The rest of the time was spent discussing checkpoint 1.2. We came up with good points, but no resolution. CS, PB, and WAC took an action to summarize today's discussion to come up with criteria for 1.2.
Action CS, PB, WAC come up with criteria for checkpoint 1.2
JW Identify issues to address before it goes public. We have a few actions for particular checkpoints. Particularly 2.2, flicker, color contrast. Please identify issues that need to be addressed.
PB The first example under 1.1, this graphic bothers me.
WAC It's an arrow that's used to move to next slide. It has the word next since Anne suggested that people need text and graphics.
PB We ought to add text to every graphic?
WAC No. It's probably not the best example. Feel free to submit new ones. Ideally, each example should be illustrated.
JS I sent comments a few weeks ago. I sent comments about wording and the examples. I don't see that they have been covered.
WAC Might be best to resend them.
JW Are those editorial?
JS Primarily, from what I remember.
WAC Ben's appendix looks good.
JW What are the concerns w/this checkpoint?
CS We were talking about captioning for multimedia, question of webcams came up. One example was UWash web site - webcam of the quad. Text underneath of live weather report. It seems like a reasonable equivalent. They haven't actually associated the weather report with the image, but it does convey a lot of what is important about that image.
JW Suggestions for wording?
GV If a cam out a window, what is the equivalent?
CS alt-text should be fine.
JS Agreed.
GV Basically a window.
CS If the infrastructure to get a live weather report and use as alt-text, that would be nice. But not required.
GV A link to a weather forecast.
CS Yes.
JW Exclude from conformance claim. The author intended purpose at levels 1 and 2 could be regarded as paramount consideration.
CS I don't want the situation where when trying to adop, it would not be acceptable for someone to point a cam at a coffee pot unless someone had updated text equivalent of coffee levels.
PB In order for the coffee-pot cam to be accessible, there would need to be some way to provide real-time text equivalents. Therefore, some way to indicate if the coffee pot is full or not. However, I could put it up there but not include in conformance claim. Therefore, the definition is that it would need to be accessible.
JS My choice would then not to put it up if it won't be accessible.
GV Depends on the level. If you don't do level 3 it won't be completely accessible. At what level to we require that coffee pot cams be accessible?
JW Concerned about resources and such are sneaking into what level things are at. I thought we ruled that out.
GV Resource is explicitly used to determined level.
JW An issue last time was how to define it. There have been objections about that.
CS How would you structure the coffee cam requirement so that not required to do now when technology is not readily available but is required when the tech exists.
JW It is a requirement that the author's purpose be fulfilled. If to provide actual information, you are not providing it. It would have to be ommitted from scope of the conformance claim until it is accessible. The metadata would reflect state of the situation.
CS The way we handle in the success criteria isn't bad. #4 (reads).
JW It seems a bit circular. If they are available it is imposed as a condition, but providing them is what you need to do to make them available.
CS It's like the weather example. Weather feeds exist, so you can use that. The coffee measuring sensor (in my office) doesn't exist. Therefore, that equivalent doesn't exist. But, if my coffee pot had that feature, I should hook up to it.
GV If we set it up so that when the rules came out they made sense. As time goes on and things get easier to conform to (due to technology) then they can more easily conform at higher levels.
CS I agree with that.
JW I thought we had decided that when you have a script, synch no longer makes sense. No point in having it change on people. No synch requirement on complete text equivalents for dynamic multimedia presentations. That's a different issue. We could say that the author's purpose in providing the content needs to be fulfilled. If it can only be fulfilled with a real-time caption or description then that is what is required. Otherwise, create a qualification on the guidelines that these have to implemented with reasonableness.
PB Could you restate the last idea?
JW To say that the guideliens define what it means to be accessible and the issue of whether one makes a conformance claim that applies to all content, or instances where it is not feasible to satisfy, those issues are affected by local laws and policy considerations that are out of our scope but should be in the mind's of implementors.
GV I thought that we...the access board does not say "if you follow it will be accessible." Accessibility is on a person by person basis. I can always find someone who can not use it. If the guidelines do not determine accessibility, what do they determine? I thought the levels determined levels of accessibility. Level 1 - some level of accessibility, level 2 - higher level, level 3 - extraordinarily accessible. When they have done level 1, 2 or 3, what are we saying they have done.
