RE: Implications of proposed baseline definitions

Tim expresses concern about the implications for conformance of what
Jason described as writing "'hooks' to baseline" into the success
criteria.

If I understand correctly what Jason meant there, these "hooks" consist
either of modifications so that the success criteria describe what we
have been calling "functional outcomes" instead of specifying methods of
achieving those outcomes; and possible new success criteria that will
ensure that content includes features upon which user agents depend in
order for thecontent to work in a properly accessible way.  As far as I
know, there are no plans for success criteria to refer explicitly to
"baseline" as such.

And we have been working hard to ensure that the "style" of the success
criteria, guidelines, and principles is consistent.

Principles are statements about what "must" be the case in order for
content to be accessible.  Guidelines are written in the imperative
mood, and instruct/advise authors about what they should do to make
content accessible.  Neither principles nor guidelines, in and of
themselves, are testable.  Success criteria, on the other hand, are
written as testable statements-- assertions that are either true or
false for specific content.

I hope this addresses the concerns that Tim raised (I'm sure he'll let
us know if it doesn't1<grin>).

John



-----Original Message-----
From: w3c-wai-gl-request@w3.org [mailto:w3c-wai-gl-request@w3.org] On
Behalf Of Tim Boland
Sent: Wednesday, May 11, 2005 8:25 AM
To: w3c-wai-gl@w3.org
Subject: Re: Implications of proposed baseline definitions


I am concerned about the possible implications of including such
baseline-related "hooks" directly into some (but not all?) of the
normative 
success criteria,
particularly in relation to QA SpecGL Requirement 07 ("use a consistent 
style for
conformance requirements and explain how to distinguish them" [1]) and
Requirement 08 ("indicate which conformance requirements are mandatory,
which are recommended, and which are optional" [2]).  I am also
concerned about
use of the word "implicitly" (in excerpted text following) regarding 
conformance, conformance model and testability.  Perhaps I am 
misunderstanding, but have all the implications of this (possibly
major?) 
change at this point
to the wording of success criteria been considered?

[1]: 
http://www.w3.org/TR/2005/WD-qaframe-spec-20050428/#consistent-style-pri
nciple
[2]:
http://www.w3.org/TR/2005/WD-qaframe-spec-20050428/#req-opt-conf-princip
le


>2. We need to introduce "hooks" into the guidelines that make explicit
>reference to the baseline. A draft of this has already been written,
but
>it has to be incorporated into the success criteria in all those places
>where they rely implicitly on the baseline. In the past, this hasn't
>been made plain in the document, in part because we didn't have a firm
>strategy for dealing with these dependencies; but now that the concept
>of baseline has solidified we need to make sure the success criteria
>reflect it.

Received on Wednesday, 11 May 2005 17:32:10 UTC