away" and his status message is "stuffage"
[14:10:50] *** linuxwolf shows as "online"
[14:21:21] *** dwd has joined the room
[14:43:26] *** linuxwolf shows as "away" and his status message is "stuffage"
[14:46:34] *** Kev shows as "online"
[14:48:40] *** MattJ has joined the room
[14:53:32] *** stpeter has joined the room
[14:57:47] *** linuxwolf shows as "online"
[15:00:24] <stpeter> greetings
[15:00:51] <MattJ> Greetings stpeter
[15:00:54] *** linuxwolf shows as "dnd" and his status message is "XSF Council Meeting"
[15:01:05] *linuxwolf waves
[15:01:14] <stpeter> MattJ: are still in NYC?
[15:01:23] *** linuxwolf has left the room
[15:01:24] *** Tobias has joined the room
[15:01:24] *** Tobias shows as "online" and his status message is "Available"
[15:01:25] <MattJ> How was Quebec?
[15:01:29] <MattJ> I am
[15:01:31] *** linuxwolf has joined the room
[15:01:33] <stpeter> it was enjoyable
[15:01:35] <Kev> Evening all.
[15:01:36] <stpeter> nice place
[15:01:43] <stpeter> enjoyable and busy, of course
[15:01:59] <linuxwolf> very French (-:
[15:02:04] <MattJ> Heh
[15:02:17] *** Fritzy has joined the room
[15:02:17] *** Fritzy shows as "online"
[15:03:58] <Kev> Right, it is time.
[15:04:08] <Kev> 1) Roll call.
[15:04:10] <Kev> I'm here!
[15:04:14] <Fritzy> howdy
[15:04:15] <linuxwolf> presente
[15:04:21] <MattJ> Presents
[15:04:34] <linuxwolf> dang…a full house?! unpossible!
[15:04:44] <linuxwolf> oh, no ralphm
[15:04:54] <MattJ> Don't get your hopes up :)
[15:05:21] <Kev> 2) Promote XEP-0260 to Draft?
http://mail.jabber.org/pipermail/standards/2011-June/024722.html
[15:06:13] <linuxwolf> +1; I think I found a couple cosmetic grammatical error, but not enough to hold it up
[15:06:24] <Fritzy> no traffic on the mailing list?
[15:06:30] <MattJ> Am I correct in that we had no LC feedback?
[15:06:31] <linuxwolf> not until this morning
[15:06:33] <Kev> Tobias had feedback on both.
[15:06:37] <Kev> (A few minutes ago)
[15:06:42] <MattJ> Oh, Tobias
[15:07:07] <Kev> It bothers me somewhat that we're promoting stuff to draft that only a single person says they've implemented or intend to implement.
[15:07:27] <MattJ> So, I've implemented most of 260 - but it was some time ago, I'd like to know interop with someone worked
[15:07:36] <MattJ> I'd like to test with Tobias in particular
[15:07:53] <MattJ> and Gajim implements it I think, if we can track down the code (I'm not sure it's in trunk)
[15:07:54] <Fritzy> ok, well, sounds like 260 has at least two implementers
[15:08:01] <Fritzy> 3
[15:08:13] <linuxwolf> yeah, that was my sense
[15:08:14] <dwd> IIRC, Gajim's Jingle-FT is in a branch, yes.
[15:08:14] <MattJ> Implementers, no real world experience
[15:08:23] <Fritzy> right
[15:08:26] <linuxwolf> true
[15:08:28] <Kev> Can we put off 260 and 261 until next week, and chase people in the meantime for feedback?
[15:08:37] <linuxwolf> wfm
[15:08:40] <Fritzy> Should we hold off on draft until we see some interop, at least anecdotally?
[15:08:41] <Tobias> there's also a pidgin branch implementing 260 and 261..at least claiming that
[15:08:42] <MattJ> dwd, Asterix told me he was merging it "tomorrow" about 6 months ago
[15:08:57] <Fritzy> cool, let's give it a week
[15:09:08] <stpeter> MattJ: :)
[15:09:12] <linuxwolf> the same would be true for 261, yes?
[15:09:15] <Tobias> well..IBB to pidgin works for me
[15:09:21] <MattJ> I'd like to bring up the last call issues in AOB
[15:09:21] <Kev> It doesn't seem that waiting a week will significantly hurt anything, and it may give us a much better-informed position to comment from.
