Re: Questions on ATAG2.0 Guidelines?

Thanks Tim, hopefully this gets us rolling again. My comments are in-line:

Tim Boland wrote:
> In conversation with Jan Richards on February 22 regarding getting a 
> draft ATAG2.0 techniques ready for publishing to the AUWG list, I have 
> noted the following items as (from the conversation) as possibly needing 
> addressing in the ATAG2.0 Guidelines (NOTE: these items all refer to 
> text included or missing from the current ATAG2.0 Guidelines [1]:
> 
> (1) in SC2 of A.2.5, change "styling" to "presentation" - motivation is
> "presentation" seems a more generically used
> word in W3C (used in contrast with "structure" in many discussions)?

Agreed.

> (2) The "Rationale" for checkpoint A.2.9 states that "authors with 
> disabilities need to
> have access to a preview", but I thought previews were "optional" in 
> authoring tools that
> might seek to conform to the ATAG2.0 Guidelines (that is, an authoring 
> tool didn't need to have a preview function in order to claim 
> conformance to the ATAG2.0 Guidelines)?   Is there a consistency issue 
> in this regard?

The current wording is:

Rationale: The workflow of many authoring tools includes periodically 
checking a *preview* of how content will appear to *end users* in a 
browser. Authors with disabilities need to have access to a preview so 
that they can check all aspects of their work (i.e. not just the 
accessibility of that work). For this reason the preview needs to be as, 
but not more, accessible than the target browser(s).


I propose the following wording change:

Rationale: *Preview* features are provided in many authoring tools 
because the workflow of authors often includes periodically checking how 
content will appear to *end users* in a user agent. In order to enable 
authors with disabilities to follow the same workflow as other authors, 
they must have access any preview features that exist.


> (3) Note 1 of A.2.9 stated "This requirement serves, for the preview 
> features only, in lieu
> of the other user interface accessibility requirements in Part A". This 
> sentence seems unclear
> to me, so perhaps some clarification is needed as to exactly what is 
> meant by the "other"
> requirements - which "other" requirements?

The current wording is:

Note 1: This requirement serves, for the preview feature(s) only, in 
lieu of the other user interface accessibility requirements in Part A.

Note 2: In addition, it is expected that the operation of the preview 
accessibility features will be constrained by the accessibility and/or 
completeness of the content. For example, an incomplete document may not 
be renderable by the preview.


I propose the following wording change:

Note 1: Authors, including those with disabilities, will not be 
well-served if preview features diverge too much from the actual 
functionality of available user agents. Therefore, preview features are 
exempted from necessarily having to meet all of the other requirements 
in Part A of this guidelines document, if they meet this checkpoint.

Note 2: It is understood that the accessibility of the content display 
of a preview will be negatively affected if the content being rendered 
is inaccessible or incomplete. For example, if markup tags are missing, 
content may not be rendered properly.


> 
> (4) Somewhere in the ATAG2.0 Guidelines, should it be stated that a 
> "preview" is basically meant a "user agent"?

The current Definition is:

A view of the content that is intended to show how it will appear and 
behave in a browser. Content within a preview is not editable.


I propose the following wording change:

A non-editable view of the content that is intended to show how it will 
appear and behave in a user agent (e.g. launching a browser to render 
the content).



> (5) What does "same accessibility features" exactly mean as used in 
> A.2.9 SC1 Part (b)?
> Perhaps clarification would help?  What specifically are these 
> "accessibility features"?

> (6) The text after "the following must be true.." in A.2.9 SC1 Part (b) 
> seems
> convoluted and confusing to me. Perhaps this text should be dropped, as 
> it may
> represent an "edge case/situation" which may never actually occur in 
> practice (or occur rarely)
> in the context of authoring tools claiming conformance to the ATAG2.0 
> Guidelines?
> NOTE: Dropping this text might simplify A.2.9 SC1 Part (b)?

You're right that this is somewhat convoluted. Perhaps, given that this 
is a P2, we can tighten up the conditions a bit. Also I'd like to be 
more clear overall that the means for returning from the preview must be 
accessible (in case the preview isn't).

The current wording is:

(1) If a *preview* is provided, then:
(a) The author must be able to choose an external browser to perform the 
preview
(b) or, the preview must provide the same accessibility features as a 
target browser (which must be identified in the conformance profile) 
that is being emulated and the following must be true:
Any other part of the user interface other than the content being 
previewed must meet the other requirements of *Part A*.
A means of exiting the preview must be provided that meet the other 
requirements of *Part A*.
(c) or, the preview does not claim to emulate a specific browser, in 
which case it must meet all of the requirements of *Part A*. In other 
words, *Note 1* does not apply.


I propose the following wording change:

(1) If a preview feature is provided, then a method of returning from 
the preview must be provided that meets all of the checkpoints in *Part A*.
(2) If a *preview* is provided, then either:
(a) The author must be able to choose an external user agent (e.g. a 
third-party browser) to perform the preview
(b) or, if the preview emulates a particular user agent (*identified in 
the conformance profile*), then the *content display* of the preview 
must implement the same accessibility features (e.g. keyboard 
navigation, API support, etc.) as that user agent and a list of the 
implemented accessibility features must be published with the 
*conformance claim*).
(c) or, if the preview does not emulate the *content display* of a 
particular user agent, then the *content display* of the preview must 
meet all of the checkpoints in *Part A*.



Thoughts?

-Jan

Received on Thursday, 23 February 2006 19:21:07 UTC