Logs for jdev

Show join/part/nick changes:

[05:17:05] * jkhii left the chat.
[05:19:14] * teo left the chat.
[05:19:15] * teo joined the chat.
[05:20:41] * johnny joined the chat.
[05:32:21] * john joined the chat.
[05:36:11] * Treebilou joined the chat.
[06:17:56] * john left the chat.
[06:21:38] * jugg left the chat.
[06:22:01] * lastsky left the chat.
[06:22:07] * lastsky joined the chat.
[06:26:02] * scippio left the chat.
[06:57:54] * jugg joined the chat.
[06:58:22] * ermine joined the chat.
[07:16:45] * teo left the chat.
[07:16:45] * teo joined the chat.
[07:31:14] * Kev left the chat.
[07:32:05] * Kev joined the chat.
[07:36:08] * nabatt joined the chat.
[07:37:47] * john joined the chat.
[07:47:51] * jprieur joined the chat.
[07:47:52] * jprieur left the chat.
[08:09:54] * Tobias joined the chat.
[08:12:41] * luca tagliaferri joined the chat.
[08:23:15] * petermount joined the chat.
[08:25:02] * Neustradamus left the chat.
[08:26:35] * fantasticsid left the chat.
[08:37:43] * scippio joined the chat.
[08:56:49] * alkino joined the chat.
[09:06:56] * will.thompson joined the chat.
[09:28:39] * niekie left the chat.
[09:52:09] * Link Mauve joined the chat.
[10:10:18] * bowman joined the chat.
[10:10:30] * bowman left the chat.
[10:18:19] * jugg left the chat.
[10:47:52] * scippio left the chat.
[10:47:56] * scippio joined the chat.
[11:11:11] * Neustradamus joined the chat.
[11:18:37] * mlundblad joined the chat.
[11:38:18] * Link Mauve left the chat.
[11:40:11] * Link Mauve joined the chat.
[11:40:39] * Link Mauve left the chat.
[12:33:49] * Link Mauve joined the chat.
[12:58:43] * MattJ joined the chat.
[13:02:03] * niekie joined the chat.
[13:04:28] <will.thompson> is it just me, or do I keep getting sent the same scrollback over and over again? am i bouncing in and out of this conference?
[13:05:40] <petermount> I last saw you enter at 10.06 BST and according to me you've stayed in here ever since...
[13:05:41] <Kev> Are you using Google Talk?
[13:05:50] * Kevmin joined the chat.
[13:05:54] <will.thompson> Kev: nope
[13:06:05] <Kev> Odd, it's usually a Google Talk bug that causes that.
[13:06:17] <Kev> Can you PM me an XML log?
[13:06:29] <will.thompson> Kev: just turned on debugging, I'm going to wait for it to happen again... :)
[13:06:31] <petermount> kev: I've not seen anything today to show the google talk 'bug'
[13:06:49] <Kev> petermount: you're not connected from Google Talk.
[13:07:34] <petermount> not at the moment, but I usually see presences from those who do in here - and not seen any for a while
[13:07:53] <will.thompson> Kev: I think collabora.co.uk still runs ejabberd, though it might be prosody
[13:08:18] <Kev> petermount: No-one in here is currently using GTalk.
[13:08:35] <Kev> will.thompson: Let's see what the XML says, it could be a client bug too ;)
[13:09:15] <will.thompson> Kev: oi :p
[13:10:16] <Kev> I'll wait for the XML before throwing wild accusations around :D
[13:10:38] <Kev> Even if I have suspicions.
[13:12:57] * Xificurk joined the chat.
[13:15:24] * Tobias left the chat.
[13:19:31] <will.thompson> Kev: what nature would you expect the client bug to have?
[13:19:38] * teo left the chat.
[13:19:46] <Kev> I wouldn't want to hint that there could be a client bug.
[13:20:17] <Kev> If a client were to resend the join stanza, that would cause the room to force a rejoin, while not showing it to other occupants, though.
[13:20:45] <Kev> That's what causes the problem for GTalk users, the server resends the presence stanza every 5 minutes, including the join element.
[13:21:19] <Kev> So the server says "Oh, I thought they were here, but they're clearly saying they're not here and they want to join. Well, I'd better resend them the history then, they won't have that if they haven't joined yet".
[13:21:20] <MattJ> will.thompson, FWIW collabora.co.uk is still using ejabberd at the moment
[13:22:15] <will.thompson> MattJ: You haven't bribed Tollef enough yet then :)
[13:22:33] <MattJ> No bribing necessary :)
[13:22:38] <will.thompson> Kev: Oh, yeah, saw that one on some list
[13:24:52] * tofu left the chat.
[13:25:58] * tofu joined the chat.
[13:31:01] <Kev> will.thompson: Is there any chance that this is behind it?
[13:31:59] <will.thompson> sigh. it's happening every time i change status
[13:32:09] <MattJ> Client bug :)
[13:32:22] <Kev> will.thompson: Sounds like the mentioned issue, then.
[13:32:33] <Kev> Check you're not sending the join element in every presence update.
[13:32:39] <MattJ> It shouldn't include <x xmlns=...muc...> in the presence
[13:35:16] <will.thompson> you're spot on
[13:35:22] * roga joined the chat.
[13:35:24] <will.thompson> sigh. another regression in the muc code refactoring
[13:36:40] * roga left the chat.
[13:37:41] <MattJ> will.thompson, refactoring? Is this issue in any release?
[13:38:39] <will.thompson> yes, the last few 0.9 releases
[13:39:07] <Kev> Surprising we've not had reports already.
[13:39:58] <will.thompson> yeah, guess it shows how many Telepathy users use conference.jabber.org MUCs ...
