Re: CSS ID, xml:id confusion

On 6/30/05, Ian Hickson <ian@hixie.ch> wrote:
> On Thu, 30 Jun 2005, Orion Adrian wrote:
> >
> > What is a CSS ID (the one referenced by #)?
> 
> CSS is a little muddled in how it presents it but the key term in the
> CSS2.1 spec is "of type ID".
> 
> 
> > What is an xml:id?
> 
> The xml:id spec says that xml:id is of type ID.
> 
> 
> Note that xml:id explicitly says that no change is required to CSS
> processors or to the CSS spec (appendix C bullet point 5). Note also that
> CSS2.1 explicitly mentions xml:id as being a source of IDs (5.9 last
> paragraph). Note finally that the W3C Selectors test suite already
> contains a test that that includes xml:id (test 15c).
> 
> The request that started this thread is not about making the specs
> clearer; the specs are fine. The request is about making the CSS spec
> somehow endorse the xml:id spec.
> 
> Personally I agree with Bjoern and David (amongst others), and consider
> the concept of one spec endorsing another to be silly and inappropriate.
> 
> As it happens, that is the current position of the CSSWG as a whole: the
> topic was recently raised and resolved in a face-to-face meeting. At the
> time the discussion was about requiring support for particular bitmap
> image formats, but the resulting discussion was a generic one and not tied
> to endorsing particular types of specs; it applies to endorsing xml:id
> just as much as to endorsing PNG or JPEG.

I actually agree with you here. Examples are fine since examples can
and should change, however where is "of type ID" defined? I can have
two different specs both have the text "of type date" and they could
mean different things. I feel the same problem is here. The specs in
question and all other specs do not have a singular point of reference
say "of type ID (http://w3.org/types/id)". Or "as specified in this
specification (whose sole purpose would be to specify types.

That is what I'm asking for.

Orion Adrian

Received on Friday, 1 July 2005 02:44:12 UTC