Re: REL and REV attributes (Was: More comments on HTML 3.0)

Craig Hubley (
Thu, 27 Apr 95 16:34:30 EDT

> My vote goes for keeping REL and REV. The fact is that many
> relationships are not strictly binary. Additionally I should not have to
> have write access to a remote file to express how I relate to it (not
> just how it relates to me).

Yes of course. No one is arguing to remove REL and REV, are they ?

> Extra-document navigational paradigms are good. I'd like to see a browser
> (or write a Java applet :) that did an MHEAD depth-first traversal on a site
> going 6 or levels deap that built a sketch of the infospace on the site, with
> "parents" (or their semantic equivalent given a large glossary for the
> "higher" in a hierarchy) towards the top and children (or equiv) towards
> the bottom. To this list, it's not so important *how* it's laid out as
> that we provide the mechanisms for doing so.

Yes. Notecards had such a browser, but it logged *all* link types and used
'contains' links to build the hierarchy (although all the links were shown).
In the long run the 'contains' links (which forced users to put documents
in a 'folder') were abandoned and cited as the single biggest pain by authors.
But so much add on software lazily looked for 'contains' instead of author-
specified organizing criteria that effectively the 'contains' stuck hard...

> cyclical graph of the space you've *traversed*. The sooner

Doesn't IBM's OS/2 Warp browser attempt to do this ? I haven't seen it.

Craig Hubley                Business that runs on knowledge
Craig Hubley & Associates   needs software that runs on the net     416-778-6136    416-778-1965 FAX
Seventy Eaton Avenue, Toronto, Ontario, Canada M4J 2Z5