Minutes for 09 November 2000 WCAG WG telecon

Available at: http://www.w3.org/WAI/GL/2000/11/09-minutes.html

Minutes for 09 November 2000 WCAG WG telecon
Summary of action items and resolutions
·       Resolved: keep definition in checkpoint 1.1
·       Resolved: Replace "element" with "content" (checkpoint 1.1)
·       Resolved: Replace "Examples" with "Non-normative examples"
·       Resolved: Explain "non-normative"
·       Resolved: Graphic art should be done by a professional.
·       Resolved: Adopt Checkpoint 1.1 with minor resolutions made thus far.
·       Resolved: Adopt new section in introduction, change headings to 
statements rather than questions.
·       Resolved: include definition of auditory description in Glossary.
·       Action: LS and WC discuss auditory descriptions off-line.
·       Resolved: Add examples to checkpoint 1.3 (positive and negative) to 
make it more clear as to when auditory descriptions are needed or not.
·       Action WL: Write a proposal for 1.1., 1.2, and 1.3.
·       Resolved: add synchronization and examples to 1.3.
·       Resolved: add example to 2.1 that says don't use presentation 
elements for structure.
·       Action WC: send rewritten 2.3 to list to discuss at next week's 
meeting.

Participants
·       Cynthia Shelly
·       Dick Brown
·       Annuska Perkins
·       Loretta Guarno-Reid (sp?)
·       Andi Snow-Weaver
·       William Loughborough
·       Jason White
·       Wendy Chisholm
·       Lisa Seeman
·       Gregg Vanderheiden

Regrets
·       Kynn Bartlett
·       Len Kasday
·       Katie Haritos-Shea

Agenda
discuss the proposed revisions of the WCAG 2.0 7 November 2000 working draft.

Checkpoint 1.1
CS Wonder if defn of text equivalent should be moved to the glossary
DB Can reference from elsewhere.
WC Would like to keep it here for a few reasons, particularly for policy 
makers who may not as easily follow a link or for those who print this out. 
But, could copy definition to the glossary.
LS Does more than defines it, it clarifies it.
Resolved: keep definition in checkpoint 1.1
Resolved: Replace "element" with "content"
WC Will incorporate some suggestions into examples from discussions on list.
WL Aren't normative, therefore shouldn't matter. they're Examples.
CS Should say "non-normative."
Resolved: Replace "Examples" with "Non-normative examples"
Resolved: Explain "non-normative"
WC Plus, we'll be adding links to techniques, so this will evolve.
CS I like having the graphic, but we need to hire someone.
WC Never claimed to be a graphic artist. <grin> Any suggestions for who I 
could work with? Will continue to scratch out ideas to help get the ideas down.
WL Placeholders are better than nothing.
Resolved: Graphic art should be done by a professional.
Resolved: Adopt Checkpoint 1.1 with minor resolutions made thus far.

Introduction
ASW Reword heading as statement rather than question.
Resolved: Adopt new section in introduction, change headings to statements 
rather than questions.

Checkpoint 1.3
JW Drawn from WCAG 1.0 due to last week's discussion. It attempts to avoid 
the "until user agents" clause.
WC As with 1.1, include definition in Glossary.
Resolved: include definition of auditory description in Glossary.
LS Why isn't 1.3 an example under 1.1?
WC describes an auditory description.
LS It has to be in a standard character set.
Action: LS and WC discuss auditory descriptions off-line.
DB Do we really mean all multimedia? What about a slideshow.
JW Does not fit our description of auditory description.
WC Need to include examples on this one like 1.1 to help people figure out 
when and when not to use auditory descriptions?
WL Need also negative examples (when not to have auditory descriptions).
Resolved: Add examples to checkpoint 1.3 (positive and negative) to make it 
more clear as to when auditory descriptions are needed or not.
WL It's possible that synchronization should be included in the checkpoint. 
That would help clarify Dick's question. Once slide loaded, don't need to 
synchronize description.
WC Something like, "Describe the essential visual info and synchronize 
those descriptions..."
JW Changing presentation where both media are changing and synchronized.
WC 1.1 provide text equivalents, 1.2 synchronize text equivalents 
(captions), 1.3 synchronize auditory descriptions.
Action WL: Write a proposal for 1.1., 1.2, and 1.3.
Resolved: add synchronization and examples to 1.3.

