variables revisited

to follow up on what Dave Raggett said:

> At one point there was a proposal for inserting text before or after
> an element. Sadly this was rejected by the CSS and FP working group.
> 
> I think there is a good case to be made for the more specific
> proposal for speak-before and speak-after, defined in a similar
> manner to cue-before and cue-after. For example:

[Good suggestion snipped.]

Along the same lines of "what _can_ we get if we can't have
everything," there is a limited form of variables that is needed
to make Jason's idea of a SpeakCell (or SpeakRow or SpeakCol)
attribute work.

That is the capability to pull a multi-stage attribute
dereference into the pattern.  An example piece of content you
want to be able to use in such an attribute is related to the
current entity instance by the relationship:

     "the content of a cell which has an ID mentioned
     in the HEADERS attribute, or in whose SCOPE I fall."

This is much less general than a complete
"define-includable-fragment" capability, but it would really help
us to pull together related pieces of an HTML structure which are
not juxtaposed in the syntax tree.

That's in the category of "I am convinced we want it; I am not
convinced we can get it."

-- Al

Received on Monday, 3 November 1997 13:52:42 UTC