I think that such work and more (much more) has been done with RDF
lately -- check W3C (http://www.w3c.org) on this and the Dublin Core
initiative (see link there).
I think that the two best options for solving this problem are: using
meta-data, as has been done with HTML, although in a simple fashion; and
using tag libraries, which will promote conformance of structures such
as tables, lists and links, not simply tag names, through
de-facto conformance. I think that they are both good ideas and that
can exist together.
The meta-data issue needs to be discussed, as the META scheme won't do
for XML docs. I assume that most people would like to see RDF
as the replacement. I remember reading about RDF about a year ago and
even writing a paper about it and other metadata schemes, but I have not
looked at it much since. Has much work been done in the area of
standardizing a scheme for web-pages, with cooperation from search
On Mon, 7 Dec 1998, Peter Flynn wrote:
> > I would have thought the elements would have to be along the lines
> > <basedocns:author> and not <author> to work. Otherwise, there is
> Any attempt to bind meaning to element type names is going to run
> into trouble, I think. There simply too many conflicting demands
> > Also, what about searching for documents in another language?
> That's just one of the problems: I will want to use <auteur> in
> France, not <author>.
> > At least a
> > DTD can have synonyms.
> Can you explain this a bit more?
<?xml version="1.0" standalone="yes"?>
<EMAIL>relander at uwaterloo.ca</EMAIL>