Logs for jdev

Show join/part/nick changes:

[00:34:13] * Treebilou joined the chat.
[00:47:39] * bjc left the chat.
[00:57:46] * Tobias left the chat.
[01:36:40] * duck1123 joined the chat.
[02:11:06] * Florob left the chat.
[02:24:06] * simon left the chat.
[02:34:31] * MattJ left the chat.
[02:35:23] * Zash left the chat.
[02:35:51] * darkrain_ left the chat.
[03:08:33] * darkrain joined the chat.
[03:24:38] <zanchin> am I correct in thinking that the pubsub spec doesn't specify whether the publish IQ response or the resulting notifications should be sent first?
[03:28:39] <louiz’> You mean, what comes first, the <iq type='result'/> or the <message /> that goes to all subscribers?
[03:28:47] <zanchin> yeah
[03:29:07] <darkrain> !xep 60
[03:29:08] <Kanchil> darkrain: XEP-0060: Publish-Subscribe is Standards Track (Draft, 2010-07-12) See: http://xmpp.org/extensions/xep-0060.html
[03:29:34] <louiz’> The pubsub service MUST then send one event notification to each entity that meets the criteria for receiving an event notification (typically to each approved subscriber
[03:29:40] <louiz’> see the *THEN*
[03:29:44] <darkrain> Yep, was about to paste the exact same sentence :)
[03:29:49] <louiz’> :)
[03:30:01] <louiz’> So, you MUST first send the iq, then the message, I think.
[03:30:09] <zanchin> hmm
[03:30:30] <darkrain> Why does it matter?
[03:30:47] <louiz’> (and that's also the most logical way, in my opinion)
[03:31:18] <darkrain> It certainly allows for background thread processing of fetching all the subscribers, doing processing about who wants notifications, and then generating those notifications
[03:31:31] <zanchin> it matters if the result stanza has useful information in it, such as a generated item id
[03:31:46] <darkrain> Why?
[03:31:54] <darkrain> s/Why/In what way/
[03:33:27] <louiz’> In suppose, something like "the iq says me that the item I just published is id=42" and when you receive the notification for the item 42 you have a"this is YOUR item"-like special notification?
[03:33:52] <zanchin> in my case I have a node-in-code, and the publisher needs to know the id of the published item
[03:34:10] <darkrain> zanchin: Why does either approach actually preclude you from getting that information?
[03:34:20] <zanchin> louz': yes
[03:34:37] <louiz’> but then, zanchin, that's ok, you'll get the iq before the message.
[03:34:56] <zanchin> darkrain: i'm trying to remember the specifics of my use case
[03:35:06] <zanchin> i think the problem is when the notification comes first
[03:35:29] <louiz’> darkrain, well, if you receive a notification but you've never seen this id before, you cannot tell if this item was posted by you, can you?
[03:35:37] <louiz’> (if you receive the message before the iq, say)
[03:35:48] <darkrain> If you care about the ID of the published item so much, why aren't you providing it? :)
[03:36:12] <louiz’> oh, that's right, you can… So I don't know :p
[03:36:17] <darkrain> ^^
[03:36:54] <zanchin> darkrain, you can't always provide it. Let's say you are adding an item to a container. Only the service can make sure that there are no item id collisions
[03:37:04] <darkrain> GUIDs?
[03:37:47] <zanchin> that'll probably work, but it won't guarantee it
[03:38:01] <louiz’> (I don't know 0060 very well, so I'll leave this conversation here and go to sleep, go night/day everyone \o)
[03:38:07] <darkrain> Encode characteristics of the publisher in the id in order to guarantee uniqueness
[03:38:40] <darkrain> Anyway, this is mostly philosophical, since the specification works the way you'd prefer (right?) :)
[03:38:45] <zanchin> louiz': good night
[03:38:46] <darkrain> I should stop arguing for the sake of arguing
[03:38:50] <darkrain> louiz’: nn
[03:39:44] <zanchin> darkrain, yes, there's no problem if the notification comes later
[03:39:57] <zanchin> but my implementation wasn't doing that
[03:40:20] <louiz’> but this MUST relies only on a little "then" in this giant spec, so maybe that's not very clear for everyone. Maybe adding just a little something to insist that the notification MUST be sent after the iq-result may be useful?
