CMN who's going to Bristol?
Yes: Katie, William, Ian, Charles, Wendy, Andi
Maybe: Cynthia
GV We talk about marking language changes, are there any screen readers that do anything with that information?
JW Think IBM HomePage Reader either does or is about to.
ASW We do automatic language switching but not on lang tag. We have an algorithm per page but not inline.
WL Could that algorithm help us determine when to suggest a lang be used?
GV does it look at lang attribute?
Action ASW Send GV e-mail about IBM intended support of lang.
WC review
ASW formatting suggestion. The list should be ended after the 2 checkpoints and rest should be paragraph.
WC Planning on breaking out of unordered list into paragraphs.
GV Have all of these been tested?
WC I can not claim that every single one has been. However, I am trying to create live examples and screen shots. Refer to the latest CSS document.
WL Main problems seems to be not do they look good, but do they look somewhat the same?
KHS I was working on quicklinks on splash page to make it simpler to use. A man from the access board suggested that the reason people do not like using the W3C site is that it is too confusing. Maybe we should switch it to the left?
GV The W3C site, at least the WAI, is being modified.
CMN Probably not the world's greatest example on usability.
GV Make it easy to get to WAI site.
CMN Need to get to lots of info.
GV It should be very usable.
JW I've not had a problem with the site.
GV You are not an average user.
WL W3C current run on sentence of everything that they have separated by nothing is a nightmare.
KHS Wouldn't it be easier to have the links on the left?
WC It is being redesigned, we've got a lot of questions about it.
GV Getting us back on topic...wendy you want review?
WC There are 2 drafts.
WL With regards to tabindex.
WC How covered in UAAG?
IJ Nothing in guidelines. Discussion in Techniques - users ability to override user specified bindings. Info about direct access.
CS IE has support for both features. I created an example page.
WC I borrowed it and modified. /* explains the issue */ @@link to accesskey examples.
Action CMN: write up thoughts on tabindex and accesskey
Wendy's proposal sent 4 August 2000
WC explains position.
IJ We are addressing similar issues in UA. We have a requirement that UA allow users to rewind audio. For disposable audio, it is not required of the UA to rewind. Things that are meant for style have less support.
I don't know a piece of content could be important and unimportant.
CMN You scope the question of importance in relation to "important to what." if not important then the equivalent is important. If something in the video is important then you have to get through.
In looking at the video as an object the video is independent of the container. This particular thing is not accessible.
WC Checkpoint 1.1 says you must require a text equivalent, 1.3 says that you must synchronize it. I believe there are instances where satisfying 1.1 is enough.
IJ Sounds reasonable. Eric Hansen recently sent an e-mail asking about providing alternatives to alternatives and distinguishing primary content and alternative content.
JW I think we need to be careful about decision about multimedia presentation being important or not. I'm a bit reluctant to complicate the requirement if we can avoid it. I'm not greatly a supportive of this requirement. I think the problem lies within the priority system than within the individual requirements. I'm not opposed to a rewording, but we should try to make it as clear as possible. I'm not convinced that the exceptions require a change.
GV I agree with Jason, except that I don't know where to come down. I know where Wendy is coming from. We have a lot of things where we state something simply it can be taken to apply absolutely in cases where it becomes cumbersome and provides no real access. However, if you say, "we'll let the author decide" we slide into the drink pretty quick. They may spend a lot of time developing yet, not important enough to describe. They may spend all that money because people who can see will really appreciate it. People who are blind may appreciate it if when they visit the site that thing was quiet.
IJ One way to express that is, "when synchronization is important to understanding the function, provide the synchronized auditory description."
GV asks about synchronization checkpoint.
CMN reads it:
1.4 For any time-based multimedia presentation (e.g., a movie or animation), synchronize equivalent alternatives (e.g., captions or auditory descriptions of the visual track) with the presentation. [Priority 1]
JW Too many exceptions then people can find ways around them.
CS Flip-side, no exception people could find the requirements ridiculous.
GV Yes, we need to find the balance so that we maintain credibility.
