Re: Ideas for the ACSS module of CSS3

On Sat, 14 Oct 2000, Ian Hickson wrote:

> As currently specified, there is no way a browser could concurently render
> a document using CSS both visually and aurally without doing so in a
> completely independent fashion.
> 
> This may or may not be a deficiency in the spec. It could be argued that
> for that kind of document SMIL would be more appropriate.

Certainly, it seems as if styles apply more or less orthogonally to
simultaneous media. Presumably the two could still be sychronised to some
extent (independent of style sheets) - e.g. with paged visual medium +
speech, the speech might stop at the end of the page, until the page is
"turned"; or the currently selected link on a screen could be made to
correspond with the one most recently mentioned in speech (or braille, or
3D animated sign language...).

It might be nice to be able to specify this sort of behaviour in style
sheets, though I feel it is likely that it will mostly be a matter solely
of user preference. More useful perhaps, allow for visual properties to be
applied to the text (or table row or whatever) currently being spoken, and
so on. (Perhaps one of the CSS3 pseudo-classes/elements already permits
this, maybe with "suitable interpretation" on the part of the hypothetical
UA.)

There may also be some issues with properties that apply to both media: If
I set different values for the same property for two media to which the
document is being rendered synchronously, should the two media receive
their respective values, or should the media be considered together, so
that the settings conflict, and the final value is determined by the
normal cascading rules? I suspect that there may not be a single correct
answer which applies to all properties.


Tim Bagot

Received on Saturday, 14 October 2000 11:17:14 UTC