17 Apr 2008


See also: IRC log


Jim Allan




<scribe> Scribe: Jan

<KFord> This is just an FYI that Marc from IE will be joinging us today but he will be a couple minutes late. He's here to talk about the ARIA document and represents us in PF.

<aaronlev> what's the passcode?

<KFord> 82941#

<aaronlev> i got invalid passcode

<aaronlev> now it works

<Alan> ?

<KFord> Welcome.

<Alan> Thanks

1. Charter review


JA: Says what we are going to do for a couple years
... Which will mostly be UAAG2

KF: Question about success criteria....
... Success criteria don't include testing materials...should they?

JA: Don't know

JR: Sure let's add it

KF: Other than that it looks ok?


<oedipus_laptop> what about a UA Support For... type document which is maintained to document conformance / support?

JR: I was following WCAG's kead on how to fill in charter

<oedipus_laptop> yes JR

KF: So an implementation summary?
... How all browsers have done

<oedipus_laptop> yes, 1.0 does, but it hasn't been updated since 2003 - i have an action item to update it

JA: Right and that's part of the CR process - Implementation report

<KFord> I'd be willing to help with this update, at least for IE.

JR: Sometimes there is a lack of manpower to do a wide effort at conformance testing

JA: OK we can focus a bit more on that after we get to Rec

AL: Charter not a lot of detail

<oedipus_laptop> http://www.hicom.net/~oedipus/temp/w3c/test/until-user-agent-clauses.html

<oedipus_laptop> http://www.hicom.net/~oedipus/temp/w3c/test/WAI-UA-Support_2007.html

AL: Felt requirements document is more useful (http://www.w3.org/TR/2007/WD-UAAG20-requirements-20071031/)
... I'd be willing to provide feedback on the Req document

JAL Good! We'd like feedback!

JA: Anything else?

AC: Mission...
... Should Mission also include statement that support documents will be created to the guideline.

JA: Right but maybe that's just duplicating too much.

AC: Interesting how repetitive the document is

<oedipus_laptop> http://www.w3.org/WAI/Resources/WAI-UA-Support

JR: These charters are very high level

JA: We'll vote next week...then we need to go through Judy

<Zakim> oedipus_laptop, you wanted to ask that Access Module issues be added to agenda

GR: Acces module is important piece - is it ok if cross posts coninue?

JA: Yes cross posts ok

<KFord> Marc is on his way to my office.

<oedipus_laptop> must activate be boolean? http://lists.w3.org/Archives/Public/wai-xtech/2008Apr/0175.html

<oedipus_laptop> redefining keys and user control cascade: http://lists.w3.org/Archives/Public/wai-xtech/2008Apr/0176.html

2. Discuss/review ARIA UA Implementors Guide

<oedipus_laptop> (both XHTML Access Module Issues)

AL: Introduces self...
... Works on Firefox, WAI-PF, ARIA

<oedipus_laptop> http://www.w3.org/TR/wai-aria

<oedipus_laptop> http://www.w3.org/TR/wai-aria-roadmap

<oedipus_laptop> http://www.w3.org/TR/wai-aria-primer

<oedipus_laptop> http://www.w3.org/TR/wai-aria-practices

Marc Sibley Arrives


MS: Introduces self, works on IE and ARIA with WAI-PF
... So has been working with Simon and AL on this best practices guide

JA: Would like to throw it open

AL: My question is UA working group...

<oedipus_laptop> platform specifics

AL: How deep is UA going to get into platform specifics

<oedipus_laptop> thought we were platform agnostic

JA: In the past, not deep....

<oedipus_laptop> spec should be platform agnostic, techniques where to address

JA: Gave some examples, but pretty agnostic

AL: Right...from talking to other browser vendors...but majority of our work is platform specific....
... WHat we really need is help coordinating between platforms

KF: I don't see that in Guidelines doc it's in tech guide

AL: Don't care about names of documents
... Just care about practical

KF: But our first mission on UA is to be agnostic...in general this is what browser needs to do

<oedipus_laptop> first, do no harm...

KF: These are not UA reqs on platform A

AL: OK if not you, then say that and someone else can do it
... Should I be doing it...implementors guide for ARIA...
... This is the kind of thing HTML should have done....

GR: I think the UAWG is trying to work directly with other working groups eg HTML5....
... e.g. on the User Agent guidance from HTML45
... e.g. on the User Agent guidance from HTML5
... So we are giving feedback back to others

AL: So ARIA implementors guide....not really best practices

GR: Right that's what developers need

AL: Then I'm told go to UA group who should be writing this
... Then WAI-UA responds that's not where are bread is buttered

GR: We want to work with ARIA best practices...
... See ARIA as standard W3C way of feeding in to APIs...
... THen they can look up the mapping

AL: Where do they look them up?

GR: In appendices of ARIA Best Practices doc

AL: Ok but doesn't map to the platform specifics
... OK it's developed....too big for best practices, not for authors
... Do you see my problem?

MS: OK so you see these don't fit in Web developers doc...
... My opinion...
... I think as groups like HTML start to develop specs, must be a best practices implemention guide...
... I think HTML5 guide should be very developer-centric
... I don't think UA could do implementation guides for all of the formats
... UA is more consultants to all the different groups...but they need to do theri own

<oedipus_laptop> aria to msaa mappings: (based on moz dev) http://esw.w3.org/topic/PF/ARIA/BestPractices/API_MSAA

MS: Since no one else doing ARIA, WAI should do it

<oedipus_laptop> http://esw.w3.org/topic/PF/ARIA/BestPractices/API_IA2

JA: This is very detailed stuff having to do with browsers...UAWG always stayed at high level...

<oedipus_laptop> http://esw.w3.org/topic/PF/ARIA/BestPractices/API_ATK

JA: Implementation "on top" specified (in the past).

