You would probably go the opposite direction. If you referred to them in
the text as "the blue one" and "the red one" and "the green one" then the
CLASS names "red", "blue" and "green" are semantically very important, and
should not be "thrown away" because the user does not use the style sheet
format that you do.
If you referred to them as "example one", "example two" and "example three",
then you should use those class names.
> One could also argue that where the styling is used to
>convey a specific typographic impression, giving the styling would be
>more useful to the visually impaired reader than giving an arbitrary
>name to the styling, since the reader could then visualize the
>appearance of the material.
A blind person visualize? I hope I'm not showing ignorance here, but if
you've been blind since birth, the difference between red and blue is
probably pretty meaningless. The difference between "first example" and
"second example" is very explicit.
Why are you even giving the styles "arbitrary" names. The colour has to
represent _something_. Is it "discussed in the text"? Is it "the column
that will be referred to as the red column in th text"? If the column isn't
at all important, why are you drawing my attention to it? And if it is
important, there must be some way to describe its importance!
>I'd be interested in hearing more about how a browser would drive a
>speech synthesizer - what elements or attributes would be spoken?
I don't know. I have spoken to people who use them, though. If I were
writing a speech-synth browser, I would try to map classes to sounds or
inflections, and when that failed, I would have the CLASS just read aloud to
them. Styles, being semantically meaningless (else, why not CLASS?) would
be thrown away.
So where does the STYLE attribute come in? Well, not all documents are
designed for maximum readability by the widest audience. Sometimes you just
want a little "flair" that is not related to actually communicating a
_point_ (other than the point that you have "flair"). In my mind, the STYLE
attribute allows you to apply formatting for formatting's sake, and not for
communication.
Since the Web is primarily a communication medium, I feel that that this
element should be deprecated (in the old fashioned sense, not the
standards-setting sense). And yes, it should be a little harder to get at,
like dynamic_cast() in C++, type conversions in C, procedural programming in
SmallTalk/Java, etc. I would _not_ support this position if I thought that
most people who used this STYLE attribute would understand it. In HTML 5.0,
when everyone understands exactly what the Web is, we can make it easier to
get at.
The STYLE compromise that was suggested makes it just as difficult to use
STYLE as to use CLASS. While you are scrolling to the top to modify your
STYLE you can think: "Is this really a CLASS what am I trying to
communicate?" And if the answer is "visual differentiation for its own
sake" then you can go ahead with your STYLE.
Paul Prescod