[13:40:28] <MattJ> 0.9 isn't in Lucid though anyway
[13:41:04] <Kev> Time to get a fix out before 10.10 :)
[13:41:13] <MattJ> :)
[13:41:42] * dwd joined the chat.
[13:44:08] <dwd> will.thompson, If you want to moan that it's all my fault, feel free.
[13:44:28] <will.thompson> hah, i'm just checking whether it's actually a regression as opposed to longstanding behavioulr
[13:44:33] <will.thompson> i understand the rationale though
[13:44:34] * Neustradamus left the chat.
[13:44:56] * Zash joined the chat.
[13:47:50] <dwd> Yeah, it helps in a few cases.
[13:48:20] * mlundblad_netbook joined the chat.
[13:50:24] <will.thompson> ah, it's not even a regression! it's been there forever
[13:50:53] <Kev> \o/
[13:54:19] <will.thompson> my favourite kind of bug.
[14:12:35] * mlundblad_netbook left the chat.
[14:13:30] * Guus joined the chat.
[14:16:19] * Zash left the chat.
[14:16:32] * Zash joined the chat.
[15:14:19] * john left the chat.
[15:14:52] * stpeter joined the chat.
[15:16:25] * deryni joined the chat.
[15:16:30] * jkhii joined the chat.
[15:20:30] * tkoski joined the chat.
[15:25:52] * Zash left the chat.
[15:36:56] * john joined the chat.
[15:42:44] * mlundblad_netbook joined the chat.
[15:52:38] * hawke joined the chat.
[15:55:55] * scippio left the chat.
[15:56:00] * scippio joined the chat.
[16:02:35] * Neustradamus joined the chat.
[16:05:07] * lastsky left the chat.
[16:09:30] * nabatt left the chat.
[16:14:47] * dbanes joined the chat.
[16:17:07] * mlundblad_netbook left the chat.
[16:17:37] * Zash joined the chat.
[16:21:24] * paranoik joined the chat.
[16:25:16] * bjc joined the chat.
[16:26:53] * paranoik left the chat.
[16:34:40] * teo joined the chat.
[16:34:45] * Zash left the chat.
[16:36:10] * bear joined the chat.
[16:39:20] * dbanes left the chat.
[16:40:48] * tkoski left the chat.
[16:43:10] * norbertg joined the chat.
[16:44:21] * norbertg left the chat.
[16:49:52] * scippio left the chat.
[16:57:44] * john left the chat.
[17:00:44] * john joined the chat.
[17:05:03] * steve-e joined the chat.
[17:05:47] * petermount left the chat.
[17:09:49] * teo left the chat.
[17:09:50] * teo joined the chat.
[17:20:00] * hildjj joined the chat.
[17:20:19] * luca tagliaferri left the chat.
[17:23:11] * linuxwolf joined the chat.
[17:25:55] * Zash joined the chat.
[17:29:14] * linuxwolf left the chat.
[17:29:24] * linuxwolf joined the chat.
[17:30:45] * waqas joined the chat.
[17:33:31] * zanchin left the chat.
[17:34:46] * zanchin joined the chat.
[17:38:38] <bear> prosody.example.com 10.10.30.212
[17:39:46] <hildjj> ack
[17:39:51] <bear> ta
[17:40:42] * Jon Snyder joined the chat.
[17:40:43] * Jon Snyder left the chat.
[17:40:43] * Jon Snyder joined the chat.
[17:40:43] * Jon Snyder left the chat.
[17:42:09] * steve-e left the chat.
[17:42:34] * Jon Snyder joined the chat.
[17:42:34] * Jon Snyder left the chat.
[17:43:21] * SteffenL joined the chat.
[17:43:21] * SteffenL left the chat.
[17:43:21] * SteffenL joined the chat.
[17:43:21] * SteffenL left the chat.
[17:44:55] * linuxwolf left the chat.
[17:45:05] * linuxwolf joined the chat.
[17:45:05] <hildjj> The DNS server is here: 10.10.27.123
[17:45:48] * waqas left the chat.
[17:45:50] * waqas joined the chat.
[17:46:05] * waqas left the chat.
[17:46:10] * waqas joined the chat.
[17:47:32] * SteffenL joined the chat.
[17:47:32] * SteffenL left the chat.
[17:47:35] * Jon Snyder joined the chat.
[17:47:35] * Jon Snyder left the chat.
[17:47:58] * linuxwolf left the chat.
[17:49:05] * linuxwolf joined the chat.
[17:49:11] <hildjj> dig +short @localhost _xmppconnect.prosody.example.com TXT
[17:49:17] * jsautret joined the chat.
[17:49:46] <hildjj> I've got prosody and xcp set up. anyone else ready? paste your IP addr here.
[17:49:56] * Link Mauve left the chat.
[17:50:28] <jsautret> ejabberd.example.com is 10.10.29.122
[17:51:03] <hildjj> got it.
[17:51:13] * waqas left the chat.
[17:51:36] * Fritzy joined the chat.
[17:51:39] * waqas joined the chat.
[17:51:41] * Jack Erwin joined the chat.
[17:52:11] <hildjj> dns server: 10.10.27.123
[17:52:33] * SteffenL joined the chat.
[17:52:33] * SteffenL left the chat.
[17:52:35] * Jon Snyder joined the chat.
[17:52:35] * Jon Snyder left the chat.
[17:52:42] <waqas> I see M-Link is storing chat state notifications in the room history. Prosody does the same. I'm on the fence about that being the right thing to do or not.
[17:52:43] * Jack Erwin left the chat.
[17:52:46] <waqas> Thoughts?
[17:53:19] <hildjj> To forward them at all is the question. storing them is right out.
[17:53:35] <Kev> Forwarding is the right thing.
[17:53:38] <Kev> Storing isn't.