[03:40:30] <louiz’> To avoid buggy pubsub-services :p
[03:40:52] <louiz’> or I don't know :)
[03:40:53] <darkrain> louiz’: Now you're hoping that we've interpreted it correctly :)
[03:41:29] <louiz’> no, I'm saying that, since zanchin wasn't sure, maybe some pubsub-server writters aren't sure either, and will do the opposite way
[03:41:50] <darkrain> Right, i understand
[03:42:01] <zanchin> I do think that it should be more explicit
[03:42:15] <darkrain> My point is that you're making the assumptino (in saying it should be made more explicit) that we're correct
[03:42:21] <louiz’> I do think to, but I'm not an expert in writting XEPs either, so… :p
[03:42:22] <darkrain> zanchin: Please post to the pubsub@xmpp.org list :)
[03:42:29] <louiz’> ah, ok :D
[03:42:38] <zanchin> darkrain: ok
[03:42:41] <louiz’> Yes, I'm assuming we are correct :p
[03:43:08] <darkrain> I try to (at least internally) check my assumptions, since rarely does anyone else :/
[03:43:58] <louiz’> Yep, everyone should do that :)
[03:44:25] <darkrain> Check my assumptions? Yes. I get away with a little too much BS
[03:44:41] <darkrain> (being ~right a reasonable percentage of the time helps, but still)
[03:44:46] <darkrain> Anywho
[03:46:41] <louiz’> nope, check THEIR assumptions :p
[03:46:47] <louiz’> anyhow, bye :)
[03:56:25] <zanchin> darkrain: sent
[03:57:24] <darkrain> .
[04:00:46] <zanchin> well, time to go home, later all
[05:35:35] * Treebilou left the chat.
[05:36:05] * Treebilou joined the chat.
[05:36:33] * Neustradamus left the chat.
[05:36:46] * Neustradamus joined the chat.
[05:45:59] * bear left the chat.
[05:56:21] * tofu left the chat.
[06:27:51] * lastsky joined the chat.
[06:32:04] * Treebilou left the chat.
[07:51:01] * waqas joined the chat.
[07:55:39] * Treebilou joined the chat.
[07:56:35] * Treebilou left the chat.
[07:56:39] * Treebilou joined the chat.
[08:17:50] * lastsky_ joined the chat.
[08:25:48] * waqas left the chat.
[08:26:04] * niekie left the chat.
[08:58:56] * Kev left the chat.
[09:09:14] * lastsky_ left the chat.
[09:12:34] * Tobias joined the chat.
[09:13:06] * Tobias left the chat.
[09:28:30] * ermine joined the chat.
[09:54:31] * quack@jabber.org joined the chat.
[09:54:51] * quack@jabber.org left the chat.
[10:21:40] * Treebilou left the chat.
[10:26:20] * niekie joined the chat.
[10:42:03] * smoku left the chat.
[10:43:18] * smoku joined the chat.
[11:07:35] * niekie left the chat.
[11:36:38] * duck1123 left the chat.
[12:07:45] * duck1123 joined the chat.
[12:26:38] * Treebilou joined the chat.
[12:56:18] * MattJ joined the chat.
[13:10:27] * Ludovic joined the chat.
[13:39:57] * mlundblad joined the chat.
[14:16:16] * Robert joined the chat.
[14:18:06] * Robert left the chat.
[14:22:43] * luca tagliaferri joined the chat.
[14:22:45] * luca tagliaferri left the chat.
[14:34:43] * Robert joined the chat.
[14:53:40] * MattJ left the chat.
[14:53:41] * MattJ joined the chat.
[15:06:18] * MattJ left the chat.
[15:06:19] * MattJ joined the chat.
[15:08:01] * Xificurk left the chat.
[15:09:23] * tkoski joined the chat.
[15:26:43] * lastsky left the chat.
[15:32:08] * MattJ left the chat.
[15:32:09] * MattJ joined the chat.
[15:53:14] * MattJ left the chat.
[15:55:54] * mehdi61b joined the chat.
[16:08:14] * luca tagliaferri joined the chat.
[16:27:55] * tofu joined the chat.
[16:48:29] * lastsky joined the chat.
[17:09:54] * tofu left the chat.
[17:39:12] * tofu joined the chat.
[17:51:18] * mary@hut.peard.org joined the chat.
[17:52:33] * mary@hut.peard.org in now known as Mary.
[17:53:37] * Robert left the chat.
[17:56:13] * Mary left the chat.
[18:17:04] * Sidney San Martín joined the chat.
[18:17:05] * Sidney San Martín left the chat.
[18:20:36] * Sidney San Martín joined the chat.
[18:20:36] * Sidney San Martín left the chat.
[18:21:33] * Sidney San Martín joined the chat.
[18:22:27] <Sidney San Martín> Morning. I'm a Jabber/XMPP chat user. There seem to be major shortcomings in being signed in from multiple devices at once. I'm just curious, does this stem from limitations in the current protocols, or the server?
[18:23:17] <Sidney San Martín> Namely, I only get messages on one device when I'm signed in from more than one.
[18:26:52] * test1 joined the chat.
[18:26:52] * Sidney San Martín left the chat.
[18:27:31] <test1> gfdsgdfs
[18:27:37] <test1> gfds
[18:27:52] * test1 left the chat.
[18:28:06] * test1 joined the chat.
[18:29:22] <test1> Martin, use different resouce at login on different devices, and will go right...
[18:30:01] * test1 left the chat.
[18:31:23] * Sidney San Martín joined the chat.
[18:31:36] * Sidney San Martín left the chat.
[18:36:36] * Sidney San Martín joined the chat.
[18:43:47] * luca tagliaferri left the chat.
[18:50:48] * Sidney San Martín left the chat.
[18:52:03] * Sidney San Martín joined the chat.