MM This is not just an issue with accessibility of screen readers, but can play into how user agents handle things. With two levels of important, you would have to presume that it is mostly on. If something is important then user agents may handle differently, when ads go multimedia they will always be on. You will have a multimedia issue that will always be on for screen readers. That will be across the web once user agents get into it. I wonder if it is normal priority versus high. It's like the analog to spacer gif. You know it's not important.
JW I am concerned with the interplay of the priority system and the conformance criteria and what is done with them. That is part of the issue. This is probably related to errata for 7.4. There is an incentive to complete all of P1, but not for P2 because then you have to do all of them. The split priority makes conformance more difficult to determine.
WC Split priority is an issue, but what about rewording? Ian's suggestion is just a P1.
GV The key is "multimedia."
WC School House Rock example.
CMN Comes back to "what is the role this object plays in the presentation." You can look at the song on its own as content.
JW How can we work through this? If multimedia, then need auditory description although hard to make that judgement. It's a question of, "are there problems with this and we need to reword this?"
WL Can we have an exception document? List exceptions?
IJ That is errata.
CMN When I was discussing this, I did not look at 1.4. Maybe WC should rewrite proposal in terms of 1.3 and 1.4. It might clarify it better. The point is reasonable.
Action WC: Rewrite 1.3 proposal to consider checkpoint 1.4.
IJ Our requirements should be of the author's perspective alone. Anyone can take the video out of context and say, "it is no longer accessible." The author decides if something is style. Someone else could say, "that's important to me." the author does not know everything that can happen. We must carefully say that important is the author perspective.
GV Yes, look at these two checkpoints together.
IJ Eye-candy can be determined by style.
JW Need to work on 2.0.
WC General theme of my proposals for 7.4 and 1.3 is that we need to address exceptions.
IJ Errata.
WL Techniques.
IJ If you can not understand guidelines w/out techniques, then guidelines are broken.
WL Lawyer asked at hearing, if you can't access the web site, listen to it on the radio.
GV Then why do we have newspapers? radio? why have the Web at all?
JW In the next version of the guidelines, I'm wondering whether we need to .. after each requirement we have provision for notes. Perhaps we ought to clearly highlight exceptions.
IJ On result of UA experience, the primary comment in the PR review of UAAG cause us to spend 6 months on rewriting. The question, "how do i know when i have conformed." We made sure that the requirement was crystal clear. Only checkpoints normative, everything else informative. Therefore exceptions are in the checkpoints. Request from several reviewers. We have also identified minimal requirements. Everything above and beyond is should and may discussed in Note or techniques doc. I would encourage this group to take that approach. Put everything you need to know in the checkpoint.
IJ WC says example is a periodic refresh. It is new content not a refresh. Even if a page in between that says, "please wait." Simply a transaction.
CS It might be using the same HTML element.
JW We need to distinguish between the mechanism and the substance.
CS I agree and think it might be a confusing wording issue.
JW It is instigated by the user.
CS I agree, but the current wording says to me, "don't use meta refresh tag no matter what."
JW Should there be a clarifying note?
CMN It makes me uneasy.
IJ The two cases: on user demand vs. not on user demand. This is clearly on user demand.
CMN From the user's experience, it is trying to figure out what is going on.
IJ So the author sets expectations, "When you submit this, you will be asked to wait some time before receiving your results."
WC 7.4 Until user agents provide the ability to stop the refresh, avoid creating periodically auto-refreshing pages where a user might miss important information or be confused by quick changes in content. [Priority 2]
could say:
7.4 Until user agents provide the ability to stop the refresh, avoid creating periodically auto-refreshing pages where a user might be confused by quick changes in content. [Priority 2]
IJ Need to make it clearer. There is rationale by saying, "where a user...".
Action WC/IJ Rewrite proposal for 7.4 to avoid use of the word important and make it clear what the scope is.
Action JW - go through recent draft and add qualifications to requirements in the way that IJ suggested. Make clear exceptions and qualifications. I'll try to bring this draft into line with current suggestions.
GV I have comments would like to discuss.
JW Write-up, send to list, discuss next week?
WC Also, consider agenda items for face 2 face and submit to the list.
JW We're also working on the editing process.
$Date: 2000/11/08 08:27:09 $ Wendy Chisholm