Checkpoint 2.1
minor edit.
LS When incorporate proposal for Guideline 2? It might filter down to the 
checkpoints.
WC Those that did not have clear outcomes I did not incorporate. They will 
not get lost. I track issues. Not all of them get represented on the issues 
list, but I have a very regimented e-mail sorting process and archiving.
LS Still concerned about changes in guideline to checkpoints.
WL Yes, guideline may change, and checkpoints will change, but shouldn't 
stop us from making these changes now.
WL This checkpoint requires: xyz.
WC Like in 1.1 list of bullets that say, "here's what we're talking about."
WL It's a style of doing it.
WC Yes, specifically played with new style on 1.1 and 2.3. If people like 
we can adopt for other checkpoints.
JW In terms of 2.3, the examples are good, be good to have points that...KB 
and I had statements of what it means. We could clarify 2.3 a bit more. 
Keep the list of points and examples.
/* GV joins */
JW If we want a consistent style suggest: text of each checkpoint, 
discussion (what it requires), examples. These should be clearly delimited.
GV Except where you have "e.g." to explain a term. Examples should not be 
technology specific.
JW Agreed that examples are not normative, question of style.
WC Let's trend in that direction, but let's not say we have to apply to all 
right now. The style will evolve. It seems that it works for 1.1. and 2.1, 
we'll try to apply.
GV Right, let's not get religious about it. If something is succinct, don't 
need to make it longer.
LS Right.
GV Examples can take you in a technology direction.

Checkpoint 2.3
JW Has been considerably rewritten.
WC how does an author use just a data model?
JW Content embedded in a program without any markup. It doesn't reside in a 
different file with markup, it's generated by software.
GV I have a page that is text
WC Like using accessibility class properties in Java.
CS If you have all of your content in a database and it doesn't get marked 
up until the output phase, diff markup might be generated.
GV Still use markup to use structure.
DB People are getting data from a server, device uses the data.
GV I have a device on a small screen and it downloads a document the only 
part that's visible is the paragraph i'm reading. if i say give me the next 
one, the current one disappears and the next one shows up. data is there in 
the background but the document is only plain text.
DB It could be on the server (doesn't have to be local).
WL What's a data model?
DB People use it in a lot of ways. Description of how one piece of data 
relates to another.
JW Or something in the DOM. Something that could be marked up in XML could 
be stored in a diff form and then made available to DOM from server.
WL Problem is that "markup" means something that is not in the data model.
CS Markup is one way to express structure, database tables is another way.
WC Are we combining the technique and checkpoint? Should it say, "Provide 
the logical structure of content." ?
DB/CS agree.
JW We need the structure in an explicit form that can be accessed on. The 
shortest way to express that is how it currently reads.
DB You want to ensure that the logical structure can be inferred reliably.
JW Yes. It must be absolutely clear and machine processable rather than 
discernable through human judgement given a presentation.
WL If the adverb "explicitly" gets in there, does that handle JW's concern?
WC Can handle like we did with 1.1.
WL markup and data model are ways to do it.
JW Categorization parts of content. It attaches categories to content which 
helps later processing.
GV My concern is "provide" if we say "expose" or "present." If I use a 
database, it doesn't mean the user will have access with it.
CS Ongoing controversy where that happens. The assumption in WCAG 1.0 is 
that all info sent down the wire. Based on user preferences stored on 
server, can the server make a determination ahead of time. Then, the data 
model is only on the server.
GV The issue is, when you send the data, did you expose the structure in 
the process. The data model is only useful if it presents structure to the 
user.
LGR To the user agent.
CS Not convinced of that.
GV Server-side solutions can be superior in cases, but only if the 
structure is presented to the user when asked for it.
LS 2 things in this checkpoint: 1. capture 2. expose.
JW If I'm using braille translation software to read a document in order to 
process the document the translator software needs the logical structure. 
The structure needs to come down to the proxy or UA.
LS There is more to it. "Capture then expose." This could fall through the 
cracks if it is not made explicitly. With CSS you can call everything a 
paragraph and call a paragraph 1 an H1.
JW That would fail 2.1.
WL You would be using CSS to do structure.
JW You would not be using heading elements.
LS That has to be explicit in an example.
WC reads 2.1. says don't use structure for presentation, likewise add don't 
use presentation for structure.
/* Agreed.*/
Resolved: add example to 2.1 that says don't use presentation elements for 
structure.
WC Expose and capture have been rejected.
GV Why expose rejected?
WC Not clear what the author has to do.
DB Why not provide?
GV If provide but user does not have access.
DB Too visual. It should be used.
CS Technical definition, that works, but may conflict with lay term.
JW Not many better choices.
CS Need to get the idea that we have a data model and how it breaks out. 
Organized appropriately. Expose to proxy (or whatever is down the line).
GV Provide a structure to your content if a form that can be accessed by 
the user's program. Need an easier to understand statement. Should be less 
technical.
JW Wendy put forward a proposal, a parallel with 1.1. "Proivide logical 
structure of content explicitly." under that would be more technical 
requirements of what needs to be satisfied (structure stored or available 
that can be processed by software, etc.) Then the examples would follow to 
further clarify what logical structure means. If people think that would 
work, we could invite wendy to draft it.
LS sounds good.
CS On next week, can we agree on the technical requirement.
Action WC: send rewritten 2.3 to list to discuss at next week's meeting.

$Date: 2000/11/09 22:41:17 $ Wendy Chisholm

--
wendy a chisholm
world wide web consortium
web accessibility initiative
madison, wi usa
tel: +1 608 663 6346
/--