[10:48:00] --- Dave Raggett has joined
[11:01:05] <Dave Raggett> Looks like things are running slightly late
[11:03:49] --- mmealling has joined
[11:04:18] <Dave Raggett> do you know if Dean's slides are on the web anywhere?
[11:04:39] <mmealling> that's atypical for an IETF meeting... usually a few have at least logged into the MUC....
[11:04:51] <mmealling> not sure... I haven't been paying attention to the WG...
[11:05:13] <Dave Raggett> I am remote, and the sound is largely dropping out
[11:05:19] <mmealling> I just follow the APPS area WGs during the meetings. More nostalgia for me, really.
[11:05:30] <mmealling> so am I... my last IETF meeting was 2 years aog
[11:05:58] <Dave Raggett> I was a chair some years back (HTTP WG), but haven't been involved much recently either
[11:07:31] <Dave Raggett> the sound in this meeting room worked great this time yesterday, but now it is terrible
[11:07:34] <mmealling> I made a conscious decision to get out of the industry a few years ago... but I keep dabbling in it out of habbit. ;-)
[11:08:32] <mmealling> woah... the xcon MUC is empty?
[11:08:38] <mmealling> I wonder if there are local network problems.
[11:08:54] <Dave Raggett> sounds like it
[11:09:11] <Dave Raggett> only getting the odd second here and there before it drops out
[11:10:42] <Dave Raggett> FWIW my slides for the meeting are at http://www.w3.org/2005/11/widex-dsr-20051108.html (I was unable to attend in person)
[11:14:37] <mmealling> hmm.... is there a tutorial on what Widex is?
[11:14:55] <Dave Raggett> Vlad is giving one now (his audio is now solid)
[11:14:58] <mmealling> I can't tell if its a standard version of ajax or something else....
[11:15:07] <mmealling> ahh.... no audio here.....
[11:15:10] <Dave Raggett> Audio http://videolab.uoregon.edu/events/ietf/ietf645.m3u
[11:15:28] --- RjS has joined
[11:16:02] <RjS> ok - took a minute to join
[11:16:12] <RjS> I'm in the room - if someone has a question, I can relay it to the mike
[11:16:12] <Dave Raggett> The idea is to use XML to describe the UI and to have the application logic running somewhere else and connected via a simple messaging protocol that passes UI events and updates
[11:17:03] <mmealling> ahh..... like what inkscape is currently using XMPP for.... except the events are for UI instead of document change events.
[11:17:36] <Dave Raggett> I'm not familiar with inkscape, but it sounds similar
[11:17:52] <mmealling> is it a single point to point or can the events come from multiple origins?
[11:18:27] <Dave Raggett> we want to serialize DOM events as XML and allow for application specific events by making the protocol independent of what kinds of events there are.
[11:18:28] <mmealling> inkscape is an SVG editor. So you can use an XMPP MUC to send around what amount to SAX events that modify the shared SVG document model.
[11:19:12] <mmealling> cool.... Its interesting that you're doing this in the IETF and not W3C....
[11:19:16] <Dave Raggett> RjS, the audio keeps dropping out and the last comment was very quiet
[11:19:45] <Dave Raggett> The IETF is a good place for work on protocols.
[11:20:20] <RjS> that was ted suggesting that v4 and v6 be treated explicitly during exploring nat traversal requirements
[11:20:22] <mmealling> so if I had a UI that understood DOM events could I specify the source of those events to be an XMPP MUC like this so I could use that as a collaborative desktop?
[11:20:38] <RjS> is the audio from the presenter understandable?
[11:20:46] <mmealling> it is for me....
[11:20:58] <Dave Raggett> yes, most of the time Vlad's speech is clear
[11:25:08] <RjS> that's dean - is he coming through?
[11:25:21] <Dave Raggett> not really
[11:25:23] <RjS> ok
[11:25:38] <RjS> he was saying that the line-mike wasn't making it to the stream :)
[11:25:48] <Dave Raggett> he's right!
[11:25:50] <RjS> (from the chair mike which also isn't making it to the stream it appears)
[11:26:47] --- eburger has joined
[11:27:30] <RjS> dean: security guys always beat me up on what 'secure sessions" mean?
[11:27:34] <mmealling> I hear: what do you mean by "Secure sessions" since that makes security people upset
[11:29:09] <RjS> dean: raise you hand if you've read this draft
[11:29:15] --- stephenfarr has joined
[11:29:19] --- hardie@qualcomm.com has joined
[11:29:24] * Dave Raggett raises his virtual hand
[11:29:53] <RjS> dean: opinions about adopting this draft as WG item?
[11:30:07] --- stephenfarr has left
[11:30:13] <Dave Raggett> (thanks, audio now inaudible)
[11:30:21] <RjS> (There are now at least two other people in the room on (Eric,Ted))
[11:31:33] <RjS> Eric - I don't think I followed your comment, could you reflect it here?
[11:32:16] --- stephenfarr has joined
[11:32:49] <RjS> (essence was (i think) that Eric didn't see this being requirements for a generic remote DOM manipulation protocol, but there is some demand for such a thing)
[11:33:00] --- Randy Gellens has joined
[11:33:27] <hardie@qualcomm.com> Question is whether this is a general or a remote user interface synch specific approach
[11:33:40] <RjS> Eric: thinks we should work to solve the general problem
[11:34:57] <Dave Raggett> (the audio worked really well this time yesterday, oh well)
[11:35:21] <RjS> Eric: If this makes all the MMI people happy, then it is probably general enough
[11:36:44] --- stephenfarr has left
[11:36:57] <RjS> jerry(?) : Is this synchronized with SyncML/OMA work?
