WAI Authoring Tool Guidelines Working Group
Chair: Jutta treviranus, <firstname.lastname@example.org>
Date: Wednesday 20 October 1999
Time: 3.30pm - 5:00pm Boston time (1930Z - 2100Z)
Phone number: Tobin Bridge, +1 (617) 252 7000
The Latest Draft is dated 14 October, available at http://www.w3.org/WAI/AU/WAI-AUTOOLS-19991014.
CMN We discussed the idea. User Agent appear to have ditched the mulitple types of conformance. WCAG already allows a conformance claim that says "this process produces accessible content". It creates confusion.
JT The tool needs to be conformant, not just the content. There have been fairly strong arguments for not separating the two.
PJ If someone conforms to one or the other they can claim one side and not the other - makes it easy to measure conformance.
GR It makes no sense to have one without the other. We have been through this ad nauseam before
/* IJ joins
PJ We're not spiltting the priorities- can we make a statement that we are partly done, and would it be useful to know that. I think this is valuable information.
CMN I think there are a lot of combinations that can be useful. It weakens the conformance to have multiple levels. We can't stop people from claiming that they meet guidelines 1 thru 4 or guidelines 6 and 7.
PJ Agree that adding more conformance statements we complicate the issue
JT There is also the suggestion that a manufacturer will go for one conformance level and then stop.
JR THere is the possibility that people will be triple-A compliant for a single goal.
PJ You conform or not. After that you can say what you like about how close or how far you are.
Resolved: Single type of conformance (three levels)
PJ They are close to 4.2
CMN When the author is creating content it is a case of asking the author for information as they add images. When they are working on imported content it is really the same sort of thing as 4.2.
GR Would it help to say "during creation" in 3.1
PJ So prompting is while you are adding, and checking comes later. Would prompt be satisfied by a place to add alt content for an image?
CMN, GR Yes
PJ I like support, not prompt- prompt sounds to developers like a warning
JR So what is wrong with assist?
PJ It is more than assist that we want. A developer will add the field. Adding help to that comes under 4.2
CMN I think having developers see it as a warning is a good thing
GR "Allow the author to provide alternative information dring teh creation or editing.."
PJ Sounds like guideline 1
JT And we are after something more than allowing it - we want the tool to actively encourage it.
JR Or making it easy for them. I think assist covers most of what we are talking about
CMN It seems tighter language.Prompting is narrower than assist - it means get the information from the author.
JR As long as we have a good definition of prompt it seems good
PJ I like the word support.
CMN This will be linked to the definition.
Resolved: Wording of 3.1 to be 'prompt'
PJ I don't understand why it is P1. There are 28 checkpoints, 16 have a P1 component. If we rank the checkpoints, assisting the author seems not to be as critical as 4.1
CMN The minimum satisfaction is to provide information
PJ Can I do that by linking to WCAG
CMN Yes. It seems a clear P1 for the author to know how to fix a problem.
PJ There are plenty of tools that don't explain how to fix things
CMN THere are plenty that do - HomeSite does, ASWedit does,
GR There is context-sensitive help in HomeSite
PJ The problem I had was trying to embed a movie. They don't have that level of help, so it is a lower priority.
JT THat sounds like an argument against yourself.
GR My concern is that you can't lump relative priorities with P1. I have very strong reservations about the qunatitative approach to prioritization.
PJ I undestand that. But we need to decide it at some point. One approach is to count them
GR On the other hand, we have done a very good job of tersifying, whch has led to a top-heavy document.
PJ Changing this priority doesn't make it less top heavy. It seems to me that providing additional assistance is not the critical part, checking is
CMN I think that there is an equally strong argument the other way - they seem both P1 requirements to me.
PJ My point is that accessibilitiy is being viewed as another functionality. When we make everything P1 they are overwhelmed with the functions which have to be added.
JR I think there are all sorts of P3s, but we have focussed on P1
PJ We could resolve this by noting Phill's disagreement.
Resolved (PJ dissenting): 4.2 is relative priority
IJ I don't think that is the definition that I gave at the face to face.
PJ It is too complex. I would be happy to take a go at rewording it - if I am the only one that has the concern we can resolve ity
JR This could be said in a terser way.
PJ We want to make sure they understand that there are different priorities in WCAG
Action PJ: Propose definition for Relative Priority - to be sorted out on list.
CMN I have an appointment with Tim for Monday morning at 10am
PJ A checklist (a la WCAG)?
CMN I would like not to split the checkpoints into themes - they are small groups already.
JR Are there things other than guideline 7 things under user interface? (I would object that)
IJ It will be priority 1, then relative priority, then p2, then p3, in the order they appear.
JT Does anyone want to rescind their recorded dissnets?
WL I also dissent to 5.2 being P2
PJ Can you post a summary of dissenting opinions to the list?
CMN No problem.
Action CMN: Post list of dissenting opinions to list
JT Does everyone agree to ask for Proposed Recommendation
Resolved: Ask director for Proposed Recommendation Status on Monday morning.
/* Eric Hansen has just sent some editorial comments on terms and definitions
Last Modified $Date: 2000/11/08 08:11:51 $ by Charles McCathieNevile