19:22:47 RRSAgent has joined #webapps 19:22:47 logging to http://www.w3.org/2009/03/04-webapps-irc 19:22:55 evening all 19:23:29 Zakim has joined #webapps 19:24:08 chaals has changed the topic to: 1930Z DOM3Events - Web Applications WG (logged at http://krijnhoetmer.nl/irc-logs/) 19:25:49 zakim, this will be dom 19:25:49 ok, chaals; I see IA_WebAPI(DOM3)2:30PM scheduled to start in 5 minutes 19:29:58 IA_WebAPI(DOM3)2:30PM has now started 19:30:05 +Carmelo 19:30:33 +[IPcaller] 19:30:48 zakim, ipcaller is me 19:30:48 +chaals; got it 19:31:16 zakim, start meeting 19:31:16 I don't understand 'start meeting', chaals 19:31:30 trackbot, start telcon 19:31:32 RRSAgent, make logs public 19:31:34 Zakim, this will be WAPP 19:31:35 Meeting: Web Applications Working Group Teleconference 19:31:35 Date: 04 March 2009 19:31:39 I do not see a conference matching that name scheduled within the next hour, trackbot 19:31:52 Zakim, this will be DOM3 19:31:59 Zakim, call shepazu 19:31:59 ok, shepazu, I see IA_WebAPI(DOM3)2:30PM already started 19:32:09 ok, shepazu; the call is being made 19:32:13 +Shepazu 19:32:28 Zakim, who is here? 19:32:32 On the phone I see Carmelo, chaals, Shepazu 19:32:42 On IRC I see RRSAgent, shepazu, Marcos, tlr, aroben, heycam, smaug, ArtB, Lachy, gsnedders, gavin, hober, Hixie, phenny, trackbot, maxf, krijnh, chaals, inimino, Dashiva, hendry, 19:32:47 ... timelyx 19:33:40 agenda+ http://www.w3.org/2008/webapps/track/actions/1 19:33:44 agenda+ http://www.w3.org/2008/webapps/track/actions/6 19:33:58 +??P13 19:34:03 agenda+ http://www.w3.org/2008/webapps/track/actions/312 19:34:11 Zakim, ??P13 is me 19:34:13 +smaug; got it 19:34:53 agenda+ http://www.w3.org/2008/webapps/track/issues/39 and http://www.w3.org/2008/webapps/track/actions/18 19:35:34 agenda+ Publishing timeline? 19:36:04 Carmelo has joined #webapps 19:36:53 scribe: chaals 19:37:00 Scribe: chaals 19:37:04 Chair: Chaals 19:37:20 Present: Chaals, Shepazu, Smaug, Carmelo 19:37:39 zakim, take up next agendum 19:37:39 agendum 1. "http://www.w3.org/2008/webapps/track/actions/1" taken up [from chaals] 19:38:10 Topic: Action item review 19:39:00 Find All Open Issues For DOM3 Events and Update the Specification 19:39:13 DS: Not yet done. Would be good to start. Hold over until next week. 19:39:23 zakim, take up next agendum 19:39:23 agendum 2. "http://www.w3.org/2008/webapps/track/actions/6" taken up [from chaals] 19:39:42 Add a diagram about wheel axes etc 19:40:01 DS: Not so important. Does look more fun though... 19:40:05 ... in two weeks 19:40:30 ... Is not completely silly because we can use it as a reference for people to look at. 19:40:41 CMN: Agree. Illustrations and simple examples are actually important. 19:40:51 zakim, take up next agendum 19:40:51 agendum 3. "http://www.w3.org/2008/webapps/track/actions/312" taken up [from chaals] 19:41:24 agenda+ mailing list? 19:41:25 Propose some wording to resolve ISSUE-44 by saying that it depends on the language where the event goes 19:41:45 DS: Have a local copy of this. Says what we said last week. 19:42:04 CM: Couldn't find you last week 19:42:11 CMN: Odd, we were here... 19:43:09 DS: Could send that quickly. 19:43:31 ... basically it says what we said last week. 19:45:07 http://www.w3.org/2008/webapps/track/actions/16 and http://www.w3.org/2008/webapps/track/actions/17 - Carmelo to look at latest draft, and start writing tests... 19:45:54 CMN: Suggest we hold off action 17 until we have a latest draft - depends on upcoming agenda item... 19:46:11 ... have you started writing tests? 19:46:15 CM: Yeah, and submitted some. 19:46:22 DS: Haven't looked at them yet though. 19:46:38 CMN: let's change the action item to "write more tests!" 19:47:03 DS: ISSUE-28 doesn't seem like it should be on DOM3 Events. Would expect it to be on CORS 19:47:23 RESOLUTION: Yep, let's move ISSUE-28 to CORS 19:47:36 DS: Have a suspicion that not all the issues we have are listed in tracker. 