[17:53:42] <Zash> +1
[17:54:04] <Kev> Well.
[17:54:10] <waqas> So what's the right thing to do? Just store messages with a <body/>?
[17:54:15] <Kev> That's not clear even.
[17:54:19] <Kev> Maybe they should be tored.
[17:54:33] <waqas> There's also <html/>, and something a new XEP might define.
[17:54:36] <hildjj> they shouldn't count toward the history limit, at the very least.
[17:54:43] <Kev> waqas: right.
[17:54:44] * niekie left the chat.
[17:55:00] <waqas> hildjj: Well do they go in a black list, or does <body/> go in a whitelist?
[17:55:01] * niekie joined the chat.
[17:55:55] <Zash> nick changes could be useful to save?
[17:56:05] * Jack Erwin joined the chat.
[17:56:08] <waqas> It's a bit annoying to join a room with a 20 message history size, and get less than 10 because some clients have been spamming it with chat state notifications.
[17:56:29] <Kevmin> Right.
[17:56:38] * john left the chat.
[17:56:43] <Kevmin> But then if you join a room, and someone's in the middle of typing...
[17:57:33] * SteffenL joined the chat.
[17:57:34] * SteffenL left the chat.
[17:57:35] * Jon Snyder joined the chat.
[17:57:35] * Jon Snyder left the chat.
[17:59:18] * linuxwolf left the chat.
[17:59:27] * linuxwolf joined the chat.
[18:01:19] <hildjj> Kevmin: what if i never care about typing indicators in a room and see it as a scale bug to allow them at all?
[18:02:18] <Kevmin> Then I think you'd disable the option to allow arbitrary payloads in the MUC.
[18:02:35] * Jon Snyder joined the chat.
[18:02:36] * Jon Snyder left the chat.
[18:04:28] <hildjj> i'm more for blacklist. there are some that are some protocols that are not as useful as the amount of traffic.
[18:05:29] * linuxwolf left the chat.
[18:05:38] * linuxwolf joined the chat.
[18:07:34] * Zash left the chat.
[18:07:36] * Jon Snyder joined the chat.
[18:07:36] * Jon Snyder left the chat.
[18:07:45] * Zash joined the chat.
[18:08:55] * Jack Erwin left the chat.
[18:08:55] * jaerwin joined the chat.
[18:11:42] * linuxwolf left the chat.
[18:11:50] * linuxwolf joined the chat.
[18:12:36] * Jon Snyder joined the chat.
[18:12:36] * Jon Snyder left the chat.
[18:12:52] * jaerwin left the chat.
[18:15:03] <hildjj> ok, is this thing still on?
[18:15:08] <hildjj> looks like it.
[18:15:27] <Kev> Hope so.
[18:15:41] <bear> prosody.example.com 192.168.87.57
[18:16:41] <waqas> /me wonders what the hostnames/ips bear and hildjj are posting are about
[18:16:51] * Jon Snyder joined the chat.
[18:16:51] * Jon Snyder left the chat.
[18:16:51] * Jon Snyder joined the chat.
[18:16:51] * Jon Snyder left the chat.
[18:16:56] * will.thompson left the chat.
[18:17:13] <Kev> waqas: Servers for interop testing.
[18:17:29] <Kev> What are you guys testing? Kurt mentioned SCRAM, what else?
[18:17:36] * Jon Snyder joined the chat.
[18:17:36] * Jon Snyder left the chat.
[18:19:59] * bear left the chat.
[18:20:32] * linuxwolf left the chat.
[18:20:41] <hildjj> BOSH, s2s tls
[18:21:45] * linuxwolf joined the chat.
[18:21:55] * hawke left the chat.
[18:22:29] * hawke joined the chat.
[18:22:36] * Jon Snyder joined the chat.
[18:22:36] * Jon Snyder left the chat.
[18:22:48] <Kev> Ah, all good stuff.
[18:23:01] <Kev> I'll look forward to Kurt's report:)
[18:26:51] * Jon Snyder joined the chat.
[18:26:51] * Jon Snyder left the chat.
[18:26:51] * Jon Snyder joined the chat.
[18:26:51] * Jon Snyder left the chat.
[18:27:36] * Jon Snyder joined the chat.
[18:27:37] * Jon Snyder left the chat.
[18:31:44] * hildjj left the chat.
[18:32:37] * Jon Snyder joined the chat.
[18:32:37] * Jon Snyder left the chat.
[18:32:57] * Zash left the chat.
[18:33:09] * Zash joined the chat.
[18:34:47] * waqas left the chat.
[18:37:37] * Jon Snyder joined the chat.
[18:38:11] * linuxwolf left the chat.
[18:38:20] * linuxwolf joined the chat.
[18:39:13] * waqas joined the chat.
[18:44:22] * linuxwolf left the chat.
[18:44:31] * linuxwolf joined the chat.
[18:46:37] * linuxwolf left the chat.
[18:46:57] * Fritzy left the chat.
[18:47:45] * bear joined the chat.
[18:48:39] * teo left the chat.
[18:48:42] * teo joined the chat.
[18:49:00] <bear> the xmpp summit folks have stepped out for lunch - FYI
[18:49:30] * waqas left the chat.
[18:49:40] * niekie left the chat.
[18:49:45] <Kev> Ta.
[18:50:12] <bear> we have created a private network and have almost all of the servers now setup
[18:50:29] <Kev> Anyone broken the M-Link server yet? :)
[18:50:31] <bear> going to test bosh clients against them
[18:50:36] <bear> not yet
[19:04:58] <dwd> On that MUC history thing, I'm a little hesitant at breaking the link between what was published and the items held. But I suppose I could do it if pushed.
[19:05:00] * Florob joined the chat.
[19:08:48] * hawke left the chat.
