W3C

WCAG WG Meeting

7 Apr 2005

Agenda

See also: IRC log

Attendees

Present
Bengt_Farre, John_Slatin, Luca_Mascaro, Yvette_Hoitink, Ben, Wendy, Dave_MacDonald, Joe_Clark, Matt, Gregg, Becky_Gibson, Loretta_Guarino_Reid, Michael_Cooper, Mike_Barta, Tim_Boland, Andi
Regrets
Jason_White, Roberto_Scano, Sebastiano_Nutarelli, WATANABE_Takayuki, Roberto_Castaldo, Roberto_Ellero
Chair
John, Gregg
Scribe
ben, wendy

Contents


 

 

<joeclark> +joeclark

<Yvette_Hoitink> :-)

<wendy> joe - are you calling into the telecon?

<wendy> zakim is tracking who is on the telephone.

<wendy> joe, after you call in you will need to associate your phone number with your irc nick.

<David> Test

<bengt> I get a lot of choppy sounds

<Yvette_Hoitink> bengt: we all do

<bengt> let me call in again somethihng is strange

Techniques Task Force report

<bengt> better ?

<bengt> zakim mute me

<wendy> hmm. not bengt.

<wendy> mc: looking over impact of baseline. thought about how this impacts requirements.

<wendy> mc: issues that came up: 1. guide doc vs general techs. one or 2 docs?

<wendy> mc: still a need for general techniques. not clear on relationships between documents or how people will use them.

<wendy> mc: 2. requirements issue - need reqs for how writing techs wrt baseline. need to add info to techs. how apply to diff baseline.

<wendy> mc: next issue: wanted to finalize matrix of user agents.

<wendy> mc: what UAs will we keep in mind? need a small list that is not overwhelming.

<wendy> mc: returned to matrix of UAs from 1 yr ago.

<wendy> mc: created list. longer than would like (wrt managability)

<wendy> mc: reads list. (available from minutes)

<wendy> http://www.w3.org/2005/04/06-wai-wcag-minutes.html#item03

<wendy> mc: list is windows-only. concern about accessibility features evolving on linux and mac

<wendy> mc: returning to review of test files

<LucaMascaro> tks

<wendy> js: process reminder - to get into queue type "q+" in irc or 61#

<wendy> gv: the need for general techniques to be separate from guide doc. thought that the model was to attempt to bring things together not take apart.

<wendy> js: reminder, we'll be looking at a draft next week.

<wendy> gv: thought baseline by tech not UA. what is the list of browsers?

<wendy> mc: we have repair techniques, to address UA failings. we don't expect those to be in baseline, but we do have techniques to address.

<wendy> joe - that list is being discussed in wednesday telecons until have proposal for thursday call

<wendy> luca - we can't hear you.

<LucaMascaro> i'm a problem sorry

<wendy> luca, can you type your question in irc?

<LucaMascaro> my question is about if is useful to test also with safari on Mac OS X?

<wendy> mc: we considered that, but don't know of accessibility features in safari or mac os so didn't include.

<Zakim> mcmay, you wanted to talk about Safari

<wendy> m3m: rumor is that apple will release os10.4 in april or may and should have accessibility hooks into SR and SR, etc.

<wendy> js: maybe get ppl to look at.

<wendy> tb: thought we agreed to include disclaimer that if ppl wanted tool tested, they could submit a request.

<wendy> mc: yes, but don't have resources to test everything.

<wendy> js: we're getting into discussion of baseline analysis on techniques.

<Zakim> Yvette_Hoitink, you wanted to ask "Focus on screenreaders only?"

<wendy> yh: Focus on screenreaders only?

<wendy> mc: they are on the radar. if the list needs refining, perhaps an item for next wednesday's telecon.

Parameters for choosing baselines described by the WCAG WG (introduce and form task force to investigate)

<wendy> http://lists.w3.org/Archives/Public/w3c-wai-gl/2005AprJun/0010.html

<wendy> yvette - not tuesday, wednesday. info on wcag wg home page under "meetings"

<wendy> mc: goes through proposal. 3 baselines.

<Yvette_Hoitink> wendy - thanks, I thought I hears Michael say Tuesday so I thought it was a different meeting

<wendy> not that i'm aware of. ;)

<wendy> mc: the union of techniques required for both baselines is the set of techniques we recommend using today.

<wendy> mc: caveat - current scripting techniques written w/assumption that scripts will degrade gracefully. do not have techs for scripts that are required. therefore, analysis not complete.

<wendy> mc: will write some techs that would be in that category so can complete the analysis.

<Tim> concerned about coverage of techniques re: Guidelines for particular

<Tim> baseline (numbers) - have we covered all the Guidelines re:

<wendy> js: reactions?

<wendy> tb: for a particular baseline, what coverage do we have for techniques? sufficient # of techniques to use to meet all guidelines?