[15:09:36] <Kev> stpeter: Your thoughts?
[15:09:41] <dwd> I note that proof of interop is not required by XEP-0001 for Draft.
[15:09:43] <stpeter> fine with me
[15:10:03] <Kev> dwd: Not required, but if we have known implementors, it makes sense to me to try to get feedback.
[15:10:10] <Fritzy> dwd: *nod* -- still, seems like we're blind without at least some evidence that it works.
[15:10:11] <stpeter> personally I've been having dark thoughts about how everyone will just use http://tools.ietf.org/wg/rtcweb/ before long anyway, but what do I know? ;-)
[15:10:12] <Kev> For the sake of a week.
[15:10:25] <linuxwolf> stpeter: heh
[15:10:44] <Kev> So, assuming that covers 3) Advance 261 to Draft as well...
[15:10:51] <dwd> True. Just noting that you don't *have* to. But specifically pinging implementors seems like a plan.
[15:10:56] <Kev> 4) Advance 266 to Draft
http://mail.jabber.org/pipermail/standards/2011-June/024625.html
[15:11:10] *** ralphm has joined the room
[15:11:11] <MattJ> +1 from me, I think
[15:11:17] <ralphm> hello
[15:11:18] <MattJ> Hey ralphm :)
[15:11:18] <linuxwolf> +1 from me
[15:11:20] <Kev> Hi Ralph.
[15:11:23] <Tobias> stpeter, what's the epub symbol there on the WebRTC WG page? IETF publishing in ePub now? :)
[15:11:27] <linuxwolf> damn…we do have a full house
[15:11:28] <linuxwolf> (-:
[15:11:35] <stpeter> re codecs, see also http://tools.ietf.org/html/draft-cbran-rtcweb-codec-00
[15:12:02] <linuxwolf> Tobias: some things, yes…sort of (-:
[15:12:21] <Kev> So, I know there was some amount of discussion on list about 266/Opus, but as far as I can tell the XEP does address that (we shouldn't be recommending it/we're not, we should be noting the claim/we are).
[15:12:36] <linuxwolf> exactly
[15:12:40] <stpeter> rightio
[15:12:43] <MattJ> Same logic from me
[15:12:51] <Fritzy> +1
[15:13:07] <linuxwolf> there's more consensus on jingle@ mailing list
[15:13:13] <ralphm> +!
[15:13:17] <ralphm> ehm, +1
[15:13:23] <MattJ> Heh
[15:13:24] <linuxwolf> (-:
[15:13:44] <Kev> There was someone offering to list all the reasons it was bad, then not doing so when asked to, so I'm +1 on this in the absence of any clearly unaddressed issues.
[15:13:47] <stpeter> I like +!
[15:13:48] <ralphm> and agreed about #2 and #3
[15:14:03] <stpeter> or +¡
[15:14:15] <Fritzy> haha
[15:14:27] <Kev> So, noting there's two AOB items (Matt and Matt).
[15:14:29] <ralphm> there doesn't seem to be a unicode code point for [approved]
[15:14:32] <linuxwolf> fußy
[15:14:35] <Kev> 5) Time of next meeting?
[15:14:38] <Kev> Next week, same time?
[15:14:43] <stpeter> WFM
[15:14:47] <ralphm> yup
[15:14:49] <linuxwolf> no news from Quebéc?
[15:14:59] <linuxwolf> and that date/time works for me
[15:15:03] <Kev> linuxwolf: That's your AOB item :)
[15:15:06] <Kev> Well, yours/PSA's.
[15:15:07] <stpeter> actually Québec :)
[15:15:08] <linuxwolf> haha
[15:15:12] <Kev> 6) AOB.
[15:15:15] <ralphm> :-)
[15:15:19] <Kev> Québec report from Peter/Matt:
[15:15:30] <linuxwolf> grazíe
[15:15:32] <ralphm> (expecting a large paste here)
[15:15:42] *linuxwolf digs up email sent earlier...
[15:16:07] <stpeter> the main topics were s2s/DNA, i18n, and e2e encryption
[15:16:07] <linuxwolf> so, on the XMPP front...
[15:16:15] <ralphm> (wow, why are linuxwolf's /me messages styled in a huge font?)
