> <A> and <LINK> have a REL attribute but no REV attribute in
> ftp://info.cern.ch/pub/www/dev/htmlplus.dtd
> I'm glad to see we tossed out ROLE since it really doesn't talk about the
> "ROLE" of the link, but rather the "REL"ationship of the documents. ROLE
> is, of course, fine for things like <P> since that's what it is.
Can you explain what you see as the differences between:
a) "ROLE" of the link
b) "REL"ationship of the documents
Actually in the existing draft spec for HTML+ the LINK element can take
a ROLE attribute *but not* REL or REV.
I have been thinking about support for splitting large documents such as books
into smaller nodes offering low latency. It seems from this study, that LINK
isn't needed, and that you can get away with a slightly extended version of
the A element supporting bookmarks and hypertext paths.
Perhaps, we only need LINK for style sheets?
Dave Raggett