Chair: Jon
Gunderson
Date: Wednesday, October 7th
Time: 12:00 noon to1:00 pm Eastern Standard Time
Call-in: (+1) 617/258-7910
12:00-12:30 SMIL
12:30-1:00 Dynamic HTML
Chair: Jon Gunderson
Scribe: Ian Jacobs
SMIL discussion chair: Jim Allan
Denis Anson
Judy Brewer
Daniel Dardailler
Markku T. Hakkinen
Kathy Hewitt
Philipp Hoschka
Marja-Riitta Koivunen
Scott Luebking
Chuck Opperman
Kitch Barnicle
SMIL: Will include general principles of SMIL proposal and Jim Allan will identify somebody to work on SMIL techniques
DMTL: Jon will draft a guideline related to DHTML
Next telecon on October 14th
1) SMIL (Chaired by Jim) See Judy's email [1] and thread. [1] http://lists.w3.org/Archives/Public/w3c-wai-ua/1998JulSep/0200.html
Jim: Should these be incorporated in UA Guidelines or a techniques-type document.
Judy: How does it fit into overall priorities? How does it fit into abstract guideline/concrete technique structure? In [1], there are both guidelines and techniques. Perhaps expand section 4.4 [2] with this new information? [2] http://www.w3.org/WAI/UA/WD-WAI-USERAGENT-19980814.html
Denis: General guidelines doc needs to describe features of Web pages to make them accessible. SMIL would be a technique since implementation (like HTML, for example). /* Some Issues */
Judy: In resources appendix, add Geoff's reference.
Standards for broadcast?
Judy to follow up on: Issue: Is there the appropriate technical mechanism in SMIL for descriptive video:
Phillip Hoschka: Probably can be done, but I'll have to look into this.
/* Phillip: See Geoff Freeds mock-up in G2 environment. */
/* Discussion of positioning captions */
Philipp: Its not the author that determines caption position, it's the user.
Jim: We should include in techniques that CSS or XSL will apply to any SMIL implementation.
Judy: Almost everything that we looked at here for UA has a consideration that will need to be addressed in the Authoring guidelines. How is caption position handled by SWITCH statements?
Philipp: SWITCH should not be used for positioning captions. Style should be used. User should have control over the positioning. Does SWITCH mean "Whether or not there are captions"?
Philipp (on switch): Allows you to include alternative media objects in your presentation (vary according to bandwidth, encoding, etc.). A selection mechanism for what will be displayed, depending on player, settings, etc. How does this differ from/resemble SELECT? The same (according to AL). RESOLUTION : Caption should be able to be positioned independently of video. Guidelines won't specify mechanism. How does this interact with spawning new windows?
Philipp: You don't necessarily need a new window.
Judy: Even if window is spawned, we need to send an alert.
/* New issue: dynamic control of pace of presentation */
Judy: Different paces necessary depending on needs. Need to be able to control pace of presentation dynamically (without changing pitch). Are there barriers within SMIL to doing this? Hard-sync is allowed, but not required by SMIL 1.0.
Mark: In our Audio books, we allow dynamic control by modifying the time scale. Not done through hard-sync.
Philipp: I'll think about whether hard-sync is really required.
/* New */
Judy: Is anything missing?
Jon: Seems like we've covered most of the bases.
/* Feeling that subject pretty much covered, but new info may arrive on list after this discussion */
Marja: What about 3D stuff?
Judy: I wouldn't know how to address this here.
Judy: How are we going to get this information into the UA Guidelines? (Abstractions -> guidelines Details -> techniques) Will first version of UA Guidelines include SMIL recommendations?
Judy: If possible, would offer a major benefit.
Jon: If guidelines done "right", could (theoretically) include SAMI information as well.
Chuck: I'm not concerned about SAMI as part of the guidelines. I worried about people's problems today.
ACTIONS: - Integrate Judy's list into UA Guidelines. - Choose people to help with techniques section: Jim Allan and Geoff Freed.
-----------
2) Dynamic HTML (See also last week's minutes [3]) [3] http://www.w3.org/WAI/UA/wai-ua-telecon-19980930.html
Jon: How will DOM assist in notifying users/technologies of "mutation" events (changes). How can user know that there's an event associate with an element (in the face of bubbling). What statement can we make about DHTML?
Chuck: See document published by MS about authoring DHTML accessibility. Part of DHTML Reference. (Avoid mouse-over, Use CSS whenever possible, etc.) [4] http://msdn.microsoft.com/developer/sdk/inetsdk/help/dhtml/accessibility/accessibility.htm
Chuck: most concerned with a reasonable alternative to on-mouse-down
events.
When authors do mouse-specific stuff, can the UA do anything?
What about keypress events?
How are those implemented?
Jim: See article (HTMLWG? WebReview?) on implementing keypress for DHTML.
(URL??)
When authors do mouse-specific things (intentionally ignoring
on-keypress), can the UA synthesize the functionality?
An accessibility aid running on top of a browser can enumerate which
elements have events. But this mechanism fails when bubbling introduced.
One option for Page Authors: Put explicit events on elements.
Denis: Do browsers have equivalent of the "system carrot"?
Chuck: No, they don't have an insertion point. However, when you focus on an element (via tabbing or mouse click), IE4 explicitly sets the system carrot to that position. Could you navigate with the system carrot to simulate mouse-down events? (Note: Some IE mailer is just IE browser with insertion point).
Proposed solution: Throw the browser into edit mode (but read-only) where you have the insertion point, but can't make changes. This is problematic for technical reasons. Chuck: Idea for identifying events: Through scripts, could enumerate objects with mouse events attached, and wrap them with "tabindex" (or possibly anchors) to existing HTML page.
/* Note use of mouse-over to reveal help text in frames */
Jon: Are there other HTML workarounds for help messages?
/* Note use of "title" on elements, notably anchors, since you can get a tool-tip */ How long can a tool-tip be? How about an alt-text?
( No limit imposed by HTML. May be limits in implementation.) Question about cognitive issue of having same mechanism for navigating links as for navigating events.
Scott: Last week, looked at pages with DHTML. Some scripts restructure pages on the fly. For a blind person, might be equivalent to moving furniture around without explaining why. The purpose of the change must be made clear, not just the change.
Chuck: Is there a conf call next week:
Jon: Yes. and the 28th of October.