[15:16:29] <MattJ> They seem to be XHTML-IM
[15:16:33] <linuxwolf> ugh
[15:16:41] <linuxwolf> I need to get a better client
[15:16:49] <linuxwolf> or write one /ducks
[15:16:53] <linuxwolf> anyway
[15:17:01] <ralphm> Before the last upgrade I patched gajim to disregard certain styles
[15:17:06] <MattJ> linuxwolf, join the club
[15:17:13] <Fritzy> ssshh, we're hearing a report
[15:17:14] <ralphm> in particular font-color, family and size
[15:17:17] <linuxwolf> hehe
[15:17:17] <linuxwolf> so
[15:17:19] <linuxwolf> DNA
[15:17:32] <Fritzy> "the main topics were s2s/DNA, i18n, and e2e encryption"
[15:18:09] <linuxwolf> the biggest concern seemed to be around requiring tedious software upgrades (DNSSEC-based)...
[15:18:33] <linuxwolf> …or sharing cert+keys from the hostee to the hoster (cert-based)
[15:18:54] <linuxwolf> post-session, there was some agreement to a compromise...
[15:19:37] <linuxwolf> …to use a descriptor file served from the hostee's HTTPS server that listed the valid name (or certificate) for the _xmpp-server._tcp service
[15:20:12] <linuxwolf> Joe Hildebrand and Eric Rescola were working on a draft, so I think something should be emerging in the next week or two
[15:20:50] <Kev> Which is really just the age-old trick of shifting-the-problem-without-solving-much.
[15:20:55] <linuxwolf> right
[15:21:07] <MattJ> brb, switching to wifi (198 would be so useful right now)
[15:21:12] <stpeter> :)
[15:21:13] <ralphm> based on .well-known?
[15:21:17] <linuxwolf> and with the CA trust anchors in x.509 being less-than-stellar...
[15:21:22] <linuxwolf> ralphm: yes
[15:21:28] <dwd> I basically agree, but I think it's an achievable goal and workable if one assumes that it's a fallback from DNSSEC.
[15:21:41] <linuxwolf> /nod
[15:21:42] <ralphm> yeah I can imagine the approach
[15:22:17] <ralphm> dwd: agreed
[15:22:33] <ralphm> 'assumes' or 'requires'
[15:22:38] <linuxwolf> on the i18n front…it's basically see what happens in PRECÍS, although stpeter can comment more there
[15:23:00] *** MattJ has left the room
[15:23:08] *** MattJ has joined the room
[15:23:09] *** MattJ shows as "dnd"
[15:23:09] *** MattJ shows as "dnd"
[15:23:09] *** MattJ shows as "dnd"
[15:23:09] *** MattJ shows as "dnd"
[15:23:09] *** MattJ shows as "dnd"
[15:23:09] *** MattJ shows as "dnd"
[15:23:09] *** MattJ shows as "dnd"
[15:23:17] <stpeter> yes, I've been pushing for progress on internationalization
[15:23:41] <linuxwolf> we'll need to do something
[15:23:44] <stpeter> I need to update the precis-framework document based on feedback received in Québec City
[15:23:49] <linuxwolf> /nod
[15:24:17] <stpeter> I worry a bit about Kev's point that changing the address format might require a version change to XMPP itself, but we'll see if that's needed...
[15:24:18] <ralphm> linuxwolf: as in 'nobody is picking up the work'?
[15:24:36] <linuxwolf> ralphm: it's more that "this is hard, and we don't know what to do"
[15:24:41] <stpeter> ralphm: unfortunately, few people care about internationalization, and even fewer know how to help
[15:25:01] <ralphm> or can take time to work on it
[15:25:02] <stpeter> however, I think we can see a realistic path forward
[15:25:07] <linuxwolf> /nod
[15:25:22] <Kev> stpeter: I realise I'm speaking as someone who does't understand this stuff, but I only see two options - either updating away from 'prep gives us something new and therefore isn't compatible, or it's not giving us anything that stringprep/fixed unicode versioning doesn't.
[15:25:26] <stpeter> I mean, what we're working on is based on / similar to what people are already using for domain names
[15:25:29] <linuxwolf> I think it's more that it's hard and daunting, so the motivation to work on it is less
[15:25:39] <dwd> FWIW, Isode folk have picked up some knowledge; I'll see if I can get some time from Alexey and Kurt on this.
