W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > July to September 2001

Notes and relevant threads to the baseline capabilities/user agents clause discussion

From: Wendy A Chisholm <wendy@w3.org>
Date: Tue, 03 Jul 2001 20:38:30 -0400
Message-Id: <4.2.0.58.20010703160236.00b322a0@localhost>
To: w3c-wai-gl@w3.org
I've gone through the archives looking for good tidbits of info related to 
the until user agents clause/baseline capabilities discussions.  Here they are:

http://lists.w3.org/Archives/Public/w3c-wai-gl/2000OctDec/0186.html
jason says,
The WCAG Techniques would document (1) which standard or specification, 
including the version number, each technique relies on; and (2) whether 
there is known to be an implementation of the relevant feature or features 
that would enable the technique to be applied in practice. Details of what 
software packages implement a given feature could also be supplied. Content 
designers would then be free to choose the extent of their support for 
backward compatibility, and when to adopt newer standards and technologies.

http://lists.w3.org/Archives/Public/w3c-wai-gl/2000OctDec/0212.html
Kynn considers it harmful to identify baseline capabilities.

http://lists.w3.org/Archives/Public/w3c-wai-gl/2000OctDec/0151.html
Matt May's OS baseline. Considered best in thread.


http://lists.w3.org/Archives/Public/w3c-wai-gl/2000OctDec/0148.html
Charles's original message. He proposes:
1. There must betwo such products available for each of - Windows 95 - 
Windows 2000 - MacOS 8+ - Unix (must include linux)

2. They must be known to be usable with at least two speech output systems, 
including one free one where that is available.

3. They must work with standard keyboard modifications (including modified 
keyboards) and one voice input system.

4. The products must have been available for at least 6 months.

Note that this can apply to partial features of a technology. For example, 
it might be possible to state that CSS font and colour control are 
sufficiently implemented, but CSS behaviour and layout control are not.

----
Other comments:
- we must reach for what 80% of the population is using, and try to get the 
other 20% as well.
- In some countries, we are using the latest and greatest machines while in 
many others (and even in larger companies and poorer neighborhoods in the 
U.S.) we are using much older machines.  In some countries, the cost of a 
computer is several months of salary therefore using older, refurbished 
computers is the only way to go.  Although, the use of some of these older 
computers sometimes costs a lot to get them online.

In a different thread, Len suggests a separate document that focuses just 
on accessibility as a function of the user agent.  However, how is this 
related to UAAG?  It's a strange intersection between them.  Charles argues 
that it is up to the policy makers to track the individual user agents and 
their support.  His note reminded me of another aspect of this problem: 
differences in localizations. The baseline can't be based on just 
American/English implementations.  CMN also suggests that minimum 
requirements/baselines are defined at the technology level.  Len retorts 
that he thought people would be able to determine conformance at the 
guideline/checkpoint level not technique.  He also wonders if D is required 
if longdesc is not what the author chooses to use.
http://lists.w3.org/Archives/Public/w3c-wai-gl/2000OctDec/0769.html

In response, Jason suggests that techniques be divided into categories with 
dates and stored in a database so that searches could be performed on those 
that have been implemented for over 2 years (e.g.).
The categories he suggests:
a. Not known to have been implemented;
b. Known to have one implementation.
c. Known to have multiple, consistent implementations.
d. Known to have multiple implementations, subject to consistency problems 
(i.e., incompatible implementations)
e. Known to be incompatible with assistive technologies
http://lists.w3.org/Archives/Public/w3c-wai-gl/2000OctDec/0781.html

Charles modified the categories a bit:
Except that I would look at doing it in terms of collecting known 
implementations and marking them as "rare (e.g. only available for a couple 
of Operating System", "normal (e.g. available on half a dozen different 
platforms)" and "conflicting (e.g. interoperability problem with a 
standards-conformant implementation)" and note what languages they work in. 
Then we can just deduce the data for a given date...
http://lists.w3.org/Archives/Public/w3c-wai-gl/2000OctDec/0782.html


----
other issues

1. conformance scheme ala Kynn Bartlett: 
http://lists.w3.org/Archives/Public/w3c-wai-gl/2000OctDec/0927.html

2. Device independent authoring ala Len
http://lists.w3.org/Archives/Public/w3c-wai-gl/2000OctDec/0849.html

--
wendy a chisholm
world wide web consortium
web accessibility initiative
seattle, wa usa
tel: +1 206.706.5263
/--
Received on Tuesday, 3 July 2001 20:29:42 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:11 GMT