19:48:34 DS: http://www.w3.org/2008/webapps/track/issues/27 ISSUE-27 also seems to belong to CORS 19:48:45 RESOLUTION: Move ISSUE-27 to CORS 19:50:03 DS: Is there anything we can do in the spec that willhave an effect on Firefox? 19:50:10 OP: I won't change anything this late 19:50:20 CMN: We won't change Opera 10 this late either. 19:50:39 DS: So looks like Keyboard stuff won't get sorted until next releases 19:50:46 OP: Indeed... 19:51:11 DS: Tempted to just change the Keyboard model completely to avoid breaking web pages (as opposed to just tweaking it a bit) 19:51:46 ... problem with documenting existing behaviour (which is important) is that when all behaviour is idiosyncratic, ... 19:52:15 ... If Firefox changes things that they thought was more sensible, they break pages. But I guess people know that Opera and Firefox are different, and so 19:52:23 ... I guess this is applicable to keyboard 19:52:30 [scribe: This is really what he said] 19:53:01 ... so to work across browsers I would do something not much better than sniffing to see what capabilities are in the browser and if it didn't have them I would switch to do different behaviours on different things. 19:54:01 ... with the load event I punted to the individual language. Fortunately it is clearly defined in SVG and has been for a while. For HTML I am happy tohave Hixie solve the problem but I am not sure it is a solvable problem. No matter what you do you will break content. I would like to have language that says :this is what you should do. But languages may define other things" 19:54:14 CMN: That is what I thought you were going to do. 19:54:27 DS: No, I just left it up to individual languages. 19:54:46 CMN: I would be pleased if you did define some default behaviour, then noted things may be different. 19:55:48 zakim, take up next agendum 19:55:48 agendum 4. "http://www.w3.org/2008/webapps/track/issues/39 and http://www.w3.org/2008/webapps/track/actions/18" taken up [from chaals] 19:55:58 Topic: Mutation events 19:56:17 DS: Anyone ahve an opinion about the previous topic? 19:56:36 OP: I would like to define our behaviour, of course :) 19:58:40 DS: I wonder if someone could be smart, and if people are sniffing on browser version they get something that says "if you don't have method X, do some things that are unrelated to X" (e.g. people test on filters to see if things are IE, then do something else). 20:00:04 CMN: Yes, but not always. 20:01:20 CMN: I understood Doug would say it is possible for languages to define what they want for load, and would note what HTML does, but say langauges should generally, and User Agents should expect generally something like SVG. 20:01:44 ... which I believe would effectively be defining what you are already doing... 20:02:00 OP: Does SVG define Load or SVGLoad? 20:02:28 DS: Defines the SVGLoad event, but then says "oh, you can also use the load event, with the same behaviour" 20:02:42 OP: Are there SVG1.1 errata? 20:02:49 ... I think it was changed a bit... 20:03:14 ACTION: schepers to research what SVG defines for load. 20:03:14 Created ACTION-318 - Research what SVG defines for load. [on Doug Schepers - due 2009-03-11]. 20:04:09 OP: 1.1 doesn't say much about load event. Mentions SVGload, triggered when some element has been fully parsed. This is quite different. 20:04:26 ... but SVG has the same attribute name (different event but same attribute). 20:04:39 DS: Believe we defined it more clearly in SVG 1.2 but not convinced that we got it right. 20:05:31 OP: SVGload isn't related to the document, but to elements 20:06:02 DS: Wouldn't necessarily define it the way SVG defines it. Would define its default in a way that makes sense, and talk about the differences that exist between HTML and SVG 20:07:10 ... SVG tiny 1.2 talks more about progress load event. 