W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > October to December 2002

Sun's Review of WCAG 2.0 Working Draft

From: by way of Wendy A Chisholm <Earl.Johnson@Sun.COM>
Date: Sun, 27 Oct 2002 23:00:21 -0500
Message-Id: <5.1.0.14.2.20021027230002.02420ab0@localhost>
To: w3c-wai-gl@w3.org



Dear WCAG 2.0 Team,

Thank you for all the work you've put into the writing of WCAG 2.0 so
far, it is an improvement in that it's easier to read and figure out
what aspect of accessibility a given checkpoint addresses (e.g.
navigaton).  Thanks also for the extra time you granted Sun for
submitting this review.

Earl Johnson
Sun Microsystems

-------------
		Sun Microsystems comments on WD-WCAG 2.0
			based on 8/22/02 draft


GENERAL COMMENTS
1. The first informative information that would be helpful seeing in
    all checkpoints is a one short paragraph answer to "what does this
    do and why am I doing this?" (something along the lines of the Note
    for checkpoint 1.5 is a good example of what this intro should
    contain and the max length it should be). This information would
    potentially be better placed right after the checkpoint versus
    placing it in the checkpoint's informative section.

2. When reading the WCAG 2 what stands out is the checkpoint's
    informative box versus the chckpoint and what each level requires.
    The checkpoint wording should stand out more so that when a reader
    skims it their eye is drawn to the actual checkpoint first.

3. While WCAG 2.0 appears to be an overall improvement over 1.0, how to
    meet the requirements its levels call for are less technology
    specific than 1.0 (i.e. there is more absraction in 2.0). To
    counteract some of this technology abstraction it would be helpful
    if if checkpoint had at least 2 examples provided in he illustrative
    section.


SPECIFIC COMMENTS
1. In the Note from "Overview of Design Principles": "A structured
    database or metadata collection where the data is intended for use
    by another machine and thus requires no interface lies outside the
    scope of these guidelines."
	a. The other machine could be software, how about changing
	   "machine" to "service"
	b. Should "user" be added to "no interface"?

2. In Checkpoint 1.1
	a. Success Criteria 2: What does "descriptive label" mean? Is
	   it "provide a longer description", more, or otherwise? The
	   subbullet doesn't give guidance, having it would be helpful
	   (e.g. point to a specific informative Example).
	b. "Benefits" bullet#2: Suggest dropping this "or have it
	   translated and presented as sign language," the text
	   "reading the text" makes the point.
	c. "Examples" #4: Change "described in words" to "read"

3. In Checkpoint 1.2
	a. "Success criteria", #5 in "Minimum": The way it currently
	   reads suggests there are 2 conditions #5 is meant to cover.
	   Should the sentences be split into separate bullets or does
	   this rewrite capture #5's point - "if the Web content is
	   real-time non-interactive video (e.g. a Webcam of ambient
	   conditions), provide an accessible alternative that achieves
	   the purpose of the video and that conforms to checkpoint
	   1.1, or a link is provided to content elsewhere which
	   conforms to checkpoint 1.1 (e.g. a link to a weather Web
	   site)."

4. In Checkpoint 1.3
	a. #1 in "Minimum Level": What is the text "conveyed through
	   presentation formatting" refering to or talking about? An
	   example would be helpful, e.g. "conveyed through
	   presentation formatting, e.g. video, ..."

5. In Checkpoint 1.5
	a. #1 in Level 2 is more stringent than #1 Level 3. Should these
	   be swapped?
	b. The "Note" is informative. Should it be renamed and placed
	   into the informative box? Or maybe make it the opening
	   explanatory statement for the checkpoint (see general
	   comment #1)?
		- Please also consider adding a short
		  description/reasoning to the beginning of all other
		  checkpoints.

6. In Checkpoint 2.1
	a. Suggest rewording "Minimum Level" to "content uses only
	   event handlers that are designed so that, at a minimum, they
	   are operable through character input." The current wording
	   can be interpreted as meaning an event handler can not also
	   support mouse input (i.e. that it must be a keyboard event
	   handler only).
	b. From Level 2: what is a "more abstract event"? Same question
	   applies if the wording for this should be "more abstract
	   event handler".