<wendy> mc: did not to that. if adopt baseline proposal, would be place to start. unofficially, we are likely missing techniques. 1st need to choose baselines we are going to write techniques for, then perform that analysis, and then fill in the gaps.

<ben> http://www.w3.org/WAI/GL/2005/02/11-sc-techniques-mapping.html

<wendy> bc: this shows the mapping of techniques to SC. you'll see several SC that have no techniques.

<wendy> js: if we can get SC and GL set, then more can shift attn to writing techniques.

<wendy> gv: may need to refine the baselines. but, it gives us a place to start and a way to sort techniques and determine what we have or not.

<wendy> jc: if we put good effort into filling holes and then put out a call, imagine they will be able to create techniques. these are the type of problems they love to solve.

<Michael> "they" being standards oriented designers, "high minded people"

<joeclark> the standards-compliant developers can probably come up with at least one technique for each guideline we're missing in HTML+CSS and *possibly* scripting.

<DoyleB> Wendy, I am here, student I was to see medi-vac'd out of teh village. Just back from emergency dental stuff, will likely ONLY do IRC. Count me as here - albeit late. Doyle

<ben> scribe: ben

Conformance analysis / proposal

<DoyleB> Joe - Akiachak

wc: bc and I sent summary - http://lists.w3.org/Archives/Public/w3c-wai-gl/2005AprJun/0021.html
... some issues related to 3rd party verifying claims - may need more details about UA and assumptions made by authors - depends somewhat on what bc, mc and bg come up with regarding techniques baselines
... would be interested to get more feedback on required technologies vs. technologies used

js: questions or clarifications?

tb: wondering if conformance allows claims for resources that rely on multiple conformance claims?

wc: yes, examples we gave include mult. technologies

one exmaple relied on HTML, but used various image formats, another relied on HTML and script, but used others

didn't create an objective measure to diff. which are required vs. used. basically, "relied upon" means an author isn't providing a graceful degradation

wc: relied upon means techs must be supported and turned on.

used means that technologies can be turned off or not supported, but content would still be usable

(you still get all the content, it is just presented differently)

wc: example 3 (web application for photo site) was that script and CSS was relied upon and would not work without support for these techs

tb: usability testing needed?

gv: think this is interesting, one suggestion is that relied upon is hard to interpret

perhaps change it to say xxx technologies must be supported and turned on. and xxx technologies are used, but not required (might make it clearer)

is a very interesting way to point out exactly what we mean

jc: 2 things (btw, flikr works ok w/ out flash) - example 2 example (s5) is a pretty good one - any example we come up with that relies on CSS, there's really no such thing as turning it off (you're turning off author CSS and using default UA CSS unless we're talking about lynx)

wc: we can clarify that, but there are cases where the authors CSS isn't supported in conjunction with script, problems may come up

gv: I think joe's comment is particularly good because if you said CSS is not req'd and people said, "oh good, I can use lynx," then we're sending the wrong message

wc: can see clarification that adds author CSS
... have talked about 3 baselines dependent on HTML, CSS and Javascript, but over time, we may add other technology examples

js: other comments and questions?

gv: what about audience assumptions in conf. claims?

wc: didn't focus on that yet

gv: would like to suggest that audience not be part of it because people often say that people with disabilities are not in their audience.

wc: that was something that came out of F2F - #1 assumption is that people with disabilities are in every audience
... think that as long as audience info is optional, audience information may be very useful for 3rd parties trying to verify validity of conformance claims, but need to give people some parameters for what audience assumptions can be made.

gv: audience is a red flag because people can make a number of problematic assumptions about their audience

js: note that I'm getting ready to propose updates to 3.1 that will include some details about identifying audience, will try to address them, if not in guidelines, then in the guide doc.
... other questions/responses?

wc: next steps?

js: since all 3 of these things (techs analysis, conformance and 4.2) might make sense to go through that first and come back to harvest action items

Impact analysis for guideline 4.2

lgr: in going through 4.2 as we looked at assumption of moving baseline out, we concluded that there was nothing left for 4.2

list discussion raised some questions about that

summary: http://lists.w3.org/Archives/Public/w3c-wai-gl/2005AprJun/0015.html

lgr: techniques work seems to be consistent with the recc. to clarify required technologies

js: conformance work does too

<wendy> s/draming/dreaming

js: what I pick up here is that there are some very specific proposals here, may not need to decide today

one is proposal to delete 4.2 from guidelines in its entirety

js: also jason's proposal
... definitions for "baseline" and "technology" needed

gv: concern (wanted to have group help think this through) if reason we drop 4.2 was because it was being covered by UA and other guidelines, but purpose for 4.2 was about what happens when authors create user interfaces that are not part of user agent
... these elements would not be covered because they don't exist until they are rendered by the UA - 4.2 says if you do that, then you need to meet reqs. from UAAG
... if pieces of user interface come with content, it is authors responsibility to make sure that that UA software that they've created through their interface, it also needs to meet UAAG

lgr: one reason I have troulbe with this is that maybe I'm not familiar enough with techs that can do this - are we just talking about scripting?

gv: could be java, could be flash

lgr: question is whether this is fundamentally a scripting issue?

gv: if you sent something down in flash or java, you could create whole new user interfaces
... JavaScript? no.

lgr: trying to think about types of techs where it is possible to create a user interface - not sure if those are captured by the rest of our guidelines

<wendy> dmd: 4.2 was looking through the lens of html. in 2.0, if something is a user agent, it would fall under UAAG.

