Re: Experimentation with extensions, LINK and META

On Mon, 13 Oct 1997, Al Gilman wrote:

> We are at pretty much of an impasse in the TABLES area.  I think
> that the AXIS, AXES proposal fails to meet minimum standards of
> functionality.  Dave doesn't see the problem.
> 

I do not see the problem either. All of the sample tables which have been
considered so far can be made accessible by the inclusion of AXIS and AXES
attributes. In particular, the relationship between any given data cell
and an associated row or column header, whether the latter be contained in
a TH or TD element, can be expressed using the AXES attribute.

I agree that the proposed SCOPE attribute is beneficial in as much as it
reduces the amount of markup required in order to specify header
relationships, thereby relieving the author of the need to specify AXES
attributes repeatedly throughout the body of the table.

Furthermore, the flexibility to associate a table with an external
resource, such as a dictionary or extended description, is also desirable.

None of the discussion so far has convinced me that AXIS and AXES,
together with SCOPE, the capacity to link a table to external resources
and the availability of a reserved class value to specify column
dominance, would not meet accessibility requirements. Additional
enhancements, such as CSS macros of the kind that I have proposed for
controlling the reading order of a table, add further value to the HTML
features already discussed, but are not essential to accessibility.

Received on Monday, 13 October 1997 19:28:08 UTC