[15:26:07] <linuxwolf> I've been getting a crash course from stpeter and hildjj
[15:26:14] <linuxwolf> it is daunting
[15:26:15] <ralphm> Kev: I see. And your point is that since the JIDs already appear in the stream header, you can't discover it?
[15:26:24] <stpeter> Kev: right -- we need to dig into those issues, which I plan to do Real Soon Now™
[15:26:27] <stpeter> ralphm: right
[15:26:44] <linuxwolf> Kev: it's not clear if it will really break in the practical sense, or just the theoretical sense
[15:26:49] <linuxwolf> or at least not clear to me
[15:26:57] <stpeter> yeah
[15:27:05] <stpeter> well, further updates on the way :)
[15:27:07] <stpeter> next item?
[15:27:10] <linuxwolf> the fixed unicode is, or will be very soon, breaking in the practical sense
[15:27:11] <MattJ> I don't think it needs a version bump
[15:27:13] <dwd> Kev, I think we may need to signal in the <stream:stream>, but it's not clear this is a full version-break.
[15:27:25] <linuxwolf> MattJ: I don't think it's clear yet
[15:27:27] <ralphm> linuxwolf: so we need to assert the risk of breakage and decide if we think that is generally acceptable
[15:27:44] <Fritzy> So, how was the weather?
[15:27:44] <linuxwolf> ralphm: basically…there's more to it, but that's one of the big things
[15:27:44] <MattJ> if the JID is valid, who cares about the version? if it's invalid (and should be valid) then there's a broken interop whether we bump the version or no
[15:27:56] <Kev> So, was that the total for the WG meet?
[15:27:56] <linuxwolf> Fritzy: warm, but not unpleasant (-:
[15:28:23] <linuxwolf> there was an E2E presentation of sorts (-:
[15:28:31] <ralphm> Well, if we ever change the version number, I have a wish list
[15:28:34] <ralphm> :-D
[15:28:37] <linuxwolf> that is basically waiting on output from WOES cum JOSE
[15:28:38] <dwd> linuxwolf, That's just a tradition, now, right?
[15:28:45] <Kev> MattJ: Unless it's valid, or different, under different profiles and you need to signal the profile. If you have a profile in common you can interop, otherwise not. This can be a version number, or just another top level attribute on the stream header.
[15:28:48] <linuxwolf> dwd: essentially (-:
[15:28:48] <MattJ> ralphm, you and how many other people? :)
[15:29:12] <ralphm> MattJ: how is that relevant :-P
[15:29:53] <MattJ> Kev, true, valid but "different" would be awkward
[15:29:58] <stpeter> forget about XMPP 1.0+, just use WebSocket
[15:30:03] <Kev> linuxwolf: Is there more to say on e2e?
[15:30:11] <Kev> Or shall we move onto MattJ's AOB for the last 2 minutes?
[15:30:11] <MattJ> stpeter, because those guys really have their act together
[15:30:13] <ralphm> stpeter: hehe
[15:30:33] <dwd> stpeter, Can't on S2S, neither end would know which one should be encrypting their output.
[15:30:37] <Kev> MattJ: You wanted to AOB about the LC feedback?
[15:30:45] <MattJ> Right
[15:30:53] <linuxwolf> Kev: there's a WG forming for "better" crypto options, which is now called JOSE…otherwise nothing concrete yet
[15:31:01] <MattJ> Kev, it was just a thought that I had before the meeting
[15:31:22] <MattJ> that it's not uncommon for us to have no LC feedback, despite implementations existing
[15:31:31] <MattJ> and I'm wondering if that's something we can improve on
[15:31:48] <ralphm> MattJ: which could either mean our specs are awesome, or nobody cares to write up issues?
[15:32:03] <MattJ> ralphm, nobody cares, most certainly :)
[15:32:07] <linuxwolf> heh
[15:32:09] <Tobias> the latter
[15:32:22] <linuxwolf> yeah
[15:32:27] <MattJ> I'm just wondering if we should make it someone's responsibility to research implementations and reach out to these people
[15:32:31] <Kev> ralphm: If they were awesome, but people cared, they'd still reply to the "Is this spec alright?" mail.