[18:52:03] * Sidney San Martín left the chat.
[18:56:56] * Sidney San Martín joined the chat.
[19:24:38] * Sidney San Martín left the chat.
[19:24:50] * Sidney San Martín joined the chat.
[19:24:50] * Sidney San Martín left the chat.
[19:25:34] * Sidney San Martín joined the chat.
[19:25:39] * Sidney San Martín left the chat.
[19:26:59] * Sidney San Martín joined the chat.
[19:30:51] * mehdi61b left the chat.
[19:32:28] * Zash joined the chat.
[19:33:17] <Zash> Sidney San Martín: That's by design
[19:34:11] <Zash> Sidney San Martín: Messages are sent by default to the device with the highest priority.
[19:38:13] * mehdi61b joined the chat.
[19:40:56] <Sidney San Martín> Zash: I suppose that makes sense, but just barely. How is the server supposed to know which device I want to receive messages on at a given moment?
[19:43:40] <Zash> Sidney San Martín: By dynamicaly adjusting priority. Or by making them all have the same prio, then the first message would usualy be sent to all, and subsequent messages is sent to the one you reply from
[19:50:07] * tofu left the chat.
[19:51:55] <Sidney San Martín> "all subsequent messages"… when would it start sending them to all devices again? What if I switch?
[19:53:44] <Zash> The devices should try to indicate which one is the "most available" one
[19:54:54] <Zash> If you switch device, and send a message from it, replies would usualy be sent to that device
[19:56:42] <Zash> If you have to worry about it, your server or devices/clients are probably doing something they shouldn't
[20:04:29] <darkrain> Some servers can also be tweaked to send messages to all resources
[20:05:41] <Sidney San Martín> Hmm, OK
[20:06:03] <Sidney San Martín> Will they ever send "sent" messages to other resources?
[20:06:41] <Zash> "sent" messages?
[20:07:40] <Sidney San Martín> Re. darkrain, if a server is configured to send messages to all resources, can they also send your own side of the conversation?
[20:08:14] <darkrain> [xep message carbons]
[20:08:15] <Sidney San Martín> i.e. I'm logged into two resources, I send a message to you from one, my sent message is visible on the other
[20:08:20] <darkrain> !xep message carbons
[20:08:27] <darkrain> Kanchil: Awake?
[20:08:35] <darkrain> Sidney San Martín: http://xmpp.org/extensions/xep-0280.html
[20:08:50] <Zash> Or, messge archiving could be used
[20:09:12] <darkrain> I think they're slightly different, but not mutually exclusive
[20:09:23] <Sidney San Martín> Ah, interesting
[20:09:56] <Sidney San Martín> So it sounds like a lot of good stuff is possible, but not necessarily implemented/set up in a sane way in most places
[20:09:58] <Sidney San Martín> That sucks
[20:14:28] <Zash> Sidney San Martín: Implementations may suck, but that's not the fault of the protocol ;)
[20:20:40] <johnny> Zash, if all implementations suck, then yes. it might be because of the protocol..
[20:24:20] <Zash> johnny: True, but do they all?
[20:25:21] <johnny> even if most of them suck, it could still be becausee the the protocol
[20:25:35] <johnny> there will always be the person who gets the A while everybody else gets the D
[20:25:49] <johnny> who is willing to follow every insane rule
[20:31:30] * 111 joined the chat.
[20:31:48] <111> testing...
[20:32:19] * 111 left the chat.
[20:32:19] * 111 joined the chat.
[20:32:19] * Sidney San Martín left the chat.
[20:32:23] * 111 left the chat.
[20:35:29] * scippio_netbook joined the chat.
[20:37:08] * Sidney San Martín joined the chat.
[20:59:52] * mehdi61b left the chat.
[21:40:56] * johnny left the chat.
[21:41:04] * johnny joined the chat.
[21:52:33] * Sidney San Martín joined the chat.
[21:55:36] * scippio_netbook joined the chat.
[21:55:48] * dwd joined the chat.
[21:57:30] * dwd left the chat.
[21:57:32] * lastsky joined the chat.
[21:57:36] * dwd joined the chat.
[21:58:52] * dwd left the chat.
[22:00:08] * elmex joined the chat.
[22:00:14] * Treebilou joined the chat.
[22:00:28] * mlundblad joined the chat.
[22:00:28] * alfeberlin joined the chat.
[22:08:37] * duck1123 joined the chat.
[22:13:35] * MattJ joined the chat.
[22:17:18] * darkrain joined the chat.
[22:21:33] * johnny joined the chat.
[22:24:47] * teo1 joined the chat.
[22:41:57] * niekie joined the chat.
[22:45:13] * mlundblad left the chat.
[22:53:06] * niekie left the chat.
[23:03:49] * Sidney San Martín left the chat.
[23:17:01] * tkoski joined the chat.
[23:22:48] * McKael joined the chat.
[23:32:28] * Lance joined the chat.
[23:36:02] * Florob joined the chat.
[23:42:10] * tkoski left the chat.