If you notice, Arena's implementation of stylesheets has the style
information embedded in the <HEAD> of the document (no support yet for
external stylesheets but I'm sure that's coming). Thus, there's no loss
in speed for those who want the style information tightly coupled with
the document. The most common use of external stylesheets will probably
be for enterprise-wide styles, where a collection of documents shares a
common style sheet - in that case there would be a delay for the first
document but not for subsequent ones.
> Stylesheets should
> in no way be relied upon as the only method of rendering hints or
> assumed to exist before they actually do.
In some ways you're right, this is a chicken-and-egg situation. But
there's sufficient evidence to suggest that stylesheets should at least
be investigated by any institution doing serious R&D in web technology
> The smart approach is
> to move forward with BOTH the stylesheets proposal and continue
> to add rendering hints directly into HTML as tags and attributes.
> Having both methods will allow the public to use whichever
> method best suits their needs and will lead to the best of both worlds.
No, the standards process must *not* be driven by market forces or the
momentum of first-to-market, and it should not be John Q. Public who decides
issues of fundamental web architectures. Authoring tools will make this
issue moot to most authors anyways. We're approaching areas of religious
debate so I won't go further, suffice it to say: let's just BITE THE BULLET
and do style sheets and keep HTML from becoming RTF or Postscript. I'm
already starting to play with Arena's implementation - both myself *and* the
design folks around here are getting starry-eyed at the potential.
email@example.com firstname.lastname@example.org http://www.hotwired.com/Staff/brian/