7. In Checkpoint 2.3
	a. On #3 in Level 2: if it is kept as a criteria consider
	   changing it to a Level 3 criteria.

8. In Checkpoint 3.1
	a. For Minimum Level: shouldn't table structure (row, column)
	   and basic role (data, header) also be required?
	b. The Note is a good recommendation (guidance), consider
	   moving it to the informate portion of the checkpoint and
	   renaming it to Recommendation (or Guidance). Or maybe the
	   Note should be in "Minimun Level" or "Level 2" (which would
	   address comment a. above).

9. In Checkpoint 3.2
	a. From Level 3: it is unclear what this is talking about:
	   "alternate presentation formats are available to vary the
	   emphasis of the structure." Are the Examples in the
	   informative section all examples of this? Consider defining
	   what "alternate presentation formats" means in the
	   informative section.
	b. On Note #1: it is guidance or a recommendation. Consider
	   renaming it to Guidance or Recommendation and placing it in
	   the informative section.
	c. On Note #2:l the note is really a requirement, should it
	   state it is?
	d. On Note #2: it refers to Checkpoint 2.2. What is the
	   connection between 2.2 (timed response) and 3.2 (emphasize
	   structure)? Should it be pointing at 5.3 instead?
	e. On Note #2: see 9b.'s recommendation.

10. In Checkpoint 3.5
	a. The informative Benefits: the Note looks like a benefit.
	   Consider changing it from a Note to a Benefit bullet.
	b. Sun answer to Reviewers Note question: expand and clarify
	   them to acheive consistancy in style with how all other
	   Examples are presented.

11. In Checkpoint 3.6
	a. In #1 Level 3: is this saying "use a combobox whenever
	   possible," or more, or different. It's hard to tell what needs
	   to be done to what to fulfill this requirement.
	b. In #2 Level 3: change it to a Level 2 requirement?
	c. In #4 Level 3, change it to a Minimum Level requirement?

12. In Checkpoint 4.2
	a. This appears to be a different way of saying what
	   Checkpoint 1.1 says, why is this checkpoint needed? If it's
	   really not a mirror, should checkpoint 1.1 be mentioned
	   someplace in the text for this checkpoint?

13. In Checkpoint 4.3
	a. Does "use longdesc if conditions warrant" fall under this
	   checkpoint? It would be very helpful to have multiple
	   examples that illustrate what this checkpoint means by
	   "annotate."

14. In Checkpoint 5.1
	a. Sun thinks the mention of protocols is relevant and
	   desireable (especially when a link also points to an
	   appendix entry that names protocols that support
	   accessibility). Content providers/developers ought to be
	   helped by pointing to where they can find more specific
	   information for ensuring that when they use non-W3C
	   technologies in web content they are using or choose
	   technologies that have access support built into them (e.g.
	   Java/Swing, PDF, realtime video, etc.).

15. Checkpoint 5.3 , MAYBE
	a. The checkpoint itself doesn't tightly tie the technology
	   used to construct the content to what WCAG 2.0 is saying
	   must be done to meet its 3 levels. Right now in WCAG 2.0 a
	   developer really doesn't know, based on 2.0's content, how
	   they programmatically answer this question (or that there
	   are technologies hey can use): How does a blind person read
	   and grok what's on a web page? Utimately, the question's
	   answer is make it possible for the AT to grab,
	   programmatically, all that has been put into the content to
	   ensure it meets the WCAG. Something like the following does
	   a clearer job of  "Choose technologies that programmatically
	   support, expose, and make possible building content that
	   meets the WCAG.
		- The MAYBE mentioned above stems from an unsureness
		  about what checkpoint in 5 this guidance ultimately
		  applies to. For example, instead of applying it to
		  5.3, maybe it would be better to put the jist of this
		  feedback (structure and content must be
		  programmatically available to an AT) into Guideline
		  5's wording or into 5.1 or 5.4 
Received on Sunday, 27 October 2002 22:53:20 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:20 GMT