[11:37:29] <RjS> Dean: They are aware of what widex is doing.
[11:37:30] <Dave Raggett> (SyncML designed for syncing calendars etc. not for real time UI)
[11:38:06] <Dave Raggett> (OMA and W3C coordinating on MMI architecture)
[11:38:34] <RjS> Eric is taking your comment to the mike dave.
[11:38:40] <Dave Raggett> thx
[11:39:02] <eburger> You got it (RjS): the _requirements_ in the document would satisfy a bunch of needs for real-time DOM synchronization. The _language_ around the requirements make it look extremely focused on mobile user interfaces, only.
[11:39:13] <RjS> hum: Adopt this document as WG?
[11:40:02] <mmealling> is this a BOF or a WG?
[11:40:23] <Dave Raggett> 1st WG meeting (BoF was at Paris IETF)
[11:41:33] <mmealling> ok, I was wondering.... it seemed as though the ADs are trying to keep things close to the charter.....
[11:41:49] <RjS> hum was strong for adopt
[11:41:49] <RjS> ...
[11:41:49] <RjS> hum to adopt was strong
[11:42:06] <RjS> sorry - my network connection got strange
[11:42:43] --- hardie@qualcomm.com has left: Logged out
[11:43:55] <Dave Raggett> now on http://www.w3.org/2005/11/widex-dsr-20051108.html#[3]
[11:45:15] <RjS> now #[4]
[11:45:24] <Dave Raggett> thx
[11:48:51] <RjS> Dean: thinks there are two questions: (1) requirements around in-order delivery. (2) requirements around message loss/reliability
[11:49:07] <Dave Raggett> agreed
[11:49:11] <RjS> Dean: also need to discuss binding between widex session and a specific document
[11:49:30] <Dave Raggett> yes, the notion of a DOM document
[11:50:49] <RjS> Dean : focusing on text in last sentence of second paragraph of what's on [4]. Need to add clarity around indicating target node and target DOM document.
[11:51:05] <RjS> Dave - they wan't your slides to have larger text in the future :)
[11:51:22] <Dave Raggett> Use the > key for instantly larger fonts!
[11:51:41] <Dave Raggett> It's self scaling to match window width
[11:52:20] <RjS> not helping, but we'll work through it (It _is_ Dean's computer)
[11:52:30] <Dave Raggett> but I guess I should have split #[4] across 2 slides
[11:53:25] <RjS> Dean: Does application logic change depending on knowing what the presentation mechanism actually is (especially if they introduce different timing characteristics)?
[11:53:41] <RjS> (touch screen vs buttons on a controller)
[11:54:13] <RjS> ted: compare to games where order of multitaps makes a semantic difference
[11:54:41] <RjS> dean: do we need to be able to coordinate events between two different widex sessions?
[11:54:57] <mmealling> and where games are hacked with a vengence to figure out things like sending events in the wrong order in order to cheat.
[11:54:59] <Dave Raggett> (if networking latency is likely to be a problem then it makes sense to provide local feedback via client-side event handlers, not all the events need to be passed to the server)
[11:55:06] <RjS> (temporal coordination in particular)
[11:56:10] <Dave Raggett> very little of the sound is getting though (Vlad's better for some reason)
[11:56:20] <RjS> Dave -I'm probably a really lossy conduit
[11:56:27] <Dave Raggett> no worries!
[11:56:55] <RjS> Vlad has a lapel mike that's feeding to the stream. I suspect the other mikes are only making it to you through the speaker to Vlad's mike channel.
[11:57:46] <Dave Raggett> human ears and hearing are truly wonderful compared to current technology, sigh
[11:58:27] <RjS> dean: last sentence on [4] corresponds to the timing question he was trying to ask
[11:59:08] <Dave Raggett> thx
[12:00:16] <RjS> dean: We're done with slides - start open-mike
[12:00:57] <mmealling> hehe..... I heard that hum from here!
[12:01:16] <RjS> did his hum question get through? (It was a humor-hum)
[12:01:42] <RjS> dean: calling for work volunteers - text contributors in particular
[12:01:53] --- hardie@qualcomm.com has joined
[12:01:55] <RjS> no hands going up
[12:02:13] <RjS> dean: who will engage with feedback?
[12:02:18] <Dave Raggett> I am willing to continue to help with editing and contributing comments etc.
[12:02:23] <RjS> (only a couple of hand raises)
[12:02:50] <RjS> dean: Adjourn?
[12:03:01] <RjS> meeting wraps up...
[12:03:06] --- eburger has left
[12:03:09] --- hardie@qualcomm.com has left
[12:03:21] <Dave Raggett> RJS thanks for helping out
[12:03:47] <RjS> sure - glad it was marginally useful
[12:06:10] <mmealling> quick meeting
[12:06:33] <Dave Raggett> yes, but it will take longer once there are solid specs to chew over.
[12:11:12] --- mmealling has left
[12:13:05] --- RjS has left
[12:18:38] --- xmlscott has joined
[12:33:39] --- xmlscott has left
[13:01:06] --- Randy Gellens has left: Logged out
[17:22:30] --- Dave Raggett has left