Re: TIMELY: Final "tweaks" to UAAG 2.0 before Last Call

Thanks for the detailed review, Greg.  Comments in line

On 9/22/2013 3:58 AM, Greg Lowney wrote:
> Here are some observations on the changes you asked us to review, based
> on
>     1. Re 1.2.1 updated
> Fine.
>     2. Re Levels of Conformance
> 2.a. Editorial: I suggest "AA" be changed to "AA (recommended)", to
> parallel the other two levels with their parenthetical phrases.

> 2.b. Editorial: re the intro paragraph to "Levels of Conformance", it's
> not really user agent developers who conform to UAAG20 at various
> levels, but the user agent they produce. Also, you might want to change
> "prioritize new features to develop" to "prioritize ways to improve
> accessibility", as some developers are likely to associate the term
> "features" with "big" changes, rather than minor tweaks to existing
> behaviors.

> 2.c Editorial: I'd replace the bullet items such as "Level A
> conformance: All applicable level A success criteria." with wording that
> is a bit less terse and ambiguous, such as "Level A conformance: *The
> user agent complies with* all applicable level A success criteria."

DONE. Can you look at it and see if I did it the way you intended?

> 2.d Editorial: I'd replace "The three levels of UAAG 2.0 conformance are
> based on the level designations (A, AA, or AAA) of more than 100 success
> criteria (i.e., specific requirements)." with "The three levels of UAAG
> 2.0 conformance are based on the *corresponding* level designations (A,
> AA, or AAA) of *the individual* success criteria (i.e., specific
> requirements)." This also avoids rubbing their face in the fact that
> UAAG20 has more than 100 individual success criteria.


> 2.e. Editorial: Consider changing "inconvenience to other groups of
> users with disabilities" to "inconvenience to other groups of users,
> including those with other disabilities".


> 2.f Editorial: When I read "commonality" of things, I think of aspects
> which they share, not how widespread (common) they are. Thus, I would
> change "commonality of present implementations" to "how common
> implementations are today".


> 2.6 Technically, "UAAG conformance levels attempt to balance the needs"
> is incorrect, because the UAAG conformance levels are very
> straightforward, with no room for nuance. Rather, it is the *level
> designations of the individuals success criteria* which were adjusted
> "to balance the needs of people with disabilities with the difficulty
> the user agent developer could experience".


>     3. Re 1.8.13 Reduce Horizontal Scrolling.
> 3.a This is 1.8.12.

I incorrectly estimated the renumbering. It is now 1.8.7.

> 3.b I suggest rewording the Note so that it reads more like
> instructions, rather than a simple statement of fact. For example,
> change "Note: Vertical layout languages fit within the height of the
> viewport to reduce vertical scrolling." to "Note: *For* vertical layout
> languages, *text should*  fit within the height of the viewport to
> reduce vertical scrolling."


> 3.c By the way, I thought we were going to change "the user can specify"
> to "the user can have" in cases where we did not require the user agent
> to also provide the opposite behavior. If we did that, the SC would read
> "The user can *have* text content in a graphical viewport *reflow* so
> that the content fits within the width of the viewport."


>     4. Re Split of 1.8.1 and 1.8.x Customize Highlighting
> 4.a In my email of 11 September I listed five suggested changes, three
> of which would still apply after the restructuring; we discussed one in
> email, but none of the three appears to have been done. The three were:
> <blockquote>
> 3.    If we decide to keep 1.8.x, it should be reworded because the
> agreed-upon wording does not work when taken out of context (i.e. it
> talks about highlighting, not about highlighting viewports). It should
> be changed to something like "When highlighting viewports as specified
> by 1.8.1 Highlight Viewport, highlighting options include at least"
> (which parallels the wording of 1.3.2, Highlighting Options).
> 4.    In the chat of 2013-07-18 I'd suggested that we add to 1.3.2 "(d)
> shape and size when the indicator is an image", but it was at the very
> end and we didn't end up discussing it.
> 5.    If we don't merge 1.8.x into 1.3, I suggest adding to 1.3.2 an
> additional list item, "blink rate, where blinking is implemented", thus
> paralleling the fact that blinking is referenced in 1.8.x.
> </blockquote>

DONE - as agreed in the meeting of 25 September
> 4.b By the way, is the plan to put off renumbering all the SC in 1.8
> until one of the very last steps before this draft is published?

DONE. Refresh your view. I did that Friday night, I think. I haven't 
renumbered the date because other people are looking at the version.
>     5. Re Adding 1.6.5 from resolution
> 5.a The wording of the SC is acceptable, but it would be a bit clearer
> if it or the Intent paragraph were to say something like ", overriding
> any values specified by the author or inferences made by the user
> agent". This would parallel the phrase "overriding any values specified
> by the author" used in numerous other success criteria.


