Chair: Jon Gunderson
Date: Tuesday, December 7th
Time: 12:00 noon to 1:30 pm Eastern Standard Time
Call-in: MIT Bridge (+1) 617-258-7910
Chair: Jon Gunderson
Scribe: Ian Jacobs
Gregory J. Rosmaita
Dick Brown (joined at 12:10)
1.IJ: Review techniques for topic 3.2
2.IJ: Publish new working draft of guidelines on 6 December
3.JG: Review techniques for Guideline 8.3 to 8.9
4.KB: Update impact matrix based on 5 November draft. Pending.
5.RS: Send last call document to IBM's Web Team in Austin.
Cancelled. We didn't receive comments.
6.MR: Review techniques for topic 3.1 (Multi-media)
7.MR: Review techniques for Guideline 3 (Multi-media)
Also: techniques for Guideline 4 (Multi-media)
8.DB: Review techniques for Guideline 5
9.DB: Ask IE Team about publication of review of IE 5 and Pri 1 checkpoints.
10.DB: Contact person in Windows media group to agree to review last call draft when available
11.MK: Write some comments on synchronization in multi-media to the list
12.EH: Propose new wording for Checkpoint 2.5
13.EH: Refine proposal on the meaning of "synchronized
alternatives" to the list.
IJ: See 6 December draft for editorial attempt a resolution.
MR: I would rather see something about "synchronized". Video of sign language may be synchronized.
1. RESOLVED: Telecon time change after the first of the year to meet on Thursdays at 1:00 EST (USA) for 90 minutes.
Action Jon: Please send to email@example.com.
2. GL WG is rechartering.
a) Last call issues. Refer to list and Jon's categorization
b) IJ: Important issues that you want to see addressed at ftf? (None spoken).
c) No objections to going to Proposed Rec after the holidays. GR: We'll avoid the end of year Y2K hype!
d) Other review comments.
GR: Earl brought to the fore the issue of 508 decision-making. Opposite opinion came up in ATAG review. In ATAG, concerns about application of Guidelines to purchasing requirements.
e) When building ftf agenda, think about which issues hang on interpretation of specs. Take those to PF.
Send ftf agenda items to the list.
f) It looks like we will have phone conferencing at the ftf.
Anyone here interested in calling in?
Of those present, Madeleine would.
Action Ian/Jon - publish an agenda by tomorrow morning.
Refer to Ian's proposal.
Please think of questions for the FAQ.
IJ: Refer to how addressed in 6 December draft to use only terms "captions" and "auditory descriptions"
2.7 Allow the user to specify that captions and auditory descriptions be rendered at the same time as the associated auditory and visual tracks.
MR: Problem with this wording is exclusion of other types of synchronized alternatives.
MR: We might need joint GL/UA meeting to consider EH's proposal
MR: People haven't done the text-to-speech synchronization yet.
Resolved: Need to coordinate meeting with GL WG.
Action Jon: Organize meeting with GL and AU (asking in particular Geoff Freed, Marja Koivunen, Gregg Vanderheiden, Wendy Chisholm, Eric Hansen). Propose 15 December joint meeting during UAGL call.
Action Madeleine: Send counter proposal to EH's to the list.
DP: Since we moved the definition of "native" to include OS features, does that mean we can remove it here?
DA: I think we meant in previous discussions that if a helper app is doing the rendering, it's not the UA's responsibility.
IJ: Do any UAs do this natively today?
GR: If a plug-in manufacturer is reading these guidelines, mentioning native support might help.
MR: Suppose a mainstream browser encounters a sound file. If the browser is calling a plug in, the UA should be configurable to not call the plug-in. Or is the plug-in monitoring content for sound?
IJ: I agree with MR. I think the UA can control on/off of rendering by choosing to "display" or not.
IJ: For the checkpoints related to volume (audio and speech), if you use OS controls, then control of volume through the OS.
MR: Also, background audio doesn't mention "native" either.
DP: There seem to be a lot of things that happen where the UA calls another. Is there a way to address these globally in the document?
IJ: Comments on "applicability" welcome.
DP: Gregory brought up interesting point about "autoplay". Many times an applet will let you control volume. Are we controlling through this the browser or the UA that content has been handed off to?
GR: The latter.
RESOLVED: (checkpoint numbers from ):
1) UA has the power to turn on/off rendering of audio. This applies to 3.2, 3.3, 3.4.
Action Ian: Add clarifying Note to rationale that UAs can turn off control of content even if it passes content off for rendering.
2) (Applies to 4.12 and 4.15). Remove native from 4.12 and clarify that control of volume is possible through OS if the UA using those controls.
IJ: Refer to related discussion on "native": 173.
MR: Seems harsh if the only solution is not accessible.
DA: But if you are providing a feature, you have to make it accessible.
JG: What if you're in X-Windows?
IJ: I don't UAs to reimplement, but if they are using a feature to provide a functionality, it's their responsibility. Maybe UA designers will ask for more accessibility in the OS.
JG: So if the Mac doesn't support the use of keyboard to move among controls in a dialog box whose problem is it?
IJ: I think the UA, but I realize that you want to use system controls.
GR: Take the case of HPR (native support for speech). The install is self-voicing. But if you don't have NN installed, speech stops before that point. Is that part of this discussion?
DA: There are third-party tools for the Mac that make it accessible. ("ClickIt" by Teletools (?))
DA: If the underlying OS is not accessible, your browser is not accessible.
DP: But you can make it accessible through your own UI.
/* DB Joins */
DB: I think we can live with with Ian's proposal.
Conclusion: More discussion on the list. Discuss this at ftf with IBM Team.
Resolved: * No objections to "fast forward" to 4.10 and 4.13.
MR: I thought "incremental" meant small discrete chunks. I don't think variable speed of fast forward and incremental advance are the same thing. Analogy: You want the page down key to move you in small chunks.
IJ: Does anyone feel that this is a checkpoint level requirement: advance advance/rewind in audio/video by N seconds (P2) where N is configurable (P3).
DP: I think P2 since without this, difficult for people with physical disabilities.
MR: I think this is more of a technique. It's also being done already.
Action Ian: Send proposal on this to the list.
Refers to 4.11 "Allow the user to control audio playback rate."
IJ: TV's comment is that speed control for general audio not as important as for speech.
GR: Hard for streaming. I think TV's point is that acceleration is just a convenience. Deceleration is different. More likely that people with learning disabilities could use slower audio. Also some users with blindness.
JG: It's possible to slow down streaming.
MR: Do we also require adjustment for pitch?
Action Denis: Write up text to explain why deceleration is important for users.
* 4.11 is Priority 1 (Refer to Denis' action item)
* Change language of 4.11 to refer only to deceleration of audio.
* Usually Helper App or OS will be doing the rendering.
MR: Note that 4.7 could have captions.
Adjourned 13:46 ET.