[19:11:49] <bear> as a client user, I would find it unsettling if I got the history for muc differently than if I was present in the room (/me hopes he understands the point well enough to comment)
[19:20:05] * Link Mauve joined the chat.
[19:21:04] * hawke joined the chat.
[19:24:46] <dwd> It's partly that.
[19:25:06] <dwd> It's also partly that I worry about different use-cases of MUC to simple chat.
[19:25:31] <dwd> And finally it's because some hind-brain developed over years of IETF stuff is screaming LAYER VIOLATION!!!
[19:25:43] <bear> hahaha
[19:26:01] <bear> for that reason alone ... ;)
[19:26:05] <dwd> It's not funny.
[19:26:44] <bear> sorry - bad context for chuckle - was laughing because of how you phrased it
[19:26:44] <dwd> THe only thing that would make it worse is if the voices in my head were the ghosts of OSI.
[19:27:39] <bear> but yea, if your getting a lizard-brain reflex reaction to it - then it should be considered
[19:30:13] <dwd> There's something else, actually. It smells like a workaround.
[19:30:20] <dwd> ANd workarounds don't smell nice.
[19:31:53] <MattJ> Storing only messages with a <body> smells like a workaround?
[19:36:01] <dwd> Sure. WHat if what you're using the MUC for doesn't have <body/> in the message?
[19:36:21] <dwd> Do you then add bodies in, to defeat the "fix"?
[19:36:55] <dwd> <body>This space left intentionally blank</body>
[19:37:02] <MattJ> <thisishistory xmlns="urn:xmpp:makehistory"/>
[19:37:26] <MattJ> Now spot the real protocol designer here :)
[19:38:06] <dwd> The one not inventing pointless elements with unregistered namespaces?
[19:39:10] <MattJ> You're no fun
[19:39:23] <dwd> MattJ, I am if you buy me a drink.
[19:39:41] <MattJ> I'm surprised the IETF doesn't have targets to meet for IANA registrations yet, etc.
[19:40:04] <dwd> I think it has one for registries, judging by recent behaviour.
[19:40:54] * Tobias joined the chat.
[19:41:38] <MattJ> My only other complaint with the IETF is that I think they should re-assign stpeter to rewriting the websocket specs
[19:42:06] <MattJ> You know, so there are actually specs, and actually implementations, that actually work with each other
[19:43:05] <dwd> MattJ, I did actually consider that. But then, PSA's already up to his neck.
[19:43:17] <MattJ> Indeed, hence "re-assign"
[19:44:15] <dwd> MattJ,Can't pull him out of an AD position.
[19:44:29] <dwd> MattJ, Well, actually, that *can* be done. But it would be making history.
[19:45:21] <MattJ> Pft, bureaucracy :)
[19:45:22] <dwd> It'd be quite funny if Recall was invoked for the first time to get an AD to edit more specs.
[19:46:01] <bear> hilldj is working on some folks here - i'm going to talk to my moz boss and see if I can do it in my mozilla name
[19:49:23] <dwd> hilldj has been (rightly) looking for more involvement for a while.
[19:49:30] * waqas joined the chat.
[19:49:52] <dwd> I did try, but I somewhat lack energy. ANd it's hard to debate things with someone who only replies to you as a paragraph in a monthly digest reply...
[19:52:38] * Guus left the chat.
[19:52:55] * scippio joined the chat.
[19:57:15] * Florob left the chat.
[19:57:59] * Florob joined the chat.
[20:00:32] * waqas left the chat.
[20:01:59] * waqas joined the chat.
[20:09:39] <waqas> So, did we decide what to do with chat state notifications in MUC history? :)
[20:10:13] <Kev> Keep them if the room allows them :)
[20:10:17] <MattJ> waqas, yes, we concluded the same thing you and I already concluded
[20:10:27] <waqas> Which was what?
[20:10:34] <MattJ> There's no "right" way :)
[20:10:47] <waqas> What's the default? ;)
[20:11:06] <Kev> I'd say the default should be to store as history whatever was sent to the room.
[20:11:08] <MattJ> I'd say the one that surprises the least users
[20:12:32] <waqas> Hmm, maybe strip elements based on an xmlns black list (which includes chat state notifications), and not include messages with no children left.
[20:13:02] <Kev> For history, or for sending?
[20:13:09] <waqas> History
[20:13:17] <Kev> Why not do the same for sending?
[20:13:56] <waqas> Because it doesn't surprise users in normal broadcasts.
[20:14:28] <Kev> Then why should it surprise them on context?
[20:14:53] <MattJ> No, I'd say surprising the least users is simply storing only messages with a <body>
[20:15:01] <Kev> Fewest
[20:15:06] <MattJ> Yes, thanks :)
[20:15:11] <MattJ> I can't wriggle out of that one ;)
[20:15:33] * Fritzy joined the chat.
[20:15:45] <waqas> Kev: It's surprising in history because visible messages are fewer on join than they should be.
[20:16:03] <Kev> waqas: How does the use k
[20:16:14] <Kev> How does the user know how many there should be?
[20:16:29] <MattJ> When they get 1, and they usually get 20... they know
[20:16:42] <MattJ> and it's happening fairly often in the Prosody MUC
[20:16:54] <MattJ> I join and see little or no history
[20:17:05] <waqas> Just let a couple of empathy users play around in a muc, and history magically disappears.
[20:17:44] <Kev> I'd say that means your client should be requesting history better :)
[20:17:48] <Kev> Not that Swift does it right yet.
[20:17:52] <MattJ> It should?
[20:18:22] <MattJ> Most servers only store a certain number of stanzas to avoid using unlimited resources
[20:18:41] <Kev> Yes.
[20:19:12] * waqas left the chat.
[20:19:28] * waqas joined the chat.
