In message <199702271514.KAA24157@nathaniel.ebt> gtn@ebt.com (Gavin Nicol) writes:
[...]
> I think for XML, we should probably get more concrete. The question
> is how much more concrete? How much information *must* the API support?
There are possibly two questions here. Do we need to _pass the information_,
and do we need to _have special names_ (rather than just Property). I suspect
that you will find it difficult to get rid of anything below that will
satisfy everyone. My own simple experience is that I would actively use
the following:
> Do we *need*
Element, Y
Attribute, Y
Comment, N (an XML processor _may_, but need not, pass these.
where would they be attached?)
PI, Y
Header, Y
DTD, Y
MarkedSection, N
EntityReference, Possibly, when I get brave
Entity etc. etc?
>
I would, however, be prepared to receive them as Property, so long as they
were clearly documented, and I assume that they might well be subclassed
from Property anyway.
P.
-- Peter Murray-Rust, domestic net connection Virtual School of Molecular Sciences http://www.vsms.nottingham.ac.uk/