Chair: Jon Gunderson
Date: Thursday, Feburary 10th
Time: 2:00 pm to 3:30 pm Eastern Standard Time, USA
Call-in: Longfellow Bridge (+1) (617) 252-1038
Chair: Jon Gunderson
Scribe: Jim Allan
Gregory J. Rosmaita
see UA site for additional information, and DOM survey results. What supports to AT vendors require. http://www.w3.org/WAI/UA/ want to address concerns, requirements, what can this group do to help.
JG: not focus on DOM 3 in this meeting. address at a different time.
mq: review of dom 3 might be useful for future reference.
also regularly scheduled telecons. publish recommendation on March 8.
dp: judy did a wonderful job discussing implications and demonstrating. Misunderstanding of what quick tips cards are for (is that all there is to do).
Action JG: message Mark H. of productivity works about survey and participation on DOM conference call.
Judy still in negotiations with Host.
jg: must be at proposed rec by Mar 10, must announce 8 weeks ahead of time. may have telephone call in.
jg: objected to technique (5.5), techniques was not discussed in group and added new directions and concerns about checkpoints. Proposal, no formal method of review. submit technique to list, if no objections, then Ian edits into tech document, if objections then add to issue list. Developers are now asked to look a document and implement them, must have technically sound techniques, don't want to confuse issue. must have clear understanding of what we are trying to communicate.
da: any technique under a cp must relate to technique
jg: must be clear so developer doesn't ask "what are you asking for??"
jg: reviews proposal again...
jg: techniques is published as a note. can be updated at any time.
gr: techniques must be usable and understandable.
gr: authoring tool group, rechartering as interest group to update techniques, work with tool makers. working groups must evolve how they work with techniques documents to keep up with changes.
jg: current tech have recommendation that need updating. should publish new tech within 6 months
jg: read only access to dom, write access to some attributes, current requirement would be difficult for Opera, want more specifics. JG sent proposal to list. some comments. may need PGWG involvement.
5.1 Provide programmatic read and write access to content by conforming to W3C Document Object Model (DOM) specifications and exporting interfaces defined by those specifications. [Priority 1]
JG proposed 5 seperate specific checkpoints
5.1 Provide programmatic read access to all content using the interfaces specified in the W3C Document Object Model (DOM level 1) and exporting those interfaces to other applications . [Priority 1]
5.2 Provide programmatic write access to the value attribute of form controls using the interfaces specified in the W3C Document Object Model (DOM level 1) and exporting those interfaces to other applications . [Priority 1]
5.3 Provide progammatic support for the event model and methods for form controls and links using the interfaces specified in the W3C Document Object Model (DOM level 2 CR) and exporting those interfaces to other applications . [Priority 2]
5.4 Provide programmatic write access to all element of the W3C Document Object Model (DOM level 1) and exporting those interfaces to other applications . [Priority 3]
5.5 Provide progammatic support for the event model and methods for all elements using the interfaces specified in the W3C Document Object Model (DOM level 2 CR) and exporting those interfaces to other applications . [Priority 3]
some comments from CMN, GR, RS combine 5.5 and 5.3 change priority to 1 comments from Mark Novak about event model. need to get Al Gillman and Daniel Dardelier to review this. JG invited PF folks belatedly, would like to invite them to another meeting
GR: at PF face 2 face meeting, Arnaud was receptive to our concerns, wants a draft of wish list for dom 3, coordinate UA and PF concerns. PF is closed group, public disucssion difficult.
GR: pf not happy with Dom 2, growing pains, want dom 2 out, to work on dom 3 to address large holes. keyboard event model is absent, people may programs for pointing device rather than device independence.
Action JG: invite PF Al, Daniel, CMN and Mark Novak to next meeting Feb 23
Action GR: to wake up CMN, techniques coming soon.
da: what about script events, assemble a jigsaw puzzle, using keyboard, how to make mouse work
jg: see 1.1, if no dom 2 support, then use msaa or other technology to provide access. made 5 checkpoints for more specificity.
should we talk about dom 2. not discussed in Last call. are we changing document so much that we must go back to last call.
da: I don't think so.
jg: smil presentation, periodically a link appears,
da: link appears intermittently, must click link or pause button before it disappears, same time either way. must be able to change speed of presentation.
dp: must know link is there first, then pause
jg: this is not presentation speed, in smil can say this link appears for this link.
jg: have a system that pauses at a link and does not go on until the user responds.
da: must be user configurable.
mq: not a pause under user control,
action DA: rewrite technique for 2.2 (see list)
mq: select box, in navigating list, the site changes, does pause apply here
jg: poor authoring, script controls the timing.
gr: worked on techniques for that.
jg: this is not part of pausing.
mq: if we pause part of the screen, and other parts of screen continue changing, is this part of pause.
jg: no, different from specific markup, what you are talking about is related to scripts. what you describe should be coded to transform gracefully, if scripts are off.
jg: computer processing video, it is not rendered.
dp: allow reading of contigeous materials
da: works for me, if distracted by video then hide it,
mq: is there a time when you know what the link is audio or video or smil
jg: yes, link could call a media player, or some object.
dp: in IE can turn all of those, what about embedded objects
jg: what about java script,
da: how to control
ja: if not render video are captions included with video?
jg: authors of controls or native control of rendering of video
dp: people with vi want to know whats there and whats on the screen but inform the user,
da: should have content,transcript, longdesc available for media content
jg: should get an indication that something is playing
jg: use css to hide video
mq: in webspeak play audio but don't display video
dp: this is a user option to hide information, need some information about when media has stopped, etc.
dp: css hidden, does not remove processor/bandwidth load.
da: not an accessibility issue so much as computer issuer.
jg: this is p1, css hidden might be a useful technique. if you procees the video but not render it is ok.
Action: ja use css display=none to techniques, or don't call up video player, or hide in some other fashion
da: step through not part of original intent.
jg: review techniques, stepping would work for video but not audio. this is address in 4.6 this would not satisfy 4.5.
da: slowing rate is different from stepping through.
Action: jg-add reference in 4.5 to 4.6 in regard to stepping throught animation.
jg: 4.6 start stop etc are discrete functions, 4.5 is changing rate. discuss giving feedback to those who raised issues.
Action: jg check with Ian about feedback to issues raiser.
dp: not the intenet of the checkpoint. not sure if it available
mq: can return to place in a stream, function in g2 go to a clip and return to a point where you left it, if it is not a live stream.
da: could be difficult for easily distract folks
jg: maintain history list
gr: analogous to restarting a download.
mq: does this apply to mm?
jg: gr what do you think?
gr: covered by cp that says retain user focus. we are not specific as to type of content. useful to extend to any media. user configuration is the key to this. may be time I need to return exactly where I left off.
mq: not an accessibility issue
gr: general usability issue
jg: some run in the background,
dp: implementation of streaming content would be impossible
gr: right not streaming (live), that is comming from a file.
da: what if have frames, video frame and question frame. if focus controls the video then video stops when answereing questions.
dp: frames are usally not part of history, this is a configurable situation.
jg: origionally this was an orientation issue, restore orientation. don't know of similar function in media player, describes scenario,
ja: digital talking books can bookmark audio stream
dp: not mechanism
da: media player..
jg: player called in response to an event, can you browse media sources using
gr: has a word processor history mechanism of last x number of sites visited.
jg: consensus? discuss on list. review history mechanism, can you return to place in non-live stream when returning to a stream. does the player do this or need to do this. is this a P1 in that situation.