[20:20:24] <MattJ> To be honest I'm not quite sure anyone needs to log into a channel and know that someone was typing before they sent a message some hours ago
[20:20:26] <waqas> No empathy users are active in this room at the moment. I got all 20 messages ^^
[20:20:51] <waqas> But the reason I brought the topic up was because earlier today I got less than 10 in this room.
[20:21:10] <MattJ> Yes, we'd been speaking with an Empathy user :)
[20:21:22] <waqas> And does something like <composing xmlns="http://jabber.org/protocol/chatstates" /> even make sense non-live?
[20:21:32] <MattJ> > [21:20:23] MattJ: To be honest I'm not quite sure anyone needs to log into a channel and know that someone was typing before they sent a message some hours ago
[20:21:40] <Kev> Some hours ago? No.
[20:21:47] <MattJ> Some minutes ago?
[20:21:49] <Kev> But context itself isn't interestng hours later.
[20:22:03] <Kev> It's interesting when you're entering themiddle of a discussion.
[20:22:15] <MattJ> I'm not sure that always holds true
[20:22:17] * linuxwolf joined the chat.
[20:22:23] <waqas> Does Swift send or display these notifications in MUC by the way?
[20:22:37] <Kev> It's not contxt if it's older than that, it's history, and that's a different problem to solve.
[20:22:37] <waqas> Does any client besides Empathy? It's the only one I've noticed.
[20:22:41] <Kev> waqas: Neither, I hope.
[20:22:53] <Kev> I guess it's possible I screwed up and t sends them.
[20:23:28] <MattJ> I think clients /should/ send them
[20:23:35] <MattJ> and display them
[20:24:02] <MattJ> I'd argue that the only reason they don't is because people are used to IRC
[20:24:29] <waqas> I agree with that.
[20:24:32] <Kev> The reason the don't is because people are scared of amplification.
[20:24:51] <Kev> I don't think it has anything to do with IRC.
[20:25:26] <MattJ> Which people are scared of it?
[20:25:38] <MattJ> It's not up to client devs to be scared of it :)
[20:25:40] * linuxwolf left the chat.
[20:25:42] <Kev> Probably no-one, these days.
[20:25:42] * linuxwolf joined the chat.
[20:25:51] <MattJ> It's up to server admins and server developers if anyone
[20:26:05] <MattJ> Let the server choose not to broadcast if the room is over a certain size, etc.
[20:26:28] <waqas> Or only broadcast to those who are active :)
[20:29:59] <MattJ> Indeed
[20:31:23] * alkino left the chat.
[20:31:23] * alkino joined the chat.
[20:37:15] * linuxwolf left the chat.
[20:38:10] * Guus joined the chat.
[20:43:43] * luca tagliaferri joined the chat.
[20:44:09] * Guus left the chat.
[20:44:42] * alkino left the chat.
[20:44:43] * Guus joined the chat.
[20:44:58] * bear left the chat.
[20:45:06] * bear joined the chat.
[20:45:21] * waqas left the chat.
[20:46:09] * Guus left the chat.
[20:46:27] * Guus joined the chat.
[20:48:57] * hildjj joined the chat.
[20:50:54] * Zash left the chat.
[20:51:09] * Zash joined the chat.
[20:51:26] * ermine left the chat.
[20:56:46] <hildjj> ok, i'm hildjj@xcp.example.com
[20:56:56] <hildjj> xmpp:hildjj@xcp.example.com, even.
[20:57:39] * Guus left the chat.
[20:58:13] * Guus joined the chat.
[21:02:27] <bear> is any prosody devs online?
[21:03:02] <johnny> bear, try the room..
[21:03:16] <johnny> /me isn't sure if any devs have a alert when saying prosody
[21:03:53] <bear> can you remind me of the room ;)
[21:04:02] <Kev> prosody@muc.prosody.im iirc.
[21:04:06] <bear> t
[21:04:07] <bear> a
[21:04:48] <jsautret> I'm xmpp:jerome@ejabberd.example.com
[21:06:19] <jkhii> bear, the room is prosody@conference.prosody.im
[21:06:23] <bear> prosody@conference.prosody.im
[21:06:26] <bear> :)
[21:07:54] <MattJ> The number of times "prosody" is said in the Prosody room would make an alert painful :)
[21:09:00] * zanchin left the chat.
[21:09:18] <Tobias> heh
[21:09:30] * zanchin joined the chat.
[21:10:30] <Fritzy> Prosody ding!
[21:13:51] <johnny> MattJ, i of course mean per room notices
[21:14:11] <johnny> bear, what are you doing with prosody?
[21:14:32] * zanchin left the chat.
[21:14:51] <hildjj> xmpp:oscon@conference.xcp.example.com?join
[21:15:41] <bear> bear@prosody.example.com
[21:15:54] <bear> prosody.example.com is running and ready for logins
[21:17:09] * Guus left the chat.
[21:17:57] * zanchin joined the chat.
[21:18:18] <johnny> why are folks pasting links to example.com ?
[21:18:30] <johnny> does oreilly own example.com ?
[21:18:41] <bear> no - we have setup a localnetwork
[21:18:45] <bear> with our own dns
[21:18:50] <johnny> ah
[21:18:54] <johnny> why uhm local
[21:19:05] <bear> we are just using jdev as our default chat room :)
[21:19:23] <bear> because we are being polite to the oscon network and also avoiding being hack targets
[21:19:25] <hildjj> until we bootstrap a local one that everyone can talk to…
[21:19:59] <Zash> at oscon ey?
[21:20:05] <bear> yea
[21:20:11] <Tobias> bear: is example.com even resolved by dns libs? :)
[21:20:34] <bear> with some nudging
[21:21:26] * Guus joined the chat.
