Re: Extensibility and Uniformity; defining "error" behavior

Dear all,

the CSS definition of "interoperability" was subject to an internal
discussion in W3C during the preparation of the CSS 2.1 CR. I made a
contribution to this internal discussion that was seen as valuable. I
try to adapt the point I tried to make to the current debate on
www-tag.

My reaction to this in the previous discussion was that the word
interoperable has many meanings in our language, but also in
legal language. It is a common technique in legal texts to define the
meaning of a word within the legal text to achieve more precision.

But the definition within this legal text is scoped to the legal text
it is meant for.

The CSS Working Group defined "interoperability" as the ability to 
pass certain tests with a well defined result. The CSS 2.1 definition
of "interoperability" is closely tight to the needs of CSS 2.1 and
serves this purpose well. I do not think that generalizing it is
appropriate. I also do not think that it is worth the effort to enter
the renaming game and relabel "interoperability" to "conformance"
or "foo". CSS 2.1 has defined exactly what it means
by "interoperability". So CSS 2.1 should stick with this. For HTML,
we have different challenges that may need different criteria.

Best, 

Rigo Wenning
W3C Legal Counsel


On Tuesday 30 December 2008, Anne van Kesteren wrote:
> On Tue, 30 Dec 2008 15:23:00 +0100, Larry Masinter
> <masinter@adobe.com>
>
> wrote:
> > I think you are talking about "uniformity" when you use the word
> > "interoperability".
>
> FWIW, the W3C generally uses the word "interoperability" where you
> say "uniformity". See e.g. http://www.w3.org/TR/CSS21/#crec