Re: Review of WAI-USERAGENT-19990611

At 1999-06-23 10:34 AM, Ian Jacobs wrote:
>Harvey Bingham wrote:
>
...
>They are actually in the TOC, but not in the <UL> of the TOC
>(because they are external documents and I wanted to distinguish
>them). Thus, after "6. References" of the <UL>, the following statement
>indicates they are appendixes:
>
>" The appendix list of checkpoints is available as either a 
>  tabular summary of checkpoints or as a simple list of
>  checkpoints. "
>
>There are more references to the fact that these are appendixes
>to this document in the Status section.

Interesting extension of the word "appendix". My book-centric
use is appendix in the same document. Why bother to call these 
external documents "Appendices?"

>> >3.1 Conformance
>> 
>> >Conformance claims
>> 
>> [Suggest if dependent UA, further qualify limits expected by that UA.]
>
>I'm not sure what this means. Can you give an example?

Old screen readers went across table cells as they knew not of the possibility
to fold cell content onto multiple lines.

That DUA only serves a particular class of use, and presumes some information
in its dependency with the general UA. How is that interface specified? 


>> >4. User Agent Accessibility Guidelines
>> 
>> >Please note that "device-independent support" does not mean that user agents
>> must support every input or output device. User agents should offer redundant
>> input and output mechanisms for those devices that are supported. For 
>example,
>> if a user agent supports keyboard and mouse input, users should be able to
>> interact with all features using either the keyboard or the mouse.
>> 
>> [? I am uncertain that mouse needs to provide all keyboard capabilities?]
>
>That's an interesting point. For example, must I be able to type
>via the mouse (e.g., by using an entry palette)?

So far as I am aware, that capability was omitted from XMetaL 1.0, possibly
to be put back into XMetaL 2.0. It was available in HoTMetaL, I believe since
version 3. It and the other accessibility features of that product never got
any mention in any product reviews, so their inclusion in the first product
of SoftQuad Software, Inc. was less important than getting to market with
a product.
> 
>> [General questions:
>> How handle the bootstrap process to get started: what can be assumed:
>> operating system, file system, available device "plug & play"?]
>
>For generality, we should not assume an operating system or
>file system (e.g., when using a kiosk). Where should we mention this?
> 
Not sure. Without assumptions, what can we say about making that process
accessible?

>> >4.10 Allow the user to turn on and off support for author style sheets.
>> 
>> [Does this mean substitute for author style sheets, in part or completely,
>> using the CSS cascading and IMPORTANT indications?
>> 
>> Or was the intent: audible, rather than author? If author, should audible
>> style sheets be a separate list item?]
>
>I would suggest that "support for author style sheets" means that
>they are parsed and cascaded with other style sheets according to 
>the style sheet spec. Turning off support would mean they are ignored.
> 
>> >4.13 ?automatic page forwarding?
>> [I'm unfamiliar with this, is it a "feature" of existing browsers?]
>
>Some browsers support META element declarations such as this:
>
>   <META http-equiv="refresh" content="5; http://www.acme.com/newpage">
>
>After a 5-second delay, the new page is loaded. 
> 
That "automatic page forwarding" remains semantically confusing. Can we
describe it differently, or expand the description. The "forwarding" part
implied to me that I will somehow be doing the forwarding. "Refresh" is
the concept that asserts server-side action. How about
"automatic refresh to a different page content after a time delay"?

>> [replaced "at the same time" by "synchronized with"]
>> [Similarly for 6.11 and 6.12]
>
>I think "at the same time" doesn't mean "synchronized with"
>in this case, but rather that both are rendered. In other words,
>when captions are available (and as captions they are
>already synchronized), all the user to view them while the video
>is running (as opposed to captions-only).
> 
Good point. The issue remains at what granularity is the synchronization, 
or "at the same time". I guess I read more into "at the same time" than
just "concurrently".


>> 
>> >8.8 For dependent user agents. Allow the user to view a document outline
>> constructed from its structural elements (e.g., from header and list 
>elements).
>> [Priority 2]
>> 
>> [Not all list elements have any identification that would help to
>> differentiate structure.]
>
>Which list elements are those? I think the word "element" is supposed
>to suggest markup.

In the DOM sense, structure and sibling sequence preserves structure.
Unordered list items may have no presentational differentiation. Nor may
definition list terms. For these, the sequence of occurrence might be
used associated with a count of position for that otherwise undifferentiated
structure.
> 
>> >8.16 Make available whether a chosen link has already been visited.
>> 
>> [even if multiple instances of that link occur in the document.]
>
>A "chosen link" means a particular A element, not the content
>of the element or the destination of the link. 

My proposal would mark all occurrences of a visited link, if any one were
visited.

Regards/Harvey

Received on Wednesday, 23 June 1999 11:50:36 UTC