[21:21:26] <hildjj> some browsers dislike hitting it, but the resolvers work fine
[21:21:33] <Guus> Yey for internet connectivity in hotels
[21:21:45] * Jon Snyder left the chat.
[21:21:52] * Jon Snyder joined the chat.
[21:21:52] * Jon Snyder left the chat.
[21:22:00] <Tobias> hildjj: ah..okay..because i saw bowsers choking and thought that came from DNS level
[21:23:12] <Zash> You have reached this web page by typing "example.com", ... These domain names are reserved for use in documentation and are not available for registration. See RFC 2606, Section 3.
[21:23:16] <Zash> worksforme
[21:24:01] <MattJ> ditto
[21:25:27] * Jon Snyder joined the chat.
[21:25:27] * Jon Snyder left the chat.
[21:30:28] * Jon Snyder joined the chat.
[21:30:28] * Jon Snyder left the chat.
[21:30:30] <bear> http://prosody.example.com:5280/http-bind <-- upand running
[21:31:10] * tofu left the chat.
[21:31:18] * tofu joined the chat.
[21:35:28] * Jon Snyder joined the chat.
[21:35:28] * Jon Snyder left the chat.
[21:40:28] * Jon Snyder joined the chat.
[21:40:28] * Jon Snyder left the chat.
[21:40:37] <hildjj> anyone have a link to the latest PSI beta handy?
[21:40:43] <hildjj> it's google-opaque. :(
[21:40:58] <Kev> I don't think there are nightlies any more.
[21:41:09] <Kev> And the latest beta release is the stable 0.14.
[21:41:52] <hildjj> ok, then different q. with 0.14, does it support BOSH?
[21:42:02] <Kev> No, Psi doesn't support BOSH.
[21:42:17] <hildjj> ah, cool. now i feel less dumb. couldn't figure out how to configure it. :0
[21:42:18] <hildjj> :)
[21:42:19] <Kev> Nor in Git, as far as I know.
[21:42:40] <Zash> Kev: Does Swift/-en ?
[21:42:44] <Kev> No.
[21:44:24] * tofu left the chat.
[21:44:26] * tofu joined the chat.
[21:45:11] <Tobias> pidgin supports BOSH
[21:45:20] <Zash> and gajim
[21:45:20] <Tobias> including detection of alternative connection methods
[21:45:29] * Jon Snyder joined the chat.
[21:45:29] * Jon Snyder left the chat.
[21:45:40] <Neustradamus> Kev: it is the negative point of Psi and Swift :(
[21:45:59] <Tobias> Neustradamus: only if you are in restricted network environments :)
[21:46:07] <hildjj> trying Adium.
[21:46:16] <Tobias> with a bit SRV tweaking you get through most firewalls too
[21:46:32] <Tobias> as long as you make your XMPP server listen on 443 :)
[21:47:29] <Zash> Hm, is there something like inetd for muliplexing a single port?
[21:47:43] <hildjj> different URLs...
[21:47:44] <Tobias> i know prosody does multiplexing :)
[21:48:00] * bjc left the chat.
[21:48:04] <hildjj> also, some people set up apache or something to proxy BOSH URLs through
[21:48:35] * tofu left the chat.
[21:48:37] * tofu joined the chat.
[21:48:47] <Zash> HTTPS + VPN on the same port would be nice
[21:50:29] * Jon Snyder joined the chat.
[21:50:29] * Jon Snyder left the chat.
[21:51:23] <Zash> How much time does it usualy take for clients to skip a non-working SRV retord?
[21:51:27] <Zash> record*
[21:51:52] <Tobias> that'd be OS dependent i think
[21:52:14] <Tobias> i mean connection to a non listening port fails quite fast mostly
[21:52:33] <Zash> Like if I put a record pointing to a VPN IP
[21:53:01] <Tobias> just try it out :)
[21:53:03] <Zash> They don't do a gazillion connection attempts before trying the next then?
[21:53:30] <Tobias> they usually try once and if that fails try the next SRV record
[21:55:29] * Jon Snyder joined the chat.
[21:55:29] * Jon Snyder left the chat.
[21:58:43] * wiretap joined the chat.
[22:00:29] * Jon Snyder joined the chat.
[22:00:29] * Jon Snyder left the chat.
[22:04:10] * mlundblad left the chat.
[22:05:30] * Jon Snyder joined the chat.
[22:05:30] * Jon Snyder left the chat.
[22:09:55] <MattJ> /me pokes summit-going people
[22:10:30] * Jon Snyder joined the chat.
[22:10:30] * Jon Snyder left the chat.
[22:11:25] <MattJ> I guess they found something more interesting to do :)
[22:12:45] <Tobias> MattJ: more interesting than this...sounds impossible
[22:12:54] <MattJ> :)
[22:13:24] <MattJ> Yes, well, it's not *allowed* to shout "there's a bug in Prosody" and then disappear
[22:14:30] <MattJ> I'll just hack into the OSCON network and feed my webcam stream through one of their projectors
[22:14:55] <zanchin> hey MattJ, we're just all listening to jildjj talk about HYBI
[22:15:02] <zanchin> *hildjj
[22:15:11] <MattJ> Oh goodie :)
[22:15:17] <zanchin> i'll poke bear for you
[22:15:26] <Tobias> HYBI?
[22:15:30] * Jon Snyder joined the chat.
[22:15:31] * Jon Snyder left the chat.
[22:15:37] <MattJ> What's the conclusion? Down with websockets, up with BOSH? :)
[22:15:40] <zanchin> ah, bear is not online at the moment
[22:15:57] <MattJ> He's online according to here and my roster
[22:16:15] <zanchin> yea, (bear here) - swift hasn't figured out it is offline :)
[22:16:20] <MattJ> Heh
[22:16:37] * bear left the chat.
