[whatwg] Semantic styling languages in the guise of HTML attributes.

[Hallvord R M Steen]

On 01/01/07, Mike Schinkel <mikeschinkel at gmail.com> wrote:

> > Interesting that you should choose that example, because
> > it can mean different things depending on the element you use
> > it on. Therefore, a global |type| attribute would almost
> > certainly conflict with the element-specific attribute unless
> > it was defined otherwise.
>
> Conflicts don't create any anxiety for me. If there is a conflict either
> there is an undefined state or one of the two is defined to take
> prescedence.

Right. Do you volunteer for the job of going through all "role" values
and all current HTML element semantics and define which one takes
presedence in each possible conflict? Matthew's point is that this
task itself is massive. Imagine: for each suggested "role" value, go
through all HTML elements (since "role" is a global attribute) and
note possible conflicts. Go through all other HTML attributes (and
their values) and note possible conflicts. Now define the outcome of
each conflict. Do it again, but now in C++ code.
His secondary point is that if the spec doesn't do this, every browser
will handle things differently so authors will not be able to use
"role" as intended because of browser incompatibilities.

> > I've shown you that not only are there conflicts with
> > proposed attributes, roles and elements, but that they
> > actually compete in certain situations. Furthermore, I did so
> > with only minutes worth of research.
>
> I don't see that as a problem they way you do.

That's the attitude that made HTML error recovery such a mess :-(

>  you have an opinion that few if any others are rallying behind.

Perhaps because it seems too obvious to discuss?
You could go back to the list of markup samples, and tell us for each
of them what you feel the actual semantic of that should be, and how
it should be handled visually or in form submission. Once you've tried
to understand the complexity involved we'll have more common ground
for the discussion.

-- 
Hallvord R. M. Steen