W3C home > Mailing lists > Public > w3c-wai-ua@w3.org > April to June 2001

Raw minutes from 25 May 2001 UAWG teleconference

From: Ian Jacobs <ij@w3.org>
Date: Fri, 25 May 2001 15:59:11 -0400
Message-ID: <3B0EB98F.4644E263@w3.org>
To: w3c-wai-ua@w3.org
25 May 2001 UA Guidelines Teleconference

Agenda announcement:
 http://www.w3.org/WAI/UA/2001/05/wai-ua-telecon-20010525.html#agenda

Minutes of previous meeting 24 May:
 http://lists.w3.org/Archives/Public/w3c-wai-ua/2001AprJun/0211

Next meetings: Weds 30, and Thurs 31
  Note: At the 31 May teleconference, assistive technology
        developers will be discussing the current document
        and in particular the API requirements.

Present: 
 Jon Gunderson (Chair), Ian Jacobs (scribe), Harvey Bingham
 Gregory Rosmaita, David Poehlman, Denis Anson, Jim Allan,
 Tim Lacy.

Absent:  Mickey Quenzer, Rich Schwerdtfeger
Regrets: Eric Hansen

Reference document 25 May 2001 Guidelines:
 http://www.w3.org/WAI/UA/WD-UAAG10-20010525/

----------
Discussion
----------
----------
#503: 9.5: Clarification requested: does this mean that onfocus events
are not triggered?
http://server.rehab.uiuc.edu/ua-issues/issues-linear-lc3.html#503
----------

JG: I note that there's bad implementation of "onchange" (since it
should only be triggered once focus leaves an element).

JG: I think that the only HTML events at issue here are onchange,
onfocus, and onblur.

IJ: The checkpoint says "explicitly associated input device event
handlers".

Resolved:

 - 1.2 is only about input device event handlers only.

Resolved:

 - 9.5
   a) This is not about input device events.
   b) This checkpoint should say that this is about
      any events related to the change of focus (either
      set or remove). This may include "onchange",
      but probably should not by specification.
   c) Change brief description at beginning of checkpoint.

----------
#504: 7.2, 7.3, 7.7, 12.3: Default values inherited from
operating system.
http://server.rehab.uiuc.edu/ua-issues/issues-linear-lc3.html#504
----------

Resolved:

 - When default values are inherited through the operating
   system settings, the user agent is not required to satisfy
   the default settings requirements.

/* On a related issue of whether default values need
   to be "accessible" */

DA: This is important in shared environments, where configurations
can't be changed.

DP: If the defaults make it impossible for some users to select
profiles or otherwise configure the user agent, then the defaults
don't help.

Action IJ: Mention "themes" in the section on configuration files
(along with profiles, init files).

----------
#505: Is single-key mode a sufficient technique to satisfy 11.4 
      single key requirements?
http://server.rehab.uiuc.edu/ua-issues/issues-linear-lc3.html#505
----------

Resolved:
 - Single key mode would satisfy 11.4: You enter single
   key mode, and thereafter you have single key bindings.
 - User should be able to enter/leave single key mode
   with a single key binding.

TL: This would probably be something done at the OS
level.

----------
#506: 4.1., 4.2, 4.3: How does one value work when different
components render different content?
http://server.rehab.uiuc.edu/ua-issues/issues-linear-lc3.html#506
----------

IJ: Single global configurations don't readily cross 
plug-in boundaries.

TL: Global settings are supposed to be global. If a plug-in
doesn't pay attention to the global setting, that's an issue
for the plug-in.

IJ: I think that we said "global" as opposed to "element level".
"Global" was not put in there to satisfy an accessibility 
requirement. I think that the idea is that:

 - 4.1: The user agent must allow the user to configure the size
   of all text.
 - A single global setting that does this would be sufficient
   to satisfy the requirement. The user agent may have to 
   satisfy this requirement with more than one setting, e.g.,
   in the case of support for N different formats, some of 
   which are rendered by plug-ins.

Action IJ: Send a proposal to address the "global configuration"
issue: it may be several settings.

----------
#507: 
If no global background color, does checkpoint apply if user agent can
do on a regional basis? 
http://server.rehab.uiuc.edu/ua-issues/issues-linear-lc3.html#507
----------

Resolved:

 - Checkpoint 4.3 does apply to all regions (same for 4.2, 4.2). The
 format may not allow background to be set at a global level, but it
 doesn't matter to the user: the background color has to be "color X"
 everywhere.