[22:16:40] <zanchin> the list of headers was matt from cisco saying that prosody was lacking some of the CORS headers
[22:17:15] <zanchin> when I get back online i'll give you more details
[22:17:21] <MattJ> Then it's Matt I need to talk with :)
[22:17:31] <zanchin> (bear handing laptop back to zanchin) - yes - talk to matt :)
[22:19:03] <MattJ> By "Matt" I'm guessing you mean Matthew Miller?
[22:19:10] <zanchin> yes
[22:19:13] <MattJ> Thanks
[22:20:31] * Jon Snyder joined the chat.
[22:20:31] * Jon Snyder left the chat.
[22:21:14] <Florob> Is that just me or did Mickaël Rémond really not reapply?
[22:21:29] <MattJ> Florob, very likely he didn't re-apply
[22:21:39] * linuxwolf joined the chat.
[22:21:42] <zanchin> Tobias: http://trac.tools.ietf.org/bof/trac/wiki/HyBi
[22:21:45] <Florob> MattJ, how so?
[22:22:09] <MattJ> Florob, don't ask me to dig up mailing list archives now :)
[22:23:03] <Tobias> zanchin: thanks
[22:24:22] <MattJ> linuxwolf, hey
[22:25:08] * hawke left the chat.
[22:25:24] * hawke joined the chat.
[22:25:32] * Jon Snyder joined the chat.
[22:25:32] * Jon Snyder left the chat.
[22:25:54] <Tobias> ahh..nice, scram interop is planned too :)
[22:26:19] <MattJ> Florob, http://mail.jabber.org/pipermail/members/2009-October/005241.html
[22:27:38] <linuxwolf> /me waves
[22:28:16] <MattJ> linuxwolf, tell me what you think is wrong with Prosody's HTTP headers and I'll tell you why you're wrong :)
[22:28:28] <linuxwolf> Access-Control-Allow-Origin
[22:28:32] <linuxwolf> not present
[22:29:10] * Guus left the chat.
[22:29:18] <MattJ> It's only sent if cross-domain support is configured
[22:29:35] <linuxwolf> ok, so I'll yell @ bear
[22:29:38] <linuxwolf> (-:
[22:29:53] <MattJ> Tell him cross_domain_bosh = true
[22:30:32] * Jon Snyder joined the chat.
[22:30:32] * Jon Snyder left the chat.
[22:31:28] <linuxwolf> [from bear] that's missing from the wiki (-:
[22:31:34] <linuxwolf> /me is a proxy
[22:31:45] <Florob> MattJ, that's approximately what I had feared upon your initial answer... Thank's for digging that up
[22:31:53] <MattJ> linuxwolf, doesn't look like it to me: http://prosody.im/doc/modules/mod_bosh :)
[22:32:07] <linuxwolf> don't shoot the messenger! (-:
[22:32:32] <linuxwolf> [from bear] yes, I'm an idiot
[22:33:04] * scippio left the chat.
[22:33:34] <linuxwolf> thanks, MattJ
[22:33:41] * scippio joined the chat.
[22:34:02] <MattJ> Jack reckons it should be enabled by default, I'm unsure
[22:34:20] <linuxwolf> I would agree with Jack
[22:34:32] <MattJ> Well I erred on the side of caution
[22:34:47] <linuxwolf> the security implications are not any worse than simply enabling BOSH in the first place
[22:34:57] <MattJ> Are you sure?
[22:35:23] <MattJ> I mean, if there's an issue with someone else's site, it allows that site to be used as a proxy in an XMPP-related attack
[22:35:33] * Jon Snyder joined the chat.
[22:35:33] * Jon Snyder left the chat.
[22:35:34] <linuxwolf> mod_proxy_http
[22:36:03] <MattJ> Sure, but that has to be set up specifically by the person running such a site
[22:36:26] <linuxwolf> yes, but you (BOSH server) can't necessarily tell at that point
[22:36:53] * Guus joined the chat.
[22:37:10] <MattJ> It doesn't have to, really
[22:37:46] <MattJ> A server IP brute-forcing XMPP accounts is a little different than unsuspecting website visitors doing it
[22:40:13] <linuxwolf> this sounds like a very weak straw man to me
[22:40:23] <linuxwolf> IMO
[22:40:33] * Jon Snyder joined the chat.
[22:40:33] * Jon Snyder left the chat.
[22:41:02] <MattJ> I'm not sure that it is - at the end of the day changing the setting is as simple as changing a "false" to "true" in our code
[22:41:15] <linuxwolf> then change it!
[22:41:16] <linuxwolf> (-:
[22:41:31] <MattJ> Not while I still have doubts about the wisdom of doing so :)
[22:42:35] <linuxwolf> you've got two pretty smart people telling you differently (-:
[22:42:39] <linuxwolf> at least (-:
[22:42:58] <Zash> /me votes false
[22:43:08] <MattJ> Thanks Zash. Are you smart?
[22:43:28] <Zash> I am!
[22:43:30] <MattJ> Yay!
[22:44:10] * Guus left the chat.
[22:44:56] <Zash> And I don
[22:45:10] * stpeter left the chat.
[22:45:10] <Zash> rr
[22:45:33] * Jon Snyder joined the chat.
[22:45:44] * Jon Snyder left the chat.
[22:48:39] * linuxwolf left the chat.
[22:48:48] * linuxwolf joined the chat.
[22:48:53] * tofu left the chat.
[22:48:56] * tofu joined the chat.
[22:49:44] <Zash> there's probably some relation between prosody's awesomeness and MattJ's smartness
[22:50:20] <linuxwolf> it could also be a case of paranoia (-:
[22:50:34] * Jon Snyder joined the chat.