<wendy> scribe: wendy

<ben> dm: seems to me that in 4.2, we were looking at guidelines through the lens of HTML, seemed to be applying to HTML + plugins, etc. the thinking was (and we may want to correct it) that in 2.0, we're tech. agnostic and if something really is a UA, then it would fall under UAAG guidelines, if it's something that relates to content

oops.

jc: as for UI embedded into content, e.g., flickr organizer. flash app to upload, tag photos.

<scribe> ACTION: joe look for links and create demo account re: UI embedded into content (stock tickers?) [recorded in http://www.w3.org/2005/04/07-wai-wcag-minutes.html#action01]

jc: stock tickers. updated instantaneously. great example to work from.
... fidelity stock calculator.
... why not send them to UAAG? that makes sense to me.

gv: that's what 4.2 said. "if you do x, implement relevant portions of UAAG" that is being removed.

jc: the advantage of leaving them in is that UAAG and ATAG are less known. ppl have to make conscious decision to look them up.

asw: re: jason's proposal, concern about #2. "is satisifed..." not sure if that is testable.

asw reads #2 from jason's message: http://lists.w3.org/Archives/Public/w3c-wai-gl/2005AprJun/0028.html

asw: how apply to html? there is no UAAG conforming browsers that support html.

js: that's the point of the 2nd clause.

asw: the things that don't conform to in uaag don't matter

<scribe> ACTION: john contact jason to clarify proposal and reword for better understanding by the group. [recorded in http://www.w3.org/2005/04/07-wai-wcag-minutes.html#action02]

asw: concern that if we say, "if you are creating a UA-like thing have to conform to UAAG" - then putting higher standard on ppl writing embedded UA than those writing UAs.
... we're not requiring ppl who write UAs to conform to UAAG.

gv: higher standard - as a user could get a UA that doesn't meet UAAG but then also get an AT that can help get me content. howevr, have no way to get a tool to deal with content that springs UA on me when i access content.

<joeclark> Flickr Organizr:

<joeclark> http://www.flickr.com/photos/organize

<joeclark> for the record.

lgr: my understanding of the proposal is there are aspects of uaag that content author can not repair. e.g., if not accessible documentation provided for the UA, we can't ask the content author to provide that.

(in order for them to use that tech)

lgr: if the uaag requirement had to do with contet, then make sure you satisfy it.
... uaag is good at requiring functional requirements, but not sure if it will be clear for which parts of uaag is a content author to compensate for.

gv: propose postpone discussion of jason's proposal until he's here.

m3m: we have substantial support for what is there (uaag). can assume line of demarcation between ua and content.
... when get into content that is also a ua, then in a grey area between app and browser, uaag at least defines what we expect of the browser.
... if that is something that is failing, that's another way of asying, "until user agents"

js: clarification - adovcating leaving 4.2 (in some form) in?

m3m: don't have solid answer. responding to earlier comment.

<Zakim> wendy, you wanted to say, "not limited to ua, also atag"

wac: hard to imagine flickr as a user agent. depends on the functionality of the interface element. therefore, ought to consider atag as well.

asw: know we need to have something that addresses custom interfaces, could we make the SC be something like the functional perforamance criteria in 508. meeting uaag could be a technique.

gv: be good for jason/john to touch base with lgr/wac/et al to discuss?

lgr: should we require uaag? even if not all conforming pieces in one uaag.
... think uaag is best effort to characterize what is important, but uncomfortable that no ua's are completely conforming.
... then nervous about the role we give uaag in wcag.
... also concerned about the circularity that we could gt into (wcag ref atag ref wcag...)

<Zakim> mcmay, you wanted to talk about Flickr

m3m: uaag is one of the most rigorous set of requirements
... if it were easier, more ppl would do it.
... the gaps in the system, would not call catastrophic.
... flickr is web content. the fact that wcag doesn't handle web apps is a bug on wcag.
... there are things to reference from atag and uaag, but the idea of a web app accessibility guidelines should be considered.
... when talking about dom vs object model, we do some of that well in the guidelines, but missing guidance that we can't offload onto atag or uaag or xag or 508 or s/w accessibility guidelines.