20:07:21 http://www.w3.org/TR/SVGTiny12/interact.html#LoadEvent 20:08:05 DS: SVGload is deprecated and only for backwards compatibility. For load, the event is triggered when the element and dependent resources are loaded (and scripts are interpreted)... 20:09:37 RESOLUTION: Doug, STFU! 20:09:59 s/RESOLUTION: Doug, STFU!// 20:10:21 DS: Seems like SVG isn't the right model to copy either 20:10:34 RESOLUTION: Extend ACTION-312 by a week. 20:10:50 Topic: Mailing list 20:11:15 DS: Going through the mailing list to find stuff to do with D3E is really tedious and onerous 20:12:00 ... we constantly talk about events, but very often it is not relevant to DOM 3 Events - keywords come up in other contexts too. Was wondering if it would be appropriate to make a new mailing list. 20:12:12 CM: Probably a good idea 20:12:19 OP: There is an old DOM mailing list 20:12:26 CM: I have had the same issues 20:12:36 CMN: Is old DOM mailing list public? 20:12:49 CM: Think so... 20:13:23 http://lists.w3.org/Archives/Public/www-dom/ 20:13:47 OP: There is a problem that people only check webapps list 20:14:24 CMN: So what if we move to www-dom, and send a reminder every month to webapps that DOM 3 stuff is on www-dom? (for developers who don't look at other lists much) 20:14:50 ... and cc minutes of DOM 3 stuff (with another reminder) 20:15:57 DS: how about Bcc minutes to public-webapps 20:16:13 CMN: Better to send to webapps, with ReplyTo to www-dom 20:17:05 DS: We could try that for now. I imagine we will hear complaining... There is some traffic already ... 20:18:13 Proposed RESOLUTION: We move to www-dom, and chaals will send reminders every month, plus cc minutes to public-webapps with reply-to set for www-dom 20:18:36 s/minutes/minutes and meeting notice/ 20:19:10 RESOLUTION: We move to www-dom, and chaals will send reminders every month, plus cc minutes and meeting notices to public-webapps with reply-to set for www-dom 20:20:15 Topic: Mutation events and http://www.w3.org/2008/webapps/track/issues/39 and http://www.w3.org/2008/webapps/track/actions/18 20:21:17 DS: Didn;t get a chance to do as much as I would like. There are two different camps... 20:22:22 tlr has joined #webapps 20:22:46 ... The people who say mutation events are awesome, love them, have implemented them in JS or something... 20:23:09 CMN: I believe that the ARIA folks, or at least Rich Schwerdtfeger, is in that camp 20:23:17 DS: ...and then there are the browser implementors 20:24:03 OP: Was talking with our accessibility guy today about this and seems that they only need the DOMattrmodified event. Which is not one of the evil ones. 20:24:21 ... don't understand the reason for that. But ... 20:24:55 ... wonder if they could ahve ARIAattrmodified or something. There is some accessibility software that will change attributes on the page and then check what has changed. 20:25:03 ... they are using domattrmodified for that. 20:25:16 DS: I think they also need to know when the text changed as well 20:25:29 ... after a complex change of text. 20:26:06 OP: Why would that be? 20:26:19 CMN: Because assistive technologies need to know when things have changed. 20:26:56 OP: This is not about implementation, it is about what web developers can use. The internal implementation can use something else. 20:27:28 DS: Maybe we should comment during ARIA last call that they are relying on a relatively costly mutation event and ask if they can define something more targeted like ARIAattrmodified 20:27:55 ... suspect they will not be comfortable doing that, but we could work with them to define that. 20:28:00 OP: yeah... 20:28:13 DS: That would be relatively trivial, and much lower cost than the generic one. 20:28:26 CMN: At the very least, it opens a dialogue that we probably need to have. 20:29:00 DS: Mozilla has stated that they will not implement [I forget which] 20:30:13 OP: Domnoderemoved, probably. I know of no reason not to implement it, given that we implemented another one, although it could be better to remove both. I am worried about back compatibility because some web pages do use these things. 