[Minutes] 8 Aug 2002 UAWG teleconference

UAWG teleconference, 8 August 2002

[Note: This is a resend; I don't know why these didn't
make it to the list yesterday.]

Agenda announcement:
  http://lists.w3.org/Archives/Public/w3c-wai-ua/2002JulSep/0082.html

Participants: Jon Gunderson (Chair), Ian Jacobs (Scribe),
Harvey Bingham, Eric Hansen, Matt May, Rich Schwerdtfeger

Regrets: Jim Allan, Lee Bateman, Tim Lacy, David Poehlman

Previous meeting: 25 July 2002
  http://lists.w3.org/Archives/Public/w3c-wai-ua/2002JulSep/0051

Next meeting: 15 August, 2pm ET
   To coordinate implementation report.

Reference document 7 August 2002 Working Draft
   http://www.w3.org/WAI/UA/WD-UAAG10-20020807/

==========
Discussion
==========

-------------------------------------------
1. Comments on 7 August UAAG working draft?
    http://www.w3.org/WAI/UA/WD-UAAG10-20020807/

IJ: Many thanks to Eric, Harvey, and others for comments
on the document. The last call draft will be the 7 August
draft, plus the revised 6.4 (see below), and possibly
minor edits. Congratulations!

---------------------------------------
2. Revised provisions to checkpoint 6.4
    http://lists.w3.org/Archives/Public/w3c-wai-ua/2002JulSep/0087

EH: Notes seem open-ended to me.

IJ: Even with their informative status?

IJ: Provision 1 does not establish a minimal set of
     types of graphical objects. But in practice, APIs provide
     access to graphical objects. These things are done in
    practice, however. Same for provision 2.

EH: I have no objections.

JG: This is to help resolve the schism between what's on the
     screen and invalid HTML.

RS: Screen coordinates more important than relative coords.

IJ: Can you get viewport position from system and then add
     relative coords?

RS: Yes, I think so, on most systems. Experience is that
     ATs need screen coordinates.

JG: Seems like we should require screen coordinates, with
     a Note about conversion from relative to absolute.

IJ: What's clearest definition of screen coordinates?

RS: Pixel position relative to the desktop.

IJ: What about the classic cases of four physical monitors
     on one virtual screen, or four virtual screens and one
     physical monitor?

EH: "Relative to the point of origin in the
      graphical environment (e.g., with respect to the
      desktop not the viewport)."

IJ: Should we say "pixels" (i.e., specify the units)?

RS: I think it's sufficient to not say pixels. Those skilled
     in the art know what you're talking about.

RS: MSAA provides implementation experience for provisions
     1 and 2.

IJ: We have comments from Mozilla, Opera, Adobe, Microsoft,
     and IBM that these things are already done in practice,
     and from IBM and Alva (and earlier comments from AI-squared)
     that these requirements are important.

JG: Note that the proposal requires less information about
     text than the previous version. [Nobody objected to
     this change.]

Resolved:
  * Accept proposal
    http://lists.w3.org/Archives/Public/w3c-wai-ua/2002JulSep/0087
  * Accept this clarification that "coordinates" are
    relative to the point of origin in the
    graphical environment (e.g., with respect to the desktop
    not the viewport).
  * Don't mention specific units (e.g., pixels) in the requirement.

------------------------------------------------
3. Updating Implementation Report for Proposed Recommendation
    http://www.w3.org/WAI/UA/implementation/report-cr2

JG: Matt, Ian, Colin Koteles, and I met in Chicago last week.
     Lots of progress since then, including:

  3a) Evaluation report (still in early stages of development)
      to produce implementation reports from an HTML form.
      http://www.w3.org/WAI/UA/2002/08/06-evalform

    MM: You can fill out the form, activate "generate xml" at
    the end of the page. It will use javascript to produce
    an implementation report in xml in a text area. You can
    cut and paste to a file; as the eval form evolves, it will
    become easier to use. So far, it works in Mozilla. I am
    trying to make it work in other UAs that don't support
    DOM Level 2. I suggest testing it with your browser first.

  3b) New implementation report DTD (based on our experience).

  3c) Progress on UAAG 1.0 draft test suite.
      http://www.w3.org/WAI/UA/TS/html401/

JG: Tests can be incorporated in implementation reports.

JG: The hurdle before last call is converting old implementation
     reports to new version. Matt's form will assist in this
     process.

/* Discussion about evaluations, implementation reports */

JG: The evaluation report is not just to get out of CR:

  1) It's important (as are the tests) to pointing out
     to developers limitations in their implementations.
  2) It's important to back up UAAG 1.0.

IJ: Here is my approximation of how old checkpoints migrated
     to new checkpoints:
      http://www.w3.org/2002/07/ua-diffs

IJ: See also "How to evaluate a user agent for conformance
     to UAAG 1.0"
     http://www.w3.org/WAI/UA/2001/10/eval

Actions:

  HB: Create new evaluation of Opera.
      http://www.w3.org/WAI/UA/implementation/eval_win_opera6.html

  EH: Create new evaluation of Acrobat reader.
      http://www.w3.org/WAI/UA/implementation/eval_win_pdfreader5

  MM: Create new evaluation of Mozilla
      http://www.w3.org/WAI/UA/implementation/eval_win_mozilla6

  JG: Create new evaluation of GW Micro and Jaws
      http://www.w3.org/WAI/UA/implementation/eval_win_wineyes411
      http://www.w3.org/WAI/UA/implementation/eval_win_jaws402

Deadline: 13 August.

To do before last call:

  * Incorporate new checkpoint 6.4 per today's discussion.
  * Document checkpoints with low implementation experience.
  * Update implementation report summary.

IJ: I don't think we need to send formal responses to
reviewers who raised CR issues, since we raised them as
a working group during the course of evaluations. The
broader audience will have the chance to comment on
those decisions in last call.


-- 
Ian Jacobs ([email protected])   http://www.w3.org/People/Jacobs
Tel:                     +1 718 260-9447