W3C home > Mailing lists > Public > w3c-wai-ua@w3.org > October to December 1999

Raw minutes from 6 October teleconf

From: Ian Jacobs <ij@w3.org>
Date: Wed, 06 Oct 1999 13:43:08 -0400
Message-ID: <37FB8A2C.3B3FFDC1@w3.org>
To: w3c-wai-ua@w3.org
User Agent Guidelines Teleconf
6 October 1999

Present:
 Jon Gunderson
 Ian Jacobs
 Charles McCathieNeville
 Al Gilman
 Daniel Dardailler
 Mark Novak
 Harvey Bingham
 Kitch Barnicle
 Rich Schwerdtfeger
 David Poehlman

Regrets:
 Gregory

Agenda [1]
[1] http://www.w3.org/WAI/UA/1999/10/wai-ua-telecon-19991006.html#agenda


1) Review of action items:

   1.JG: Run pwWebSpeak (with Mark H.) through the guidelines. 
   2.JG: Propose techniques for rendering of frames 
         Status: Done.
   3.GG: Review proposal for techniques for accessing content. 
   4.GR: Write a proposal to address issues about spawned windows. 
   5.DP: Run Jaws for Windows through the guidelines 
         Status: Will be done for face-to-face.
   6.MR: Working on SMIL techniques in addition to SMIL access note. 
         JG: Will contact Madeleine.
   7.IJ and JG: Send a proposal to the ua list for resolution 
        of the conformance issues related to assistive technology
         Status: Done.
 
2) Announcements:
   1.No telecon on 13 October 1999 
   2.Send agenda items to UA face-to-face. 
   3.DOM 2 in last call until 8 October. 

3) Agenda items for face-to-face?
   (No input from those present).

   DP: Will Netscape be at the meeting?
   IJ: I've written Mozilla but haven't gotten a reply.
  
   Action JG: I will invite Lake Rocca to WG and face-to-face.

   [1]
http://lists.w3.org/Archives/Public/w3c-wai-ig/1999OctDec/0015.html

4) Issue #96: Issues related to Checkpoint 2.1: Mapping of user agent 
             functions to control mechanisms and
             memory demands related to sequential/direct access to 
             functionalities.
     http://cmos-eng.rehab.uiuc.edu/ua-issues/issues-linear.html#96 

   IJ: Any proposals?

   AG: The way I understand Marja's issue that is not currently
covered: it's possible for discrete access strategies to fail because
they require too many steps. There are two separate issues in terms of 
the logic:
 a) Visual/Memory
 b) Multistep/Single step

People can fail to use key access to print because the number of
keystrokes becomes a burden. 

IJ: 5 October version has requirement of single key access for
frequently used functionalities.

JG: Summarizing
  a) How can we encapsulate demands on memory in a checkpoint?
  b) Orientation issue: keyboard commands must be documented.

Action IJ: Contact Marja about writing a proposal for what should be
changed.
  
5) Additional checkpoint suggestions:

RS: 
   a) Section 6 on observe standards. There's no checkpoint that
      addresses loading Java applets into the JVM. Proposed in [2]:

   6.7 Support plug-in and virtual machine system conventions for
loading and
       running an assistive technology. For example, the Sun Java
virtual machine
       supports loading and running of assistive technologies. (priority
1)

Action JG: Add this to issues list.

[2] http://lists.w3.org/Archives/Public/w3c-wai-ua/1999OctDec/0063.html
      
6) Issue #89: Proposed changes in conformance based on 
              interoperable UA and non-interoperable UA 
    http://cmos-eng.rehab.uiuc.edu/ua-issues/issues-linear.html#89 

   PROPOSAL 1)

   IJ: I proposed splitting 1.1. I think that two important concepts
       should be separated.

   No objections to the split.

   IJ: Also related: Does a tool have to support ALL OS input device
       APIs?

   Resolved: Clarify 1.1 so that UAs don't have to support all input
      device APIs available on an OS. If you support a type of API,
      support the system standard API.

   MN: Note that MS Platform requires support for both mouse and
       keyboard APIs. The UAGL only requires keyboard API support.
       Thus, in this case, the UAGL is less strict than the MS
       platform guidelines.

   RS: Perhaps a checkpoint that says to use "most common API" for
       the system (which might be pen input, for example).

   JG: Can we put this somewhere in Guideline 1?

   RS: Can we include examples? 
  
   Action Rich: Draft a proposal for a checkpoint about using
       "common" input/output device APIs for the given system.
  
   Action Ian: Split 1.1, clarify that not all APIs required. Results
       dependent on Rich proposal.

   PROPOSAL 2)

     Ensure that the user has access to the content of
     an element selected by the user.

     IJ: I changed "content" to "structure". 

     JG: I prefer "content"
     HB: I think "structure" is also important.

     IJ: Do 3.1, 9.2, and 8.3 count?

     RS: Ensure that the user has access to the content of an
         element selected by the user programmatically, through a
         dependent user agent, or through voice input.

     IJ: So just shorten to "Ensure that the user has 
         access to the selected content."

     RS: So just remove "for dependent user agents"

     Resolved: "Ensure that the user has access to
                selected content."
               Change Note to mention programmatic access, 
               structure, and that cell content can spoken, etc.

     
   PROPOSAL 3) Checkpoint 3.3. This checkpoint should not be 
               for dependent user agents only. Refer to issue 84.

     Resolved: For all user agents.


   PROPOSAL 4) Checkpoint 8.3. Proposed for all user agents and adding 
               a Note.

    IJ: Any reason to single out tables?
   CMN: I don't think there's any reason. Tables are two-dimensional,
        so they pose an additional problem of access. The critical
        part is access to relevant information, which includes table
        structure.
    KB: If there's a checkpoint for table navigation, then we can drop 
        this.
    DP: We definitely need to emphasize table navigation. There are
probably
        other multi-dimensional elements and we could regroup them. I
        don't want to bury a table navigation checkpoint.

    IJ: What should the priority be? 

     Resolved: Leave 8.3. Make for all user agents. Add Note proposed
     by Ian.


    PROPOSAL 5) Checkpoint 9.2. The current text:

         Provide the user with information about the number
         of viewports.

    IJ: What's the goal here? Is it actually the number of viewports?

    IJ: Change to "Provide a list of open viewports (including
        frames)."?

    IJ: "Provide information about viewport structure and focus."
        E.g., the structure of a frameset.

    Resolve: Delete based on new 9.3

    PROPOSAL 6) Checkpoint 9.3

    Proposed: Allow the user to view a document outline
              built from its structural elements 
             (e.g., from header and list elements).
 
    IJ: For all user agents. Does "page source" count?
  
    Consensus: Page source doesn't suffice.

    Resolved: Make 9.3 for all user agents. Mention frames explicitly.

    PROPOSAL: Make Guideline 7 for all open standards.         

    No objections to moving 6.6 to Guideline 7.
  
    Action Ian: 
       1) Guideline 7 more than just about w3c technologies
       2) Add checkpoint 6.6

    ISSUE) What do we do about 9.9 and 9.10 (cell headers and table
      dimensions)? 

    JG: I'd like to keep 9.9.
    DP: I'd like to keep 9.10.

    Action Ian: Propose a checkpoint like the ones for form about
       table summary information.

Adjourned
Received on Wednesday, 6 October 1999 13:43:14 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:49:24 UTC