Chair: Jon Gunderson
Date: Thursday, 4 May
Time: 2:00 pm to 3:30 pm Eastern Standard Time, USA
Call-in: Longfellow Bridge (+1) (617) 252-1038
Chair: Jon Gunderson
Scribe: Ian Jacobs
Present:
Harvey Bingham
David Poehlman
Mickey Quenzer
Tim Lacy
Gregory Rosmaita
Madeleine Rothberg
Mark Novak
Jim Allan
Regrets:
Dick Brown
Absent:
Denis Anson
Al Gilman
Kitch Barnicle
Rich Schwerdtfeger
Eric Hansen
Charles McCathieNevile
Hans Riesebos
Regrets for 11 May: HB, GR, CMN
[1] http://lists.w3.org/Archives/Public/w3c-wai-ua/2000AprJun/0275.html
1.Joint UA/WC Telecon today from 4:00 to 5:00 PM EST USA on the Longfellow bridge +1-617-252-1038.
2.Extra UA call next Tuesday (9 May) at 1:30 pm ET, same phone, if necessary. We will also have our regular meeting 11 May.
http://cmos-eng.rehab.uiuc.edu/ua-issues/issues-linear.html#271
MR: I posted something to list.
/* IJ Notes that nothing on the list */
MR: Refer to AG's comments: http://lists.w3.org/Archives/Public/w3c-wai-ua/2000AprJun/0244.html
MR: I believe that the situations AG referred to are both important.
MR: I agree with AG's comments on "undue burden" (Point 1 in his email).
MR: For Point 2, I agree that being able to arrange components is a P1 requirement.
MR: Geoff Freed and I are not aware of languages that don't. allow repositioning, so "case 3" is not significant. Refer to minutes of 25 April call for "three cases". http://lists.w3.org/Archives/Public/w3c-wai-ua/2000AprJun/0196.html
MR: If there were languages that didn't allow repositioning, then I would still want the UA to offer this to the user.
DP: Applicability also kicks in in case three - the user agent should do that work when it recognizes the content as transcript, caption.
DP: Furthermore, authoring tools already let people do this.
MR: In a multitask situation (e.g., watching a video and answering questions), you may need to obscure some content in order to answer questions at the same time. DP: Can we use another word than "arbitrary"?
JG: We could just talk about what specifications allow you to do.
MR: In quicktime, you can adjust the position of the caption window within the general region.
TL: In Windows Media Player, I can only get the caption to appear at the bottom.
MR: I've seen SAMI demos where there is repositioning. Through scripting. It's not in the player interface itself, but in the feature set.
JA: SAMI uses style sheets for positioning. Resolved: - Checkpoint 4.7 should remain a P1. It is impossible for some users to access content unless they can control positioning.
IJ: Should 4.7 be rephrased in terms of what specs support?
Proposed: Add to the checkpoint that the user should have the same flexibility as the author. MR: Using the same positioning system of the author.
IJ: Or do we say more - if your spec doesn't including a positioning system, still provide something.
GR: I'd rather keep it more vague rather than more specific.
IJ: Is it P1 even if the spec does not provide for positioning?
GR: Strictly speaking, yes.
Who supports the broader scope? GR, MR, DP, JA, HB, MQ, TL TL:
GR: Users need also to fix bad authoring in general. Also, a case of third parties writing "skins" for content and in doing so, may eliminate the rendering of captions.
MR: We may need to change "configure" to "reposition" since it's not about static configuration.
Resolved:
- Change checkpoint 4.7 wording to be "change the position of..."
"Allow the user to position .... on graphical displays."
- Add a Note after 4.7 for cases where a spec provides for positioning, the
user is expected to have the same range of positions as the author. - In
techniques, talk about the three cases (refer to 25 April minute).
HB: Should we suggest that users be able to reuse dead space, e.g., below a television broadcast of a movie?
IJ: That sounds like a useful default.
Action MR: Confirm usage of "configure" in checkpoints to verify that it means "static choice" appropriately (e.g., 4.9).
MR: I propose a technique that configuration (in 4.7) should also be possible, in addition to dynamic respositioning.
IJ: I propose that, according to MR's results, that if most of the configure checkpoints should in fact be adjust + save configuration, that we change the definition to include that. Or, that we say "control and configure" when we want both. /* Jon leaves to get an award */
http://cmos-eng.rehab.uiuc.edu/ua-issues/issues-linear.html#233
Discussion postponed since Ian needs to complete his action item from the last call.
/* Ian discussion of publication of document tomorrow. */
/* TL drops out */
http://cmos-eng.rehab.uiuc.edu/ua-issues/issues-linear.html#257
5.1) Where should minimum requirements go?
MR: If normative, then in the guidelines.
IJ: I agree.
5.2) Do people prefer an appendix, or after each checkpoint?
IJ: Related to this - what's the status of the notes after each checkpoint? DP: I think developers would like them regrouped in one section. GR: Other option is after each checkpoint.
IJ: Including this information in the checklist is an option.
Resolved: - Include minimal requirement after each checkpoint in the guidelines. Identify clearly as normative. - Include minimal requirement after each checkpoint in the checklist.
5.3) Should we pursue this project for UAAG 1.0?
HB: No, I think this a three-month effort.
IJ: I also fear a time sink now, but that we will save much time later by resolving this now.
DP: Let's give it a shot and a deadline. If we haven't identified the minimal requirements within a certain time, we advance anyway.
GR: Even if we don't go out with min reqs identified, we should commit to producing them (in a timely manner).
MR: I agree with DP - give it a shot and a deadline (and GR). My concern with GR's proposal is that if you commit publicly to a new draft three months later, then people will ignore the first Recommendation.
JA: I agree with
MQ: I agree with DP.
MN: I agree more with HB. Should this be part of a FAQ. No matter what we do, the document will have holes.
DP: If we do release a set of min reqs, this could make people lazy.
5.4) Framework for review.
a) Checkpoints where the requirement is self-evident. For example (not
definitive list): 6.1, 11.1, 11.2, 11.3, 2.6, 4.6, 4.12, 2.15, 5.8, 7.2.
b) On/off checkpoints (where the requirement is self-evident).
c) Navigation: forward sequential navigation in general is min requirement.
GR: What about backward?
d) Checkpoints with a range of values:
d.i) Some of them can be aligned with the range offered by the OS.
GR: It might be useful to point out that some ATs only run under 256 colors.
So less might be sufficient to meet the user's needs. UAs should not change OS
settings, which might impact ATs.
d.ii) Up to the WG to decide (mildly arbitrary).
e) As AG pointed out, don't have a criterion, but an example that would be
sufficient.
http://lists.w3.org/Archives/Public/w3c-wai-ua/2000AprJun/0186.html
Action IJ: Propose a grouping to the list based on this framework.