Re: Updated UA working draft

Thanks for th comments.  I think you have alot of great suggestions and
comments.  Let's just waite until after the next telecon before we make
alot of changes.

At 03:57 PM 8/14/98 -0400, Ian B. Jacobs wrote:
>Jon Gunderson wrote:
>> I have updated the working draft and issues pages.  Please review and
>> comment.
>> Especially the action item group members from the last telecon.
>> Jon
>> New working draft:
>>Section 4.1
>>2.[PRIORITY 1]
>>      Allow user to turn off blinking text generated by 
>>non-standard HTML tags link "blink". The text 
>>should be rendered as visible text. 
>I propose that the guideline read: 
>  "Allow users to turn off blinking text. The text should be rendered as
>visible text."
>First, I think that mention of non-standard HTML elements should be
>avoided when possible. Second, CSS1 allows a blinking effect
>('text-decoration: blink') and I suspect that the guideline would want
>UAs to turn off this blinking as well.
>>Section 4.3
>>2.[PRIORITY 1]
>>      When an IMG element has a value for the "longdesc" attribute and
>the user has turned off the display of images, give the user access to
>the long description. 
>Shouldn't the user (or at least some users) have access even if the
>images are turned on?
>>The LONGDESC should be rendered as a link with a "D" caption after the
>short description provided by "alt" or "title". The link should act as >a
standard link.
>Because the longdesc is not rendered inline, I propose instead: 
>        "Access to the long description should be provided through a 
>        link. The link text should be the value of the 
>        "alt" or "title" attribute followed by
>        the letter "D"."
>>If images are turned on, the image becomes one of the links that can >be
accessed through keyboard navigation. Since IMG can also be part >of
anchors a separate keyboard command is needed to select the URL >associated
with "longdesc". 
>The meaning of this is unclear to me. If a link to the
>long description is rendered in the document (as a standard
>link), why would there be any need to treat it, or the
>IMG element that engendered it, specially? Do the last
>few sentences of  the guideline only apply when the 
>IMG is the content of an A element. I'm not sure I understand
>the meaning here. Finally, perhaps these sentences should
>be moved elsewhere since the guideline specifically
>says "when the user has turned off the display of images."
>>Section 4.4
>>3.[PRIORITY 1]
>>      Allow the user to suspend timer based animations as the result >of
DTML timer scripting events. Users with some disabilities may not >be able
to follow the animation or the animation may 
>>trigger seizures.
>I propose dropping "as the result of DHTML timer scripting events."
>Also, what is a "timer-based animation"? Does the guideline
>refer to all animations or only certain ones?
>>Section 4.6
>I think guidelines 2 and 3 need fixing. 
>- The examples are too long (though useful). I propose
>  an appendix where table serialization is discussed.
>- The guidelines are the same, but two types of tables are
>  discussed (those with header info and those without,
>  the latter more likely to be used for layout). The
>  guideline should just be: "Allow the user to select
>serialized          representation of table by row or column order."
>  the two cases can be discussed in the prose.
>- Guideline 3 reads "A table that includes "TH" tags indicates
>  that the table is being used for data rendering and that row
>and        column position are important for understanding the
>  What about people who use TH to get a style effect, like bold
>  text?
>- The two guidelines have been created from one, but some
>  information has not been pruned during editing. Namely,
>  the statemetn "Users should be able to specify whether 
>  they want the cells rendered row order or column order." 
>  should be removed. Also, the Note (Table serialization 
>  is important for tables used to structure tabular 
>  information as well as those used to lay out blocks 
>  of information.) should be removed since both cases
>  are to be explored explicitly.
>>4. "Try again feature allow the use ..."
>  This guideline seems redundant since the previous
>  serialization guidelines refer to allowing the user
>  to choose from among serialization approaches. I propose
>  that this guideline be dropped. However, more information
>  about serialization alternatives might be provided
>  in the previous guidelines.
>>6.4 Direct Access Navigation
>>    1.[PRIORITY 1]
>>      Allow the current page to be searched using keyboard commands >for
a given phrase. If the search text is found, the selection should >be moved
to the occurrence. If the text occurs within an
>>link, the focus should be changed to the anchor. 
>>    2.[PRIORITY 2]
>>      Allow only the text associated with links (including ALT text >for
images that are links) on the current page to be searched using >keyboard
commands for a given phrase. If the search text is
>>found, the focus should be moved to the occurrence. 
>>    5.[PRIORITY 2]
>>      Allow the user to use the keyboard to move the selection >directly
to elements by searching for text matches that includes ALT >text, TITLE
text and LONDESC text in the search. 
>>    6.[PRIORITY 3]
>>      User option to include the text contents of LONGDESC links to be
>included in text searches of the current page. If the search text is
>found in a LONGDESC link, the focus should be changed to
>>the LONGDESC link on the current page. 
>I think these four guidelines should be harmonized (among themselves,
>and with the wording of the adjacent guidelines). I propose:
>      Allow the user to use the keyboard to perform a text search of the
>current page. If the search text is found, the selection should be set
>to the matched text. If the text occurs within a
>link, the focus should be moved to the link.
>    2.[PRIORITY 2]
>      Allow the user to use the keyboard to perform a text search of
>link text only on the current page. Link text includes the content of A
>elements and "alt" values for images used as links. If
>the search text is found, the focus should be moved to the link.
>Numbers 5 and 6 should be combined:
>Allow the user to use the keyboard to perform a text search only of
>values of the "alt" and "title" attributes, and within documents
>designed by "longdesc" attributes. If the search
>text is found, the selection should be set to the matched text. However,
>if the search text is in an external document, the focus 
>should move to the "longdesc" link on the current page designating that
>- Ian
>Ian Jacobs ( 
>Tel/Fax: (212) 684-1814 
Jon Gunderson, Ph.D., ATP
Coordinator of Assistive Communication and Information Technology
Division of Rehabilitation - Education Services
University of Illinois at Urbana/Champaign
1207 S. Oak Street
Champaign, IL 61820

Voice: 217-244-5870
Fax: 217-333-0248

Received on Tuesday, 18 August 1998 07:35:00 UTC