Response to BFs suggestions re: AUWG Poll #1: 10 September 2007

This email attempts to address Barry's suggestions in 
http://lists.w3.org/Archives/Public/w3c-wai-au/2007JulSep/0048.html, but 
IMPORTANTLY I still need another response to Poll #1 before I can 
process the results that both Barry and Roberto have accepted without 
change.

Here's Poll #1 again:
http://lists.w3.org/Archives/Public/w3c-wai-au/2007JulSep/0047.html

I've just kept the issues that Barry made suggestions for:

> -----
> 
> Proposal 3: The "modified" text in "Relationship to the Web Content
> Accessibility Guidelines (WCAG)"
> http://www.w3.org/WAI/AU/2007/WD-ATAG20-20070821/WD-ATAG20-20070821.html#Relationship-To-WCAG
> 
> Response: Accept the proposal (although it does not deal well with the 
> situation that someone chooses something other than a WCAG as the 
> guidelines for accessibility)

Good point Barry...I propose the following text instead (and withdraw 
this question from poll #1):

ATAG 2.0 relies upon *Web Content Accessibility "Benchmark"* documents 
to precisely specify what an evaluator interprets "accessible Web 
content" to mean in the context of an authoring tool and the Web content 
technologies that it produce and/or is implemented using. The 
recommended reference for the benchmark is the W3C-WAI Web Content 
Accessibility Guidelines (See *Note on other Accessibility Standards* 
[http://www.w3.org/WAI/AU/2007/WD-ATAG20-20070821/WD-ATAG20-20070821.html#other-standards])
due to the quality of the document and the process under which it was 
developed.

The Web Content Accessibility Guidelines (WCAG) is the W3C-WAI 
Recommendation that defines requirements for making Web content 
accessible to a wide range of people with disabilities. At the time of 
publication, version 1.0 of WCAG is a W3C Recommendation *[WCAG10]*, and 
a second version of the guidelines is under development *[WCAG20]*. The 
evaluator of an authoring tool may select (and record in the conformance 
profile) either version of WCAG. However, developers should give 
consideration to the following when deciding which WCAG version to use 
in a product:

[3 bullet points unchanged]



> -----
> 
> Proposal 5: Definition of "authoring session "
> http://www.w3.org/WAI/AU/2007/WD-ATAG20-20070821/WD-ATAG20-20070821.html#definitions
> 
> Response: Accept the proposal with the changes (then specify changes)
> 
> suggest "no further opportunity to make changes." --> "no further 
> opportunity to make changes without starting another session."


Good change. I'll make the change and withdraw this item from Poll #1.



> -----
> 
> Proposal 6: Modified definition of "authoring tool", "view" (which would
> then contain "editing view" and "preview"), and "authoring tool user
> interface"
> http://lists.w3.org/Archives/Public/w3c-wai-au/2007JulSep/0040.html
> PLUS see below for modification to "authoring tool"
> http://lists.w3.org/Archives/Public/w3c-wai-au/2007JulSep/0043.html
> 
> Response: Accept the proposal with the changes (then specify changes)
> 
> WRT the PLUS section.  
> ATAG 2.0 defines an "authoring tool" as any software, or *collection of
> software components*, that *authors* use to create or modify *Web
> content* for USE BY OTHER PEOPLE.
> 
> (I'm aware and ok with the fact that this covers email systems that send
> "Web content")
> 
> On the (...) comment, to me, an email (or similar, say wiki) system 
> should not be required to (but it is allowed to) be responsible for 
> assisting the author in prompting, evaluating or fixing email content 
> that was included from other sources (including forwarded email or 
> attachments). At most it should only be held accountable for the actual 
> content added by the author issuing the "send" request.
> 
> Although the AUWG clearly would include an email system as an authoring 
> tool (especially one like gmail), I'm not sure all email system vendors 
> would agree. For example, is an email system that sends over the 
> internet but uses private formats (vs say HTML) to encode the mail so 
> that only the same type of system can receive the mail and render it 
> considered to be an authoring tool (Lotus Notes can work in this mode)?

The thing about email systems is that simply by entering the email 
address of a Web-archived listserv they become authors of Web content 
that might be viewed by a very large number of people. Even pressing 
"Replying All" can send to content to a large number of people.

So it seems to me that the the basic issue is that if a software lets 
you determine important details of other people's interactive 
experiences, then they are authoring tools. (On the other hand, if a 
tool lets me modify my own view of the Web - e.g. a personalized portal 
- it does lots of "authoring-like" things, but would not be an authoring 
tool by my proposed definition).

That said, ATAG 2.0 compliance doesn't mean an in-the-author's-face 
experience, it just means the supports need to be in place if the author 
wants to use them.

Regarding private formats: If the format is fairly basic (e.g. rich text 
and images) that will make conformance relatively easy. Conversely, if 
they throw in all sorts of ability to introduce accessibility problems, 
greater effort to meet ATAG 2.0 seems a natural result.

Cheers,
Jan

Received on Monday, 17 September 2007 20:30:20 UTC