WAI Authoring Tool Guidelines Working Group
Chair: Jutta Treviranus
Date: Tuesday 18 July 2000
Time: 2:30pm - 4:00pm Boston time (1830Z - 2000Z)
Phone number: Tobin Bridge, +1 (617) 252 7000
The Latest Draft is the Recommendation dated 3 February, available at http://www.w3.org/TR/2000/REC-ATAG10-20000203. The latest techniques draft is dated 4 May March, available at http://www.w3.org/WAI/AU/WD-ATAG10-TECHS-20000504. The latest draft of the Accessibiltiy Evaluation and Repair Techniques is dated 15 March at http://www.w3.org/WAI/ER/IG/ert-20000315
CMN a possible opportunity came up for a meeting in Canada depending on lots of other stuff, in Ottawa, October 5 or 6.
WL there was an advantage in that someone can be there?
CMN It was a possibility
JT There are several other October possiblities. SMC - Nashville, and EduCar with the courseware people coming, week of October 9-15. Opportunity to draw in Courseware authoring people. Other option is Closing the Gap, week of 16th.
HS Right now I will be in the South on October 14, and no plans to be at closing the gap yet.
WL isn't WCAG looking at having a meeting at the same time?
CMN in the UK. One of the questions is are we ready for a meeting, do we want it earlier, later, ?
JT Part of the idea was to have working meeting - spend some of the time working on particular tasks.
WL I don' think we have anything as hot as the content guidelines.
CMN Are people following that at the moment?
WL They are starting a major rewrite of the WCAG. We don't have to do that so urgently
JT Why don't we run a thread on the list.
JT This is a description that I sent to the CG. Reads from archives.
WL Who is going to bid on this?
JT It is just a draft
WL But who is a possible candidate?
JT ETS, there are several in Toronto, ... Ideally we would want someone who is involved in WAI and has good background knowledge and has the required research skills.
WL I am somewhat taken aback becuase the efforts on an informal basis has been fairly overwhelming. It sounds a little pie in the sky.
JR I don't think it is that hard - it would take a person a couple of weeks or a month.
JT There is nothing new, it is just thinking logically about how to create objective measures - it is a lot of repetitive work.
JR This won't be an automatic tool, it will be a checklist.
WL The thing will let me take Cold Fusion and perform some series of tests.
JT Exactly.
CMN This is in one sense a version of the WART tool from ER with different information on the form. The trick is to develop the information on the form. Developing that information is the tricky bit, and doing it outside the working group process is a complete departure from the way W3C has worked on stuff.
JT We are not talking about developing new information, just about putting the tool together
WL So if Charles and I use the tool we will agree about what clear and simple language are, and what transforms gracefully?
CMN Part of the work has to be done within the group - what are the things that need to be tested, how do we measure the tests.
JT We seem not to have the devoted time within the working group to do this.
WL I think the thing is to question if it can be done.
HS The Windows Logo program worked with a company to create the tests. I think it can certainly be done.
JT We need to do it as much as we can.
WL Isn't this ER work
JT Yes, but they have said that they do not want to look at evaluating tools at the moment.
WL I am not averse to trying. The difference between Microsoft using a company and us doing it is that we have a different kind of requirement for results.
CMN I don't think that is a problem with the feasibility. My concern is that the working group is paying someone to be on the working group.
JT It is paying someone to do the repetitive work that is difficult to do in a working group. The group would still review the work in progress, and specify the outcome.
WL The thing that springs to mind was the film made for Education and Outreach.
JT Right. They didn't make it, but supervised it.
WL More or less. Maybe we have to go through the redundancy. Run it up the flagpole and see what happens.
JT CG were keen on the idea of some way to improve the objectivity of testing. Concerns were raised similar to those expressed here.
CMN My understanding is that we are doing what is requested - sticking with the guidelines we have, but making it clearer how to test conformance.
WL What we are looking at doing is getting someone who isn't in teh group to do something we have been doing.
JT We are getting them to package what we have been doing.
WL If it can be done it is worth doing.
JR I can see Charles' point that it is our responsibility to do this. I can also see how if it is just a matter of having to have the skills to write evaluation protocols, then that's bad.
WL I certainly have no agreemement with the idea that the AU needs significant revision.
JT I agree
JR It is a really difficult thing to do these conformance evaluations - mostly because of WCAG. It is a lot of work to go through thoroughly.
JT If it were a form it would be easier
JR It will be a long form.
JT There are some things where you won't do all things, but an important representative set.
JR I think the way of doing this might be to produce 3 documents. One for P1 items (including relative) and so on. We concentrate on the first one first. There is no point in getting a tool evaluated for P3 items that fail P1 items.
CMN Not necessarily. There is value in being able to pick a tool by checkpoints in the absence of a triple-A tool.
WL The concern about acccessibility in webwatch includes things that are P1, but there are things that are important in real cases where some P3 may be more important than a particular P1. If a tool is better than what people have, that's progress.
JT We rank things according to our guidelines. But individuals can find value by being able to get that at a very granular level, according to their particular needs.
CMN It really is a big tedious job to test a piece of software. And we are the Working Group who needs to specify what needs to be tested.
JT And there is making sure that the test we do are repeatable.
CMN I don't think it will save us much work, since we still need to specify it.
JT It may help us to focus on what we need to be looking at.
Action JT: Discuss the tool development proposal again at WAI-CG this afternoon.
JT now CMN and I are here we should try to resolve this.
HS What is the quorum procedure?
JT I think we had quorum.
CMN You can overrule my objection if I am the only one. (The priveliges of being a chair...) My problems were that the definition seemed to include a lot of stuff that isn't definition, and doesn't define in what way a prompt is different from checking, testing, etc.
JT There are two things - it is intiated by the tool, and that it is intended to encourage the author
CMN Measuring intention is tricky - it is difficult to see what that means when deciding whether something is a prompt.
JT The operative line is that it is something initiated by the tool. You can try out a tool and see if authors turn off the access features.
CMN It can be tested afterwards, but it is difficult to use the definition to specify beforehand.
JR We are going to have to throw this open to allow developers to work with their tool. I would like to see it more objective, but it is a subjective thing.
WL The objection raised was requiring an act on the part of the author.
JR And there was a debate on whether selecting "ignore" is an act. It is difficult to tell the difference between ignoring and not noticing.
HS My objection is to the requirement that the user had to click on something.
JR I would like to do some testing...
JT We want to create the tool in such a way that someone is more liekely than not to include alt text.
HS And that is up to the tool developer to know their users. I would rather get my user to put in good text or make it clear that they have just ignored it.
JR There is a mid range - if you require something/anything, then that is too much. not having it appear is too little.
Resolved: We will use the definition of prompt proposed by Jutta and adopted (CMN objecting).
HS I am following up on Windows logo requirements and how we developed the requirements and tests.
CMN The database of techniques that is required to do the testing document is going to take a couple of weeks or so. The tools database should be online enough to generate the tools and the reviews pages in a week or so.
CMN Updating issues list is outstanding.
Last Modified $Date: 2000/11/08 08:13:13 $