Re: Evaluation Results In XML

Thanks for your comments. We're still developing the software so there is
lots of room to change.

I've heard from Tim Springer that SSB uses an XML doc in their evaluation
process and he will post some details to the list. Maybe there will be some
new ideas there too.

Al asked:
> What do you propose to do when the sections of original-HTML hypertext
that
> give rise to the problem assertions overlap?
>
We're thinking of using attributes for each problem type, like:

<problem missing_alt_text="yes" missing_longdesc="yes" problem_id="1234">
<![CDATA[ <img src="rex.jpg">]]>
</problem>

(Note: this is different from my original example.)

Al asked:
> Did you consider using X-Path to mark ranges in the source, and leave the
> source in a separate file? [You would have to first tidy the HTML into
XHTML
> 1.0 so the X-Paths are well posed, but that works, doesn't it (I mean
tidy)?]
>
No we haven't considered this. We wanted the evaluation and the original
document in the same file so they wouldn't get out-of-sync with each other.
(I'm not familiar with x-path. Can you give me a link to some info?)

Al asked:
> Did you consider using RDF for the problem assertions?
>
We're thinking about it. There has been some work and discussion on using
RDF for this purpose on the list. Creating attributes to define the problem
was the easiest thing to get going.

Al stated:
> >Each accessibility problem is given a code number so it may be
referenced.
> >
> This is not sustainable. Use URI references to AERT clauses, using an URL
for
> the AERT which is specfic to a dated version thereof.
>
I think you have misunderstood the purpose of the code number. The code
number is only relative to the particular document. It's used to refer to a
particular problem. For example: A repair tool is fed problem "1234" from
the evaluation document. The repair tool fixes the problem and returns some
text and the problem number "1234". The evaluator knows to replace the text
in problem "1234" with the new text.

I think that you may be referring to our problem type attribute (example
missing_alt_text="yes"). These directly map to the AERT document and we
could use AERT technique numbers instead of text strings.

Chris


----- Original Message -----
From: "Al Gilman" <asgilman@iamdigex.net>
To: "Chris Ridpath" <chris.ridpath@utoronto.ca>
Cc: <w3c-wai-er-ig@w3.org>
Sent: Monday, October 23, 2000 5:42 PM
Subject: Re: Evaluation Results In XML


> At 11:51 AM 2000-10-23 -0400, you wrote:
> >We have been working on a means of storing the accessibility evaluation
of
> >an HTML document. Our current approach is to store the evaluation in an
XML
> >document. The XML doc contains the original HTML with any accessibility
> >problems marked with new XML elements. For example, the following snippet
> >contains the evaluation of an IMG element that is missing the 'alt'
> >attribute:
> >
> ><problem problemName="MISSING_IMG_ALTTEXT" problemID="1234">
> ><![CDATA[ <img src="rex.jpg" longdesc="rex-desc.html">]]>
> ></problem>
>
> What do you propose to do when the sections of original-HTML hypertext
that
> give rise to the problem assertions overlap?
>
> Did you consider using X-Path to mark ranges in the source, and leave the
> source in a separate file? [You would have to first tidy the HTML into
XHTML
> 1.0 so the X-Paths are well posed, but that works, doesn't it (I mean
tidy)?]
>
> Did you consider using RDF for the problem assertions?
>
> If we are going to interleave problems and original content in one XML
> document, why not use namespaces to distinguish them, as opposed to
burying
> proper hypertext in PCDATA inlines?
>
> >
> >The XML file that contains the above evaluation is attached to this
message.
> >
> >Each accessibility problem is given a code number so it may be
referenced.
> >
>
> This is not sustainable. Use URI references to AERT clauses, using an URL
for
> the AERT which is specfic to a dated version thereof.
>
> The early points are aesthetic, questions of highest-and-best use of XML.
>
> The final point [URI-references for identification of the problem] is a
must.
> I would want three good reasons before I fought this one all the way to
the
> Director's desk. This is in Tim's eyes the capstone principle of the Web.
> You
> don't take the dictionary of error cases private by using an opaque,
private
> numbering scheme where there is a Web-compatible means to identify them by
> reference to the WAI utterance where they are authoritatively defined.
>
> Al
>
> >A report tool can take the XML document and prepare a report of
> >accessibility problems.
> >
> >A repair tool can take the entire document, or pieces of the document,
make
> >repairs then update the original XML document.
> >
> >The original XML document can be easily converted back to HTML by XSLT or
a
> >simple program.
> >
> >If the group can agree on a set of specifications then all tool makers
can
> >generate and use the same XML evaluation document.
> >
> >Comments?
> >
> >Chris
> >
> >
> >

Received on Tuesday, 24 October 2000 15:35:01 UTC