<oedipus_laptop> aaronlev, do you want something more akin to http://a11y.org/a11y-dom-apis

JA: The members of the working list aren't at this level

AL: Good admission...I don't think you are ready for this

JA: Well all the people left to work on ARIA over there
... Applies direcly but my concern is how many platforms are we going to do? Mobile other OS etc.
... How far does it go?

AL: Pretty far

KF: Unless we expand charter
... We don't cover this
... ARIA hot now, but something else will be hot in 5 years
... At that time they should be able to go to UAAG and found out what they need to be able to do with their format

AL: Not just me, take Opera developers
... CMN was on user agent before
... But what they really need is list of all HTML attributes etc.
... Since it's not there they'll reverse engineer FF etc.

<oedipus_laptop> http://www.w3.org/WAI/UA/TS/

<oedipus_laptop> (UAAG 1.0 test suite)

AL: I look at top level stuff for a but, but then focus on low level
... Maybe you could be more of a host for a community to work on things

GR: Well the mappings should be standard and Access API dependent

KF: So....not in charter now....

<oedipus_laptop> ARIA provide standard calls to accessibility APIs

KF: But middle ground could be to facilitate coordination between browser developers

AL: We put our wiki on ???...because it is neutral

KF: So I'd be comfortable with facilitation....vs. we the user agent does the deep dive...
... But needs charter changes

AL: Ultimately I'd like to have software tools to know if things are done correctly or not

<oedipus_laptop> hyper-specific documents are a danger - people often implement off of a table or checklist rather than understanding the logic and structure, which leads to "dumb" implementations

KF: But you the ARIA group would be responsible for the validators

GR: My concern is that in hyper-specific documents...people implement of table but not necessarily ....understand concepts
... ARIA is there in the middle and it is APIs to move to it

AL: Access APIs are rich enough...but the APIs per platform are different

MS: Totally agree .... each group developping technology needs to think end to end...
... So this is why I think it should go along side the other ARIA docs

GR: We are talking about access to DOM....so ARIA developers not responsible end to end....because APIs have role

AL: We do want to try to ensure that the user agents do things the same on the various platforms

GR: Violent agreement...

<AllanJ> JR: PF says it doesn't want to host. UA could host, non-rec track items. There is some presedent

<AllanJ> JR: Why UA space and not PF space

GR: Are we then responsible?
... Is this stuff formally coming out of ARIA best practices

AL: Yes

MS: Fine with UA, but as we make changes to ARIA....will change ARIA implementation guide

<oedipus_laptop> q} to say that aria is the abstract the DHTML style guide is the where the abstract meets hardware

AL: So...accessibility needs to be living breathing documents
... Take live regions for example, we'lll learn a lot

GR: Right so PF defines abstract but let's DHTML style guide to define details

KF: we are inagreement I think....
... To me PF working group still responsible for technologies they develop
... We can consult but can't be experts on all formats

<oedipus_laptop> strong plus one to KF's last statement - work in liaison with other WG to advise and counsel UA conformance/implementation/requirements

KF: So we could expand charter to host best practices for other formats...
... But not being experts in them

AC: So we are just high level things

<oedipus_laptop> also need to address ARIA for those without assistive technology (native UA support)

KF: THink so....value of high level base requirements

JA: Right

MS: Driving key accessible principles into different techs

<oedipus_laptop> http://a11y.org/kafs

<oedipus_laptop> http://a11y.org/kafs-gta

<oedipus_laptop> the second URI is generic test assertions

EJ: (Earl Johnson) in Linux foundation...we callled them generic test assertions due to differences across platforms
... Can you said it this range, can yout turn on/off beep
... Then up to UA developer to figure out how they will meet that test

JA: That's how we've operated....
... UAAG doesn't get down to platform, programming languages...

<oedipus_laptop> detailed work on wiki?

JA: But what I'm hearing today is that UAWG could provide the space

<oedipus_laptop> precedent: EO working on ARIA educational materials

GR: Think EO is working on educational stuff for ARIA

<scribe> ACTION: KF, JA, JR to Think about charter change to take into account user agent implementations...eg link to HTML5 or publishing note as in ARIA [recorded in http://www.w3.org/2008/04/17-ua-minutes.html#action01]

JA: Marc and Aaron, we've reached end of time...do you want to come back?

AL: So we are pushing format groups to produce user agent implementation techniques

GR: Yes

AL: So the requiements don't get down to APIs
... Guess I could come in, but would prefer just to review text, emaiols etc

MS: I could prob come back thurs but will talk to Kelly


Summary of Action Items

[NEW] ACTION: KF, JA, JR to Think about charter change to take into account user agent implementations...eg link to HTML5 or publishing note as in ARIA [recorded in http://www.w3.org/2008/04/17-ua-minutes.html#action01]
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.133 (CVS log)
$Date: 2008/04/17 19:12:34 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.133  of Date: 2008/01/18 18:48:51  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/techniques were/techniques where/
Found Scribe: Jan
Inferring ScribeNick: Jan

WARNING: No "Present: ... " found!
Possibly Present: AC AL Aaron_Leventhal Alan AllanJ Cantor EJ Earl_Johnson GR Gregory_Rosmaita JA JR Jan KF KFord MS Microsoft P8 aaronlev oedipus_laptop precedent
You can indicate people for the Present list like this:
        <dbooth> Present: dbooth jonathan mary
        <dbooth> Present+ amy

Agenda: http://lists.w3.org/Archives/Public/w3c-wai-ua/2008AprJun/0048.html
Got date from IRC log name: 17 Apr 2008
Guessing minutes URL: http://www.w3.org/2008/04/17-ua-minutes.html
People with action items: ja jr kf

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.

[End of scribe.perl diagnostic output]