definition of 'content' (again) (sorry)

At 03:34 PM 2000-04-27 -0400, Ian wrote:
>WAI UA Teleconf
>25 Apr 2000
>
>3) Discussion
>
>    1.PR#278: Definition of "content", etc
>      http://cmos-eng.rehab.uiuc.edu/ua-issues/issues-linear.html#278
>
>     Ian's proposal:
>     http://lists.w3.org/Archives/Public/w3c-wai-ua/2000AprJun/0207.html
>
>     Resolved: Adopt Eric's proposal + Ian's addendum.
>     Action IJ: Add definitions to the document.

I apologize for not being better prepared for the telecon.  I do have
problems with Eric's rewrite.  He had problems, too, but he was trying to
be conciliatory.  I want to make another try.

So I have tried to pen yet another version.  The key change is that I add
"but the definition is still 'all of the above'" at the end.  If people
have heartburn with this, please say so.

[Referenced baseline] by Eric Hansen:

<BLOCKQUOTE>
In this specification, the "content" refers to the document object. Some
content is designed (by specification) for "human consumption". For an HTML
document, this typically includes what appears between the start and end
tags of elements, and the values of some attributes (e.g., alt, title,
summary). Other content is meant for machines, typically including the
markup itself (e.g,. element and attribute names), some attribute values
(e.g., class, id, lang, src), style sheets, scripts, etc.
</BLOCKQUOTE>

Comments by Eric Hansen:

Also, I think that the distinction between "for human consumption" and "not
for human consumption" is not as tight as the definition make it appear. Are
there not exceptions?

Comments by Al Gilman:

Clearly in the case of CLASS and LANG the attribute values are intended to
be used in a way which is both machine and human understandable.  For
others the practice is such that decoding the prime content is materially
aided by inspecting some [and unpredictable selelection] of the
normally-machine-processed attributes.  And if you want to inspect an
attribute value, you need to know what attribute it is. 

In general, the language 'designed (by specification) for "human
consumption"' is innacurate.   Furthermore it is dangerous.  So we just
shouldn't go there.

Proposal by Al Gilman:

<BLOCKQUOTE>
In this specification, the "content" refers to the document object. Some
content is routinely displayed in the course of browsing by people with no
disabilities. For an HTML document, this typically includes what appears
between the start and end tags of elements, and the values of some
attributes (e.g., alt, title, summary). Other content is usually processed
invisibly by the User Agent, such as DTD-defined names of element types and
attributes, and other attribute values such as href, id, etc. In Guideline
2, the content to which the user must have some form of access includes all
of the above.
</BLOCKQUOTE>

Al

PS:  Ian, if this statement of what we do mean flies with the group, I
don't think it should be necessary to say something about what we don't
mean.  Particularly it would be digressive and misleading to try to explain
in what sense "We don't mean information content."  Better not to go there
at all.


The purpose of the guideline is to achieve complete access to the
information content.  The checkpoints and techniques are perforce
articulated in terms of the software's view of the data.  The point is not
that we don't want to assure full access to the information content, we do.
 The distinction is that the format specs, the WCAG, and the best we can do
here does not supercede the data model in the format specs as the reference
frame in which to articulate checkpoints and techniques.

Received on Thursday, 27 April 2000 23:21:36 UTC