[15:32:35] <MattJ> I know that's what we're doing already, typically
[15:32:43] <MattJ> but only after the first LC period fails
[15:33:05] <Kev> Right.
[15:33:10] <ralphm> So would it help to approach the (known) implementers more directly?
[15:33:12] <Kev> A question to send out to list, probably.
[15:33:17] <linuxwolf> /hnod
[15:33:19] <MattJ> ralphm, that's the idea
[15:33:19] <linuxwolf> /nod even
[15:33:21] <Kev> As we've just hit half past so we're out of time.
[15:33:28] <stpeter> it seems to me that dedicated protocols like SIP and XMPP and IMAP are going away for client-to-server interactions, and we'll just be left with WebSocket and JavaScript APIs -- however, we'll still need s2s protocols for federation purposes (if anyone ever wants to federate in the brave new world of web silos
[15:33:48] <MattJ> stpeter, you can be cynical sometimes :)
[15:33:53] <linuxwolf> stpeter: I don't think that's true at all
[15:33:57] <stpeter> MattJ: just reading the writing on the wall
[15:34:03] <stpeter> or maybe I just need a vacation
[15:34:08] <Fritzy> Well, stpeter is right that there is that trend
[15:34:11] <MattJ> Looking at a different wall to me
[15:34:16] <MattJ> There's hype
[15:34:19] <Fritzy> but I don't think it'll be ultimately to that side
[15:34:25] <MattJ> I don't know if that constitutes trend just yet
[15:34:25] <linuxwolf> stpeter: except those WS + JS end up using something that looks an awful lot like XMPP or SIP or IMAP d-:
[15:34:37] <stpeter> linuxwolf: perhaps, yeah
[15:34:38] <Kev> In any case, I think we're done.
[15:34:42] <Kev> Any other any other?
[15:34:45] <MattJ> None from me
[15:34:46] <stpeter> meeting is done, I think
[15:34:50] <linuxwolf> nay
[15:34:52] <ralphm> just like Ruby is taking over the world
[15:34:56] <Kev> Righty.
[15:34:57] <Kev> Thanks all
[15:34:57] <Fritzy> :)
[15:34:59] <Tobias> lol
[15:35:00] <Fritzy> ciao!
[15:35:01] <linuxwolf> haha
[15:35:01] <stpeter> ralphm: Ruby forever!
[15:35:03] *Kev gangs the bavel.
[15:35:03] <MattJ> I'll email standards@ about LC
[15:35:10] <stpeter> MattJ: thanks
[15:35:10] <linuxwolf> grazíe
[15:35:28] <ralphm> MattJ: I'll bet you get loads of responses
[15:35:29] <stpeter> clearly Lua is taking over the world ;-)
[15:35:37] <MattJ> ralphm, :)
[15:35:39] <linuxwolf> ¡chao!
[15:35:56] *stpeter glances at the Lua book that linuxwolf just loaned him
[15:35:58] <MattJ> stpeter, http://www.tiobe.com/index.php/content/paperinfo/tpci/index.html :)
[15:36:08] <ralphm> I do think lua is kinda cool
[15:36:12] <linuxwolf> stpeter: it's really hildjj's (-:
[15:36:26] <Tobias> MattJ, right there trending with Transact-SQL
[15:36:31] <linuxwolf> although JavaScript is the new Ruby…with Lua maybe the new Python d-:
[15:36:36] <ralphm> I /was/ kind of surprised finding out that prosody doesn't respond well to a 4G lua file with offline storage
[15:36:36] <linuxwolf> ok, off to my next meeting
[15:36:41] *** linuxwolf shows as "dnd" and his status message is "in a meeting!"
[15:37:03] <MattJ> ralphm, lovely
[15:37:16] <MattJ> ralphm, FWIW we're working on that :)
[15:37:33] <ralphm> MattJ: well, it was also something I needed to fix on my end
[15:38:15] <ralphm> they were piling up pubsub notifications for resources that were no longer listening
[15:38:17] <MattJ> Sure, but if you want 4GB of offline messages...
