Any implementation based on a real SGML parser.
OK, so none exist, but it's perfectly reasonable
for information providers to validate HTML documents
with sgmls before serving them.
> I don't believe you have to write
> code each time you want to _ignore_ another tag.
You do need to modify the DTD, though.
Now you *could* add another parameter entity hook
to handle user-definable header elements, but
I still think that a single <META> element is a better
way to go, though -- authors won't need to change
the DTD at all that way.
> And I don't believe
> you can _act_ on a new tag _without_ writing more code.
This is part of the reason why <META> is a *better*
solution: so servers only need to scan for one tag
to respond to an HTTP HEAD request.
One other consideration: There is a limit to how
many new elements you can declare -- GRPCNT and GRPGTCNT
apply here. These can of course be increased if necessary.
> Hmmm... isn't this about like declaring an element X with
> attributes A1, A2, ... up to, oh, let's say A9. Use them
> for whatever you like.
No, not really.
--Joe English