[22:50:37] <MattJ> Statistics show that it pays to be paranoid
[22:50:37] * Jon Snyder left the chat.
[22:51:02] <linuxwolf> it's your server…but I'd still vote "true"
[22:52:37] <MattJ> I'll just find a moderately popular website with a cross-site scripting exploit and slip in some code that connects to your server and performs a bruteforce attack then :)
[22:53:27] <linuxwolf> by definition, the xss exploits bypass CORS…
[22:53:40] <MattJ> How so?
[22:54:22] <linuxwolf> hold on...
[22:54:54] <Zash> don't you mean s/cors/the same origin policy/
[22:54:59] <MattJ> CORS doesn't protect against XSS, it is there to limit the damage XSS can do
[22:55:34] * Jon Snyder joined the chat.
[22:55:37] <Zash> Isn't CORS a workaround for when you want to allow XSS !attacks?
[22:55:37] * Jon Snyder left the chat.
[22:55:48] <MattJ> Unquestionably allowing access from any site with XSS flaws to your XMPP server I see as a protentially regrettable decision
[22:56:38] <MattJ> and I'm sure there are cases it makes sense for someone to want to do that though - but that doesn't mean it should be the default
[22:57:12] <linuxwolf> CORS == less (web) server load
[22:57:18] <MattJ> Agreed
[22:57:26] <linuxwolf> SIGNIFICANTLY less load
[22:57:30] <MattJ> Agreed
[22:57:40] <linuxwolf> so why do you hate web servers? (-:
[22:57:48] <MattJ> I don't, I hate insecurities :)
[22:58:01] <Zash> linuxwolf: HTTP vs XMPP holy war!
[22:58:08] <Florob> /me waves good night
[22:58:12] <johnny> /me wars your holy
[22:58:14] <linuxwolf> /me sighs
[22:58:23] * Florob left the chat.
[22:58:30] <Zash> ;)
[22:58:44] <johnny> /me steals mattj's j
[22:59:02] <Zash> jjohnny?!
[22:59:02] <johnny> /me steals waqas' q
[23:00:35] * Jon Snyder joined the chat.
[23:03:48] * Fritzy left the chat.
[23:04:25] <Zash> waas - whatnow as a service?
[23:05:23] * linuxwolf left the chat.
[23:05:23] <johnny> Zash, you'd just be ash
[23:05:41] <johnny> but at least you could pretend to be a shell
[23:05:46] * linuxwolf joined the chat.
[23:05:46] * Jon Snyder left the chat.
[23:05:53] <johnny> /me replaces zash with a very small shell script
[23:05:56] <johnny> har har :)
[23:06:00] * Zash in now known as za.sh.
[23:06:07] <za.sh> noooes
[23:06:37] <johnny> ah.. that's right.. if we take the z, you get zsh .. which is more powerful
[23:06:42] <johnny> take the a*
[23:06:56] * za.sh in now known as Zash.
[23:07:01] * Fritzy joined the chat.
[23:07:10] <Zash> I should build my own shell some day
[23:07:13] <Zash> /me 's awesome shell
[23:08:11] <johnny> you could build it in lua
[23:08:44] * tofu left the chat.
[23:10:35] * Jon Snyder joined the chat.
[23:11:48] * linuxwolf left the chat.
[23:12:00] * linuxwolf joined the chat.
[23:12:03] * linuxwolf left the chat.
[23:16:56] * Link Mauve left the chat.
[23:22:40] * Neustradamus left the chat.
[23:25:32] * zanchin left the chat.
[23:34:53] * Tobias left the chat.
[23:39:05] * luca tagliaferri left the chat.
[23:44:56] * zanchin joined the chat.
[23:46:42] * hawke left the chat.
[23:48:57] * teo left the chat.
[23:48:57] * Jon Snyder left the chat.
[23:48:59] * teo joined the chat.
[00:15:41] * bear joined the chat.
[00:27:26] * Fritzy left the chat.
[00:27:34] * hildjj left the chat.
[00:29:39] * jkhii left the chat.
[00:32:29] * zanchin left the chat.
[00:40:13] * MattJ left the chat.
[00:41:07] * jugg joined the chat.
[00:42:11] * jsautret left the chat.
[00:44:22] * Zash left the chat.
[00:48:58] * bear left the chat.
[00:58:31] * dejan1 joined the chat.
[00:58:31] * dejan1 left the chat.
[01:00:42] * dejan1 joined the chat.
[01:07:19] * teo left the chat.
[01:07:30] * teo joined the chat.
[01:16:31] * dejan1 left the chat.
[01:16:35] * dejan1 joined the chat.
[01:16:36] * dejan1 left the chat.
[01:16:36] * dejan1 joined the chat.
[01:16:36] * dejan1 left the chat.
[01:21:24] * dejan1 joined the chat.
[01:21:24] * dejan1 left the chat.
[01:23:04] * dejan1 joined the chat.
[01:27:01] * dejan1 left the chat.
[01:57:37] * kkszysiu joined the chat.
[02:08:33] <kkszysiu> Hello
[02:08:58] <kkszysiu> Do you know any library in C that support BOSH connections?
[02:22:39] * kkszysiu left the chat.
[02:23:14] * grom joined the chat.
[02:33:52] <johnny> ah too late :(
[02:34:03] * teo left the chat.
[02:34:23] * teo joined the chat.
[02:35:52] * grom left the chat.
[02:53:09] * jkhii joined the chat.
[03:02:24] * scippio left the chat.
[03:02:45] * scippio joined the chat.
[03:25:18] * duongndvtc joined the chat.
[03:25:19] * duongndvtc left the chat.
[04:28:38] * johnny left the chat.
[04:29:37] * johnny joined the chat.
[04:36:22] * peregrine joined the chat.