[15:39:18] *** Tobias has left the room
[15:39:22] *** Fritzy has left the room
[15:40:07] *** Tobias has joined the room
[15:40:07] *** Tobias shows as "online" and his status message is "Available"
[15:57:18] *** linuxwolf has left the room
[15:57:32] *** linuxwolf has joined the room
[16:13:24] *** MattJ has left the room
[16:13:26] *** MattJ has joined the room
[16:13:29] *** MattJ shows as "dnd"
[16:13:29] *** MattJ shows as "dnd"
[16:13:29] *** MattJ shows as "dnd"
[16:13:35] *** dwd shows as "away" and his status message is " (Away as a result of being idle more than 5 min)"
[16:21:01] *** dwd shows as "online"
[16:21:23] *** bear shows as "online"
[16:24:06] *** ralphm shows as "away" and his status message is "Away as a result of being idle"
[16:24:18] *** Neustradamus has joined the room
[16:24:52] *** linuxwolf shows as "away" and his status message is "stuffage"
[16:25:21] *** ralphm shows as "online"
[16:34:32] *** linuxwolf shows as "online"
[16:43:49] <Tobias> how long is a Last Call period?
[16:43:59] <linuxwolf> a fortnight, IIRC
[16:44:49] *** dwd shows as "away" and his status message is " (Away as a result of being idle more than 5 min)"
[16:45:15] <linuxwolf> "The Last Call shall expire not less than fourteen (14) days after the date of issue."
[16:45:25] <linuxwolf> so, a fortnight is a minimum (-:
[16:46:32] <stpeter> sometimes we declare longer review periods
[16:46:40] <stpeter> e.g., for large or controversial specs
[16:46:40] <ralphm> 15 is right out, so is 13, unless followed by 14
[16:46:47] <stpeter> or if there are some holidays in the middble
[16:46:47] <linuxwolf> or extend them
[16:47:16] <stpeter> that too
[16:47:27] <ralphm> :-D
[16:54:49] *** dwd shows as "xa" and his status message is " (Not available as a result of being idle more than 15 min)"
[17:13:27] *** stpeter shows as "dnd" and his status message is "on a conference call"
[17:20:19] *** ralphm shows as "away" and his status message is "Away as a result of being idle"
[17:21:40] *** ralphm has left the room
[17:32:59] *** dwd shows as "online"
[17:41:37] *** dwd shows as "away" and his status message is " (Away as a result of being idle more than 5 min)"
[17:44:16] *** linuxwolf shows as "away" and his status message is "stuffage"
[17:45:05] *** linuxwolf shows as "online"
[17:51:37] *** dwd shows as "xa" and his status message is " (Not available as a result of being idle more than 15 min)"
[18:01:36] *** linuxwolf shows as "away" and his status message is "stuffage"
[18:32:23] *** Neustradamus has left the room
[18:53:34] *** MattJ has left the room
[19:12:37] *** Neustradamus has joined the room
[19:13:03] *** dwd shows as "online"
[19:19:48] *** linuxwolf shows as "online"
[19:29:18] *** Tobias shows as "online" and his status message is "Jingling along"
[19:45:36] *** stpeter has left the room
[19:46:15] *** stpeter has joined the room
[19:47:58] <dwd> Tobias, Extending Last Calls is specifically allowed:
If necessary, the XMPP Extensions Editor may, at his discretion and in consultation with the XMPP Council, extend the Last Call or issue a new Last Call if the XEP requires further discussion.
[19:51:30] <stpeter> and we've done that in the past
[19:51:43] <dwd> Right, I recall a few times, actually.
[19:53:54] <Tobias> interesting
[19:55:26] <Tobias> http://jabber.markmail.org/ <-- watching that activity graph we soon are at about 0 :)
[20:05:50] <stpeter> Tobias: interesting
[20:06:17] <stpeter> not that different from, say, http://spamassassin.markmail.org/
[20:06:56] <Tobias> stpeter, but spam is decreasing year over year..so that's no surprise ;)
[20:07:50] <stpeter> heh maybe
[20:08:58] <Tobias> but it's not just that graph..i've been noticing quite a decline in xmpp related mailing list activity...don't know if that's due to work on new standard or whatever
[20:09:13] <stpeter> Tobias: agreed
[20:09:22] <stpeter> I think XMPP has become a stable, boring technology
[20:09:42] <stpeter> or most of the energy is moving to things like node.js and websockets
[20:09:45] <stpeter> not quite sure yet
[20:09:48] <dwd> Tobias, It's a sign that we're not working on anything interesting right now.
