Re: Phill's comments on UA guidelines

pjenkins@us.ibm.com wrote:
> 
> Ian, some of these we briefly reviewed at the Authoring Tools meeting in Boston,
> others I will post to the public list.
> 
> W3C Working Draft dated 5-October-1999
> 
> Overall:
>      Organization of checkpoints seemed confusing.  I would recommend grouping
> the one about the accessibility of the UA seperate from the responsibility of
> the UA rendering content accessibly. i.e chrome vs content.
> 
> content vs resources
>      define terms and use consistently

I've gone through to harmonize. 
 
> 1.0 Into
> was  By making information available to dependent user agents throught standard
> APIs.
> to   By making Web content and UA information available to...

Ok.
 
> was  Following system standards and conventions promotes...
> to   Following platform and system standards and guidelines promotes...

Ok.
 
> 2.1 Document conventions
> was  Element names are in uppercase...
> to   HTML element (tag) names are in uppercase...
> was  Attribute names are quoted...
> to   HTML attribute names are...
> add  other conventions used?

I propose adding examples, too!
 
> Guideline 3. Ensure user access to all content
> 3rd paragraph: Requiring both text and graphics to be rendered at the same time
> by one UA seems undue burden unless the requirement is to only provide access
> via APIs for other assistive devices.  Rendering both Graphics and Text at the
> same time may be visually difficult if not immpossible.  I don't think this is
> what is wanted.

I think how it's rendered is an implementation issue (though not a
negligible
one). Clarified in 22 Oct draft.
 
> 3.2 Optionally? Why is it optional for UA to render associated header info?  Do
> you mean: render associated header info if it exists?

Deleted.
 
> 5.17 uses "windows" term, should this be "viewport" term?

Fixed.
 
> Guideline 8 second bullet
> was  Direct access... is lost.
> to   Direct access... may be lost.

Ok.
 
> 9.3 For dependent UA's only, allow the user to view outline...

Agreed to move to all UAs in recent teleconf.

> 9.9 For dependent UA;s only, provide access to header information...
>      why for dependent UA's only?

The problem in 9.9 was "selected". However, without that, I agree
this checkpoint applies to all UAs (and is subsumed by
the checkpoint about access to all content). So it could be reworded
as "Provide access to table cell header information." or deleted.
 
> Add UA-CHECKLIST to references

Done.
 
> Regards,
> Phill Jenkins     (678-4517 tieline)  1-512-838-4517
> Accessibility Program Manager
> IBM Special Needs  Fax:  1-512-838-9367
> 11501 Burnet Rd    pjenkins@us.ibm.com
> Austin TX 78758    http://www.ibm.com/sns

-- 
Ian Jacobs (jacobs@w3.org)   http://www.w3.org/People/Jacobs
Tel/Fax:                     +1 212 684-1814
Cell:                        +1 917 450-8783

Received on Friday, 22 October 1999 00:35:56 UTC