> 5.b Similarly, after the phrase "Much web content lacks the appropriate
> language indication or has an incorrect language attribute" I suggest
> adding ", and the user agent may not be able to accurately determine the
> language from the text".


>     6. Re Added note on RFC 2119 text to the UAAG 2.0 Conformance
> Applicability Notes
> Fine.
>     7. Re Disambiguated 3.1.1
> 7.a This seems to me to list item B reads as way too broad. It says "The
> user can avoid or defer... Information in the user agent user interface
> that is being updated or changing", but what does that really mean? Does
> the user have to be able to stop the browser from graying out menu items
> that aren't applicable to the selected content? What about changing the
> title bar to reflect the title of a newly loaded page? Do they have to
> be able to defer error messages? The older wording had this same
> ambiguity, but to a lesser extent, and where it said "non-essential or
> low priority messages and updating or changing information" it could be
> argued that "non-essential or low priority" applied the "udpating or
> changing information" as well as to "messages", and that's probably the
> easiest way to handle it. Unfortunately, I don't have specific wording
> that I find satisfactory.
> 7.b The new word order in list item A seems to subtly change what the
> qualifier "recognized" refers to. Before it was "recognized
> non-essential or low priority messages" but now it is "Recognized
> messages that are non-essential or low priority"; the latter seem to
> more clearly state that it applies to things that are recognized as
> non-essential or low-priority messages, whereas the new wording seems to
> apply to "recognized messages" that are non-essential or low-priority.
> Of course, since we have the global exemption for things not recognized
> by the user agent, including the phrase "recognized" may not be
> necessary at all.

Kim and I will look at this at our next meeting.

> 7.c There is a trailing space inside the anchor element for
> "Recognized", leading to an underlined space.


> 7.d By the way, in the list of UAAG 2.0 Conformance Applicability Notes,
> "UAAG 2.0 success criteria only apply to web content that can be
> recognized by user agents" might be better as "UAAG 2.0 success criteria
> only apply to web content *and its behaviors* that can be recognized by
> user agents". Because after all, the user agent can probably recognize
> the web content in a web page, even though it cannot fully understand
> the actions that its embedded scripts carry out.

>      Thanks,
>      Greg
> -------- Original Message --------
> Subject: TIMELY: Final "tweaks" to UAAG 2.0 before Last Call
> From: Jeanne Spellman <>
> To: User Agent Working Group <>
> Date: 9/20/2013 1:00 PM
>> I have prepared an Editor's Draft that will be the basis of the Last
>> Call working draft.  Because other groups have had questions about the
>> contents of the Status section of their documents, Judy has asked me
>> to make sure the group approves the Status section. I showed it at the
>> call on Thursday, but we had so many absences for the meeting, that I
>> want to make sure the group approves.
>> Editors' Draft:
>> In addition, I have been going through UAAG preparing the documents
>> for Last Call and discovered a few items that the group agreed to, but
>> were not yet in the document. In the interest of dotting all the i's
>> and crossing all the t's, I would like you to check the changed text
>> and give your updated approval to publish a Last Call Working Draft.
>> To make our schedule of publishing on the 24th, please respond to the
>> survey BY MONDAY morning, 23 September.  If there are changes, we may
>> have to postpone publishing to the 26th, but I would like the document
>> to be the best it can be.
>> 1.  1.2.1 updated
>> 2.  Levels of Conformance moved to Implementing. Kim wrote some new
>> text for an introduction to Implementing UAAG, because otherwise, the
>> Introduction went straight into the Levels of Conformance with no
>> context. Check it out.
>> 3. 1.8.13 Reduce Horizontal Scrolling. I heard back from the
>> Internationalization Domain lead at W3C who said that inclusion of
>> vertical layout languages was important and it was the most requested
>> feature at a recent e-Publishing workshop.  I took the vertical layout
>> text and moved it to a normative note. Combined with Kim's new title
>> stem for the SC, I think now it is very clear that the purpose is to
>> prevent horizontal scrolling, AND if it is a vertical layout language,
>> to prevent vertical scrolling.
>> 4. Split of 1.8.1 and 1.8.x Customize Highlighting
>> 5. Adding 1.6.5 from resolution
>> 6. Added note on RFC 2119 text to the UAAG 2.0 Conformance
>> Applicability Notes (
>> )
>> 7. Disambiguated 3.1.1

Jeanne Spellman
W3C Web Accessibility Initiative

Received on Thursday, 26 September 2013 20:37:01 UTC