Raw minutes from 11 October 2001 UAWG teleconf

11 October 2001 UAWG teleconference

Agenda announcement: 
Participants: Jon Gunderson (Chair), Ian Jacobs (Scribe), 
Lee Bateman, Tim Lacy, David Poehlman, Harvey Bingham, Jim Allan,
Mickey Quenzer, Eric Hansen

Regrets: Jill Thomas
 Absent: Gregory Rosmaita, Rich Schwerdtfeger, Denis Anson

Previous meeting: 4 October 2001

Next meeting: 18 October
Next face-to-face: 22-23 October

Reference document 12 September Candidate Recommendation:

1. FTF meeting details

Summary of connectivity:

Confirmed video bridges:
 - Austin: Jim Allan
 - Boise: Lee Bateman
 - Unknown: Denis Anson
 - No: MIT (Harvey to attend by phone)

Summary of ftf attendees: Tim, Jon, Ian, Wendy, Stephanie Potter,
David Poehlman, (RealNetworks 22nd), Mickey, plus whoever at

2. FTF agenda

Refer to agenda:

JG: Try to get Opera, Netscape to attend the "securing
developer commitment" slot.

Action IJ: contact Opera, Netscape.
Action JG: contact AT developers

JG: The implementation report will be a basis for discussion:

Refer to summary of checkpoints that need attention:

JG: Topics:

  - How do we test accessibility features?
  - How do we facilitate implementation?

3. How to evaluate a UA

IJ: I imagine a "high-level" document to set expectations about
pitfalls and a framework for evaluating software. The document will
include links to other resources (e.g., test suites).

JG: The how-to document might include:

 a) Framework for doing an evaluation. What systematic approach
    can be taken.
 b) Hints on where to look for information (e.g., documentation,
    developer claims, etc.)
 c) How you can identify which checkpoints you're trying to
 d) Examples based on real checkpoints.
 e) How to report the results (using AERT?)

IJ: When this document is ripe, UAAG 1.0 should reference it. I don't
think the how-to document should be part of UAAG 1.0 (mostly so that
it can evolve on its own).

4. Test suite update

Action IJ: Invite Lofton Henderson to UAWG ftf re: test suites.

/* IJ summarizes  test suite framework */

MQ: Can this be automated 100%?

IJ: No, comparisons with expected results probably not

MQ: I'd like a mechanism that keeps track (persistence) of what I've
already evaluated.

JG: How much of this mechanism exists already?

IJ: I hope to reuse the xslt framework (not yet available, but close).

LB: A requirement may contain several requirements. For example, a car
should be red with a black interior and white roof. These need to be
identified as distinct requirements. 

IJ: Yes, e.g., when a checkpoint mentions both audio and video, need
to have two tests.

LB: Need to be careful in reporting mechanism that the granularity is
sufficient: a UA may satisfy 4 of 5 bullets. Need people to be able to
see that a piece of a checkpoint has not been satisfied.

EH: This framework sounds like a good approach.

JG: Tim, will this help developers implement the guidelines?

/* LB leaves */

HB: Tim, do you have guidelines for preparing MS test suites?

TL: Nate would be able to discuss this.

IJ: Good to talk about tools, processes, frameworks at meeting.

IJ: I haven't even talked about processes for contributing test
cases. The XSLT framework includes this bit, but I've been more
interested in the encoding first, contributions second.

JG: It's time consuming to convert text email to a specific XML
format (such as the implementation report format).

Closed action items

1.ij: find out about availability of videoconf equipment availability at

5.ja: find out if ja and rs can meet in the same place in austin

Open action items

3.jg: talk about jg's tool and earl integration at wai cg.

4.jg: review netscape version 6.x 

6.ja: work on a comparison document of uaag and section 508 and format
in uaag style.
 Refer to

JA: Pending.

8.tl: review initial implementation report for ie and comment

TL: Pending.


7.gr: contact dolphin 

Ian Jacobs (ij@w3.org)   http://www.w3.org/People/Jacobs
Tel:                     +1 718 260-9447

Received on Thursday, 11 October 2001 15:10:42 UTC