JS the phrase "substantially meet the need"...
PB If alt-text says "status of coffee pot" - not meeting the need. Would need to provide info about how full it is. Either makes whole site fail, or just that page or just that element, not sure. I think I'm in favor of having a general set of guidelines about how to make things accessible rather than creating something that can be used as a government checklist. The trend is to go with 508 (in the U.S.) one thing b/c of legal fears and also because easier to implement. However, like the idea of having something that describes what really needs to be done.
JS I agree. UT-Austin also about to adopt 508. The argument I've been pushing is to keep WCAG in mind for setting a higher bar and a more comprehensive view. I'd like our developers to aim to that. We should aim to fulfill that function.
PB It means that not all of our guidelines implemented as law, and that's ok. There needs to be a list of how to do it right and completely.
CS As long as we continue to maintain as possible and testable as criteria.
PB I agree. That makes it harder. We are trying to address both goals.
JW You can easily determine if someone has conformed or not. It can be achieved for all content. If ignore resource related issues, you can provide real-time captions for them.
CS Seems to fall down on the sillyness test. Would a site fail if it had a photo of the coffee pot if it only had alt-text?
PB Another example: camera for traffic.
CS Taking it down b/c it is not accessible is not a good thing for us to do.
JW We're not saying take it down.
PB But, it would not conform.
CS And if the owner of the site has a legal requirement then they will have to take it down.
WAC Traffic similar to weather example. Likely to be someone providing similar data in text. We've talked about pulling in inaccessible content from other places, what about pulling content from other sites to increase accessibility? Web services as an answer to help people conform to some of these, particularly the language requirements.
GV A requirement or recommendation?
WAC I don't know, it's something that we need to discuss.
JW You could interpret success criterion 4 as saying they could always be available b/c someone could provide in real-time. Can the person using it fulfill the author's purpose? If largely decorative, then maybe not so significant. Perhaps distinguish at levels...author's purpose if important is at higher level while other uses might be further down. This distinction is made in the Device Indie Principles.
WAC In 1.0 said "important" - people complained about judgement calls.
CS JW's proposal seemed clearer.
JW Can the user achieve the author's intended purpose.
WAC That addresses web cams, what about internet radio?
JW The whole purpose is to provide the information. Therefore, provide real-time.
CS Broad cast standards for the accessibility of radio? NPR makes transcripts available.
JS NPR not in real-time.
LR I looked at it here locally, they would not be able to put the words up. Many people are moving away from.
CS Yes, lots going on in congress right now.
JW If audio-only, so synch not as important. However, reasonable timeliness to provide the info. Slightly less than real-time could meet the needs.
CS NPR broadcast over airwaves and internet should meet similar rules.
PB we recently had a broadcast of a phone conversation. People could add in comment, call in. We had it captioned, live, over the internet. sometimes it is important to have it live as well - when there is interaction.
WAC How much did that cost?
PB captioner time, and dev time to come up with. It took about 1 month. In the future, all we would need is the transcriber. Once built, just cost of transcriber and cost of the event.
JW Interactivity is a good test to determine real-time or not.
CS Also a cheap way to do that - a chat instead of phone.
LGR Don't have to pay a transcriber.
WAC But what about people who have a hard time reading text?
CS Screen-reader...but this is one option for a cheaper cost. Input problem.
JS And output problem. Trying to read quickly enough.
JW That's why we have levels. It will satisfy substantial needs where someone can't keep up with. If real-time it isn't going to satisfy unless a higher level is met. A test for level 1 - people who benefit sufficiently will benefit to that extent if level 1 is met.
PB Criteria for determining if the requirement is needed in real-time. Interactivity is one. Emergency broadcast system or breaking news would be a timeliness issue.
Action CS, PB, WAC come up with criteria for checkpoint 1.2
First thing on the agenda next week.
2nd comment on same issue - Jon Gunderson
$Date: 2002/05/24 17:00:25 $ Autho