20:30:22 DS: They only use the ones that are implemented. 20:30:56 OP: Only gecko implemented DOMattrmodified event, doesn't implement DOMnoderemovedfrom/insertedtoDocument 20:31:07 DS: Think we should open topic of removing or reworking them. 20:31:26 ... there are script librarians who say they need them, but they cannot be relying on them because they are not implemented. 20:31:37 DS: First step is to find out what is implemented. 20:32:08 OP: When Jonas was proposing removing/changing things, it was only evil things. 20:33:24 DS: If nobody is going to implement something, I am not going to put it in the spec. What we could do is make the mutation events spec, and script libraries could conform to them - a copy/paste of the mutation events from DOM3 events. D3E could go with a cleaner model, then people would not have to rely on script libraries. Nobody is relying on any of this working in browsers, where it oesn't. 20:33:53 OP: Could document what I found implemented... 20:34:09 ... would be great if someone from Opera could comment on this. 20:34:40 ACTION: Olli to publish what he has found about mutation event implementation 20:34:40 Created ACTION-319 - Publish what he has found about mutation event implementation [on Olli Pettay - due 2009-03-11]. 20:35:05 ACTION: Chaals to get someone at Opera who understands event implementation to comment on this issue 20:35:05 Created ACTION-320 - Get someone at Opera who understands event implementation to comment on this issue [on Charles McCathieNevile - due 2009-03-11]. 20:35:46 Topic: Publishing Schedule 20:36:44 CMN: Latest formal spec is 2007. We should probably publish something soon. 20:38:03 DS: Can have a new draft proposal in two weeks 20:38:31 ACTION: schepers to provide new draft ready to publish as Official draft on /TR, due in 2 weeks 20:38:31 Created ACTION-321 - Provide new draft ready to publish as Official draft on /TR, due in 2 weeks [on Doug Schepers - due 2009-03-11]. 20:39:54 ACTION: chaals to notify public-webapps that discussion has shifted to www-dom 20:39:54 Created ACTION-322 - Notify public-webapps that discussion has shifted to www-dom [on Charles McCathieNevile - due 2009-03-11]. 20:40:52 CMN: Would be useful to give a couple of days before discussing things on www-dom. (But I will send minutes there already) 20:41:45 -smaug 20:41:52 oops 20:41:54 Meeting Adjourned 20:41:57 -Carmelo 20:42:26 rrsagent, make log public 20:42:36 rrsagent, draft minutes 20:42:36 I have made the request to generate http://www.w3.org/2009/03/04-webapps-minutes.html chaals 20:48:35 zakim, bye 20:48:35 leaving. As of this point the attendees were Carmelo, chaals, Shepazu, smaug 20:48:35 Zakim has left #webapps 20:48:48 rrsaagent, bye 20:49:34 shepazu, as far as I can tell the only impact of workers on accessibility is that things happen in a background thread - you still get the change. 20:50:02 shepazu: how do I add related mails to an issue? 20:50:15 so the only question is whether there is an event to say that the thread has finished. 20:50:35 smaug, you name it - ISSUE-XXX (where XXX is some number) 20:50:51 oh, that's if you hadn't sent it yet 20:51:01 yeah, old mail 20:51:09 sent 5 months ago 20:54:13 tlr has joined #webapps 20:56:27 hober has joined #webapps 20:57:04 yeah, adding it to notes is about the best 22:38:02 tlr has joined #webapps 22:40:25 heycam has joined #webapps 22:49:31 annevk has joined #webapps 23:48:20 annevk has left #webapps