14:01:19 RRSAgent has joined #pf 14:01:19 logging to http://www.w3.org/2013/08/19-pf-irc 14:01:21 RRSAgent, make logs member 14:01:23 Zakim, this will be WAI_PF 14:01:23 WAI_PF matches both WAI_PFWG(HTML A11Y)9:00AM and WAI_PFWG(ARIA)10:00AM 14:01:24 Meeting: Protocols and Formats Working Group Teleconference 14:01:24 Date: 19 August 2013 14:01:39 RRSAgent, make log public 14:01:44 chair: Rich 14:01:54 meeting: W3C WAI-PF ARIA Caucus 14:03:47 trackbot, start meeting 14:03:49 RRSAgent, make logs member 14:03:50 zakim, who's here? 14:03:50 apparently WAI_PF()12:00PM has ended, clown 14:03:51 Zakim, this will be WAI_PF 14:03:52 Meeting: Protocols and Formats Working Group Teleconference 14:03:52 Date: 19 August 2013 14:03:52 On IRC I see RRSAgent, clown, richardschwerdtfeger, jongund, janina, scott_gonzalez, Zakim, trackbot 14:03:52 ok, trackbot, I see WAI_PFWG(ARIA)10:00AM already started 14:04:22 zakim, who's here? 14:04:22 On the phone I see Jon_Gunderson, [GVoice], ??P5, Rich 14:04:22 agenda? 14:04:23 On IRC I see RRSAgent, clown, richardschwerdtfeger, jongund, janina, scott_gonzalez, Zakim, trackbot 14:04:33 jcraig has joined #pf 14:04:38 zakim, GVoice is Joseph_Scheuhammer 14:04:38 +Joseph_Scheuhammer; got it 14:04:45 zakim, I am Joseph_Scheuhammer 14:04:45 ok, clown, I now associate you with Joseph_Scheuhammer 14:05:14 zakim, who's here? 14:05:14 On the phone I see Jon_Gunderson, Joseph_Scheuhammer, ??P5, Rich 14:05:15 On IRC I see jcraig, RRSAgent, clown, richardschwerdtfeger, jongund, janina, scott_gonzalez, Zakim, trackbot 14:05:24 zakim, ??P5 is me 14:05:24 +janina; got it 14:05:26 RRSAgent, make log public 14:05:30 chair: Rich 14:05:42 Meeting: W3C WAI-PF ARIA Caucus 14:05:54 +James_Craig 14:07:38 http://lists.w3.org/Archives/Public/public-pfwg/2013Aug/0027.html 14:07:45 MichaelC has joined #pf 14:08:38 +Cooper 14:09:27 scribe: jcraig 14:09:58 TOPIC: next steps on @aria-describedat 14:10:23 RS: group thought taking an ID would be a good idea 14:10:26 q+ 14:10:32 q+ 14:11:07 ack j 14:11:36 ack me 14:11:38 RS: IDREF of a@ref, link@rel, iframe, etc. 14:12:17 MC: I think this goes against the idea of @decribedat 14:12:31 q+ 14:12:36 q? 14:12:51 ack r 14:13:03 q+ 14:13:24 MC: author confusion with @aria-describedby, additional processing for UA, etc. 14:14:31 q? 14:14:35 RS: One example use is what description has multiple translations, like source sets or media tracks. 14:15:33 q+ 14:15:42 q+ 14:15:56 MC: confusion is that one is the description contents, and the other is a link to the description. Authors may not understand that. 14:16:00 ack j 14:16:02 ack m 14:16:37 q? 14:17:01 q+ where is the email that outlines the rationale for the new syntax for describeat? 14:17:22 MC: "at" (as in describedat) indicates a pointer. If we want to repurpose this one, we should rename it too. possibly @aria-descreference? 14:17:24 q+ to propose we add both ¨describedat¨ with URI and ¨descriptionreference¨ with IDREF / href 14:17:35 aria-descref? 14:17:56 ack me 14:17:56 MichaelC, you wanted to propose we add both ¨describedat¨ with URI and ¨descriptionreference¨ with IDREF / href 14:18:25 q? 14:18:41 RS: consensus is that @longdesc is not enough. 14:18:53 q? 14:18:59 ack r 14:19:38 MC: @aria-describedat would still take a URI. If we need an IDREF to a linking element (e.g. link, iframe), that should be a different attr 14:19:44 q+ 14:20:10 Head of email thread (with 2 messages): 14:20:11 http://lists.w3.org/Archives/Public/public-pfwg/2013Aug/0003.html 14:21:02 MC: I am not arguing against the IDREF to linking element, but saying we still need the IDREF use case 14:21:19 q? 14:21:22 ack me 14:24:03 q+ 14:24:09 q+ for an example of how to use describedat with an iframe element 14:24:28 1+ 14:24:32 q+ 14:24:42 RS: Multiple descriptions in different languages, and potentially different media formats, different grade level? 14:24:56 q? 14:24:57 ack j 14:25:07 ack jan 14:25:47 JS: @aria-describedby is auto-spoken, where this new would be be an on-demand user navigation 14:26:36 ack jo 14:26:45 q- jos 14:28:36 RS: The IDREF of this new attr could be a element or elements representing a collection of resources representing the various forms (whether, language, format, etc) of descriptions. 