[20:09:48] <linuxwolf> it's not the new hotness (-:
[20:10:05] <stpeter> heh
[20:10:14] <stpeter> it's the old hotness?
[20:10:21] <Tobias> dwd, file transfer is interesting...since it's the thing that never works :P
[20:10:35] <dwd> Tobias, But I wouldn't read too much into that - IMAP went dead for several years then popped back up for all the mobile mail stuff.
[20:11:49] <linuxwolf> it could heat up again when people realize that WebSockets turned out to be warmed over cometd/BOSH (-: /ducks
[20:11:59] *stpeter laughs
[20:12:37] <dwd> linuxwolf, WS is, in principle, faster, except that the repeated buffering and XORs are, I think, going to wear away some of that benefit.
[20:13:17] <dwd> linuxwolf, The interesting question is whether it'll fly on mobile. Very little mobile experience in that group, and there's the usual crop of "Well, I heard that on mobile [...]".
[20:13:57] <linuxwolf> dwd: yeah, we'll see
[20:14:03] <Tobias> isn't WebSocket still going over TCP?
[20:14:04] <linuxwolf> I've still got my doubts
[20:14:16] <linuxwolf> Tobias: it's worse than that
[20:14:45] <dwd> Tobias, Yes, but TCP is fine on mobile.
[20:14:57] <dwd> Tobias, You've got to get much worse networks for TCP to be a problem.
[20:16:21] <Tobias> dwd, Apple seems to think different, watched a couple of sessions from their last WWDC, they discourage any long living connection. I guess other mobile vendors think different. Opening TCP connections over and over again can't be that good for speed/latency/battery/etc.
[20:16:29] <Tobias> or not?
[20:16:34] <linuxwolf> I guess watching that train wreak…er, mailing list…made me a bit jaded
[20:17:29] <dwd> Tobias, On the vast majority of mobile networks, you can leave a TCP session open and happy for 15 minutes or more, these days.
[20:17:41] <dwd> Tobias, That is, without sending any traffic at all.
[20:17:47] <linuxwolf> /nod
[20:18:04] <Tobias> dwd, while traveling through half the country?
[20:19:03] <dwd> I've certainly held TCP sessions whilst on a train. I even held them through the Severn Tunnel a couple of times during Council meetings. :-)
[20:19:10] <linuxwolf> Tobias: from my limited experience, as long as I don't drop below a certain signal threshold, I've been fine
[20:19:31] <linuxwolf> but I'm in the US (-:
[20:32:24] <stpeter> OT: does anyone care about jabberpowered.org anymore? I need to decide whether to renew the domain registration :)
[20:32:45] <linuxwolf> we still had that?
[20:32:51] <stpeter> yep
[20:33:06] <stpeter> every year I try to decide whether I still need to renew it :)
[20:33:23] <linuxwolf> heh
[20:35:56] *stpeter decides to register a different domain name instead
[20:36:24] <linuxwolf> thefutureiswebsocketsnotxmpp.com?
[20:36:43] <Tobias> heh
[20:36:45] <stpeter> :)
[20:39:06] *** linuxwolf shows as "away" and his status message is "stuffage"
[20:52:15] *** linuxwolf shows as "online"
[20:54:07] *** Marcus has joined the room
[21:30:03] *** dwd shows as "away" and his status message is " (Away as a result of being idle more than 5 min)"
[21:40:03] *** dwd shows as "xa" and his status message is " (Not available as a result of being idle more than 15 min)"
[21:44:00] *** stpeter has left the room
[21:52:06] *** linuxwolf has left the room
[21:54:55] *** Neustradamus has left the room
[21:55:06] *** Neustradamus has joined the room
[21:56:36] *** linuxwolf has joined the room
[21:58:17] *** linuxwolf shows as "dnd" and his status message is "in a meeting!"
[22:00:17] *** dwd shows as "online"
[22:35:55] *** Marcus has left the room
[22:47:41] *** dwd shows as "away" and his status message is " (Away as a result of being idle more than 5 min)"
[22:57:41] *** dwd shows as "xa" and his status message is " (Not available as a result of being idle more than 15 min)"
[23:03:45] *** linuxwolf has left the room
[23:29:03] *** Tobias has left the room