----------
#508: 4.5: Require clarification - is fast playback required, or just
the ability to jump forward in time? 
http://server.rehab.uiuc.edu/ua-issues/issues-linear-lc3.html#508
----------

Resolved:

 - Fast playback is not required.
 - Fast playback should be implemented (technique).

-------------
#509: 6.1, 6.2: P1 to provide access to content (e.g., in raw form),
DOM either P2 or alternative
http://server.rehab.uiuc.edu/ua-issues/issues-linear-lc3.html#509
-------------

IJ: So, is access to the raw content sufficient?

DA: If you just give access to raw content, you offload all of the
burden to the AT, for whom it would be equally difficult to do the
work. 

TL: You don't want to do the same thing twice. If the UA
is parsing anyway, you don't want to do things twice.

IJ: The biggest problem I see is incremental updates based
on changes to content in the user agent. This would be
very costly, and the AT would continually lose a handle 
on where the user's point of regard is.

/* TL leaves. TL will be gone for next two weeks. */

JG: I think another issue is how assistive technologies
will know what to do with "generic" XML in general; how
they will render the information or interpret it.
[For discussion at the 31 May teleconference.]

DA: At the US federal govt. level, they say in their guidelines that
cost alone is not a sufficient reason not to do something (it's
against US law).

Resolved:

 - No change to the document. Raw access to content is
   insufficient because of burden it places on ATs.
 - Cost has not been a criterion for establishing
   user needs.

-------------
#510: Conformance: How to observe OS conventions when building
       a cross-platform user agent? 
http://server.rehab.uiuc.edu/ua-issues/issues-linear-lc3.html#510
-------------

DA: In general, this is done through a hardware abstraction
layer. There has to be custom code that talks to the hardware.

IJ: Here's an example: what if there's a standard help tool on a
particular platform, but you want to provide your help as HTML?

IJ: In the case of APIs, we said that interoperability was most
important, except if broken.

DA: Users have expectations about behavior. If you put a Windows-like
application on the Mac, you will surprise users (cognitive overload).

DA: Our document is intended for users; the developer's point is
from the developer's perspective.

IJ: Seems to me that 7.3 is really the only debatable checkpoint (as
the others seem really critical).

DA: It is an issue for users when software functions differently from
the other software on your system.

IJ: Here's what section 508 says:

  "Applications shall not disrupt or disable activated features of
  other products that are identified as accessibility features, where
  those features are developed and documented according to industry
  standards. Applications also shall not disrupt or disable activated
  features of any operating system that are identified as
  accessibility features where the application programming interface
  for those accessibility features has been documented by the
  manufacturer of the operating system and is available to the product
  developer."

IJ: This is different. We say "implement them", and they say "don't
disrupt accessibility features."

HB: The 508 people wanted to make their requirements more specific
then ours.

DP: There is some history that points to using the environment to help
users use tools. When a bunch of assistive technologies first came
out, they didn't follow operating system conventions. They have
evolved in that manner, however. E.g., HPR 3.0 requires lower
learning curve than HPR 2.5.

JG: What happens if I implement a user agent in Java?

JG: WG seems to be saying that if you do cross-platform, you need
to provide at least one configuration that aligns with the 
conventions of the operating environment.

DA: Like Microsoft Word lets you have a Word Perfect interface.

Resolved:

 - No change: conventions are important at a P2 level.

-------------
#511: What is definition of applet? 
http://server.rehab.uiuc.edu/ua-issues/issues-linear-lc3.html#511
-------------

Resolved:

 - Accept definitions of applet in plug-in in 25 May draft.

-----------------
Open Action Items
-----------------

3.IJ: Coordinate usability testing of the guidelines (JRG volunteers to 
be one of the testers).
  Source: http://lists.w3.org/Archives/Public/w3c-wai-ua/2001JanMar/0137

4.IJ: Revise proposal to address Issue #474.
Source: 
http://lists.w3.org/Archives/Public/w3c-wai-ua/2001AprJun/0164

7.RS: Send pointer to information about universal access gateway to the 
WG.
  Source: http://lists.w3.org/Archives/Public/w3c-wai-ua/2001JanMar/0258

8.GR: Review event checkpoints for techniques

9.GR: Rewrite different markup (list of elements) that 2.9 applies to, 
for clarification.

-- 
Ian Jacobs (ij@w3.org)   http://www.w3.org/People/Jacobs
Tel:                     +1 831 457-2842
Cell:                    +1 917 450-8783
Received on Friday, 25 May 2001 15:59:16 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 October 2009 06:50:50 GMT