> Keeping the audio and visual stylesheets orthogonal is not *silly* as you
> suggest.
> This is an important design goal, and I definitely do not intend doing a
> speech stylesheet mechanism that keys off the visual.
>
> This said, <strong> is clearly independent of both visual and aural
> renderings.
>
> Given that the visual stylesheet specifies to the browser how it should
> realize <strong> it's logical to allow the speech stylesheet to do the same.
You seem to mean complementary rather than orthogonal. Complementarity
is a good design goal, matching well with the machine independence that
a good web document should have. Complementarity means that the two
carry the same information in different ways, while orthogonality implies
total independence, which is not what most authors want. You need to
have different low-level functionality for visual and aural UAs, but the
best design should attempt to make the low-level functionality
transparent, so that the information the author wants to present in his
pages is preserved in machine-independent form as much as possible.
I agree that an aural stylesheet mechanism shouldn't key off a visual
stylesheet; instead, they should be complementary, following from a
combination of good descriptive markup and style-sheet prescriptions that
convey machine-independent concepts as much as possible. That is biggest
advantage of a natural language approach to style parameters, rather than
specifying machine-dependent numerical values. If someone wants to
distinguish some text, say by making it loud, it is much better to
describe this with the word loud (which can be adjusted to the tastes of
the user and capabilities of the machine) than with a fixed dB level.
David
Work: seibert@hep.physics.mcgill.ca Home: 6420 36th Ave.
Physics Department, McGill University Montreal, PQ, H1T 2Z5
3600 Univ. St., Mtl., PQ, H3A 2T8, Canada Canada
(514) 398-6496; FAX: (514) 398-3733 (514) 255-5965