14:29:49 q+ to request clarification from Janina on her comment about different grade level 14:29:58 q? 14:30:03 s/different grade level?/different grade level (Janina)?/ 14:30:05 q+ 14:30:26 q+ to say that the proposal for describedat was when we were looking for a solution around overloading describedby with a URI; so would like to keep the simple design pattern for that and use the indirection on another attribute 14:31:04 q+ to say if we don´t have a solution for a simple URI pointer, we force authors to add extra elements to the document that they don´t want to deal with 14:31:13 JG: If you allow describedby to take a URI, would that satisfy the requirement? 14:31:28 JS: not if it's auto-presented to the user 14:31:29 q+ to say which greatly increases error likelihood 14:31:42 JC: That could be a UA processing difference 14:32:02 JS: Concern that the processing difference would be overloading. 14:32:09 ack j 14:32:12 q? 14:32:13 ack james 14:32:19 ack jcra 14:32:19 jcraig, you wanted to request clarification from Janina on her comment about different grade level 14:32:21 q+ to say I think the indirection attribute is going to be the ¨advanced usage¨ one 14:32:37 http://www.w3.org/WAI/PF/aria-practices/Overview.html#Descriptions_external 14:32:48 q+ to request clarification from Janina on her comment about different grade level 14:34:06 Clown: example in authoring practices saying you can do that now 14:34:26 MC: You should remove that example from the authoring practices 14:35:57 ack me 14:35:57 MichaelC, you wanted to say that the proposal for describedat was when we were looking for a solution around overloading describedby with a URI; so would like to keep the simple 14:35:58 Clown: what about using a URI or Fragment-Identifier 14:36:00 ... design pattern for that and use the indirection on another attribute and to say if we don´t have a solution for a simple URI pointer, we force authors to add extra elements to 14:36:00 ... the document that they don´t want to deal with and to say which greatly increases error likelihood and to say I think the indirection attribute is going to be the ¨advanced 14:36:00 ... usage¨ one 14:37:14 MC: thought it was crazy to overload a single attr with both IDREF and URI 14:37:38 q+ to say frag-id is the not the same as URI 14:38:09 Call it DescHref ? 14:38:28 ack me 14:38:28 jcraig, you wanted to request clarification from Janina on her comment about different grade level and to say frag-id is the not the same as URI 14:38:47 describedby="foo" 14:39:04 describedat="#foo" 14:39:45 q+ to request clarification from Janina on her comment about different grade level 14:40:12 ack me 14:40:12 jcraig, you wanted to request clarification from Janina on her comment about different grade level 14:42:47 JS: DAISY/IDPF suggested this use case for content negotiation based on different reading level 14:43:24 JC: I agree with the need for that use case, but I don't think the content negotiation part of it should be ARIA 14:46:00 q+ to point out that the describeby eg. in the APG is handled by FF 14:47:31 MC: want to clarify with implementors: in-document IDREF processing can use a difference processing path that getting an element from a fragment identifier 14:48:30 RS: Consensus? 14:48:52 MC: Consensus for *proposal* of this approach 14:49:20 MC: Mapped out by Sept? 14:49:45 RS: A public editors draft would be sufficient 14:50:51 q+ to mention concern that we're rushing based on some arbitrary timeline "commitment" to outside groups. Are the "commitments" publicly documented? 14:51:05 ack jos 14:51:05 Joseph_Scheuhammer, you wanted to point out that the describeby eg. in the APG is handled by FF 14:54:05 JS: point is, that it's not abundantly clear that this use case is not already handled by browsers 14:54:36 JS: +1 for the URI (with potential frag id) idea 14:54:40 ack me 14:54:40 jcraig, you wanted to mention concern that we're rushing based on some arbitrary timeline "commitment" to outside groups. Are the "commitments" publicly documented? 14:56:21 q+ to say we should definitely publish with lots of ¨this is early days¨ disclaimers 14:59:48 https://dvcs.w3.org/hg/aria-unofficial/raw-file/tip/describedat.html 15:00:51