<Zakim> joeclark, you wanted to make an actual proposal about WCAG+{ATAG|UAAG}

jc: could we have guideline like 4.2 "if web content also has functions of authoring tool, conform to relevant parts of atag. if has ua functions, conform to relevant parts of atag."
... eg., if blogging tool, have authoring tool components in your web content.
... e.g., if can arrange slideshows, that is user agent, conform to abc of uaag
... technique provides more info about doing this. may raise awareness of the complete set.

<scribe> ACTION: joe write proposal for 4.2 to link to atag and uaag depending on function of content [recorded in http://www.w3.org/2005/04/07-wai-wcag-minutes.html#action03]

<scribe> ACTION: wac, jason, loretta, john (maybe joe) work together on 4.2 ala link to atag and uaag depending on function of content [recorded in http://www.w3.org/2005/04/07-wai-wcag-minutes.html#action04]

lgr: we should identify some of the guidelines/checkpoints from uaag.
... we're subsetting uaag.
... also making those claims will depend on underlying ua.
... if none of those uaag have published a conformance claim, then there is no documented info for the author to fallback to. make his/her determination more difficult.

js: sounds like another item to take back into the mix.
... one big action item is to take 4.2 group + wendy, joe (if possible), jason, john and think about the interface among wcag, atag, and uaag
... also asw's suggestion about 508 and loretta's suggestion about documenting wrt uaag and atag
... propose that for the sake of having a date on it, we discuss it 2 weeks from today. 21 april
... something to the list on tuesday 19 april at latest.

<scribe> ACTION: loretta assemble 4.2 group + wendy, joe (if possible), jason, john and think about the interface among wcag, atag, and uaag, asw's suggestion about 508, loretta's concern about documenting wrt uaag and atag [recorded in http://www.w3.org/2005/04/07-wai-wcag-minutes.html#action05]

action 5 = loretta assemble 4.2 group + wendy, joe (if possible), jason, john and think about the interface among wcag, atag, and uaag, asw's suggestion about 508, loretta's concern about documenting wrt uaag and atag. have something to the list by 19 april.

<scribe> ACTION: michael et al continue techniques analysis by 12 april [recorded in http://www.w3.org/2005/04/07-wai-wcag-minutes.html#action06]

<scribe> ACTION: wendy & ben document issues raised by tim [recorded in http://www.w3.org/2005/04/07-wai-wcag-minutes.html#action07]

<ben> scribe: ben

js: are there things that came out of conformance discussion that require specific actions?

gv: yes, in conformance claim, might be worded to say a list of those that must be supported and turned on and a list of those that are not needed -- and reflect that in examples

<wendy> action 7 - wendy & ben document issues/incorporate suggestions raised today, look at QA chunks to determine if addressed

<wendy> action 7 = wendy & ben document issues/incorporate suggestions raised today, look at QA chunks to determine if addressed

js: next week, main item is to talk about structure, specifically some drafts of guide to WCAG 2.0 document that deal with specific SC - should help to focus concern in techniques report about whether general techniques has been/should be subsumed into guide or if gen. techniques are available in a separate document
... please watch the list for drafts and proposals and read them prior to the mtg. as this is another critical piece in this discussion of where we're going
... I think we're much further along than we were 3-4 weeks ago and lots of good work has been done in recent past

Summary of Action Items

[NEW] ACTION: joe look for links and create demo account re: UI embedded into content (stock tickers?) [recorded in http://www.w3.org/2005/04/07-wai-wcag-minutes.html#action01]
[NEW] ACTION: joe write proposal for 4.2 to link to atag and uaag depending on function of content [recorded in http://www.w3.org/2005/04/07-wai-wcag-minutes.html#action03]
[NEW] ACTION: john contact jason to clarify proposal and reword for better understanding by the group. [recorded in http://www.w3.org/2005/04/07-wai-wcag-minutes.html#action02]
[NEW] ACTION: loretta assemble 4.2 group + wendy, joe (if possible), jason, john and think about the interface among wcag, atag, and uaag, asw's suggestion about 508, loretta's concern about documenting wrt uaag and atag [recorded in http://www.w3.org/2005/04/07-wai-wcag-minutes.html#action05]
[NEW] ACTION: michael et al continue techniques analysis by 12 april [recorded in http://www.w3.org/2005/04/07-wai-wcag-minutes.html#action06]
[NEW] ACTION: wac, jason, loretta, john (maybe joe) work together on 4.2 ala link to atag and uaag depending on function of content [recorded in http://www.w3.org/2005/04/07-wai-wcag-minutes.html#action04]
[NEW] ACTION: wendy & ben document issues raised by tim [recorded in http://www.w3.org/2005/04/07-wai-wcag-minutes.html#action07]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.122 (CVS log)
$Date: 2005/05/23 22:25:19 $