Re: Fuzzy Hypertext?

[dgd@xxxxxxxxx (David G. Durand)]

I've often read proposals along this line: and a lot of the automatic
linking work tends to generate strengths of association with link
suggestions. Peter Gloor's paper(s) in past ACM Hypertext conferences come
to mind as a starting point.

   The problem with "fuzzifying" links as part of the interfact for
user-driven linking is more a question of what you _do_ with the
information, and how you make a usable interface for creating it. In a
multiple-author environment, relveance ratings on links are likely to vary
too much. One proposal that has not been widely implemented, is
link-voting. The idea is that if people rate the relevance or quality of a
link, it will converge to a maeaningful value. More complex rating schemes
(parameterized by user-profiles, etc.) would also be possible. There are
have been papers about the rating of news articles that might be relevant.
I think maybe some of Tom Malone's group at MIT did this in Object Lens,
and follow-on systems, but it's been a while since I read any of those
papers.

   As for creating fuzzy anchors, I'm not sure what this would look like.
Spray paint? How would you control the density. Currently when people
aren't too sure about the scope of a link, they just pick a likely region,
and leave it to the reader to dismabiguate. Since the reader has a human
brain, this may be easier and more accurate than any automated process
would be.

   I can imagine fuzzy strengths and logical operations being useful for
fuzzy boolean querying, but it might be more effective to base things like
this on fixed or extensible keywords (maybe hierarchically organized).

   The hard part with this kind of extension is figuring how it will
operate for the user, not inventing conceptual models (of some sort). What
do you have in mind?

   -- David

I am not a number. I am an undefined character.
_________________________________________
David Durand              dgd@xxxxxxxxx  \  david@xxxxxxxxxxxxxxxxxxx
Boston University Computer Science        \  Sr. Analyst
http://www.cs.bu.edu/students/grads/dgd/   \  Dynamic Diagrams
--------------------------------------------\  http://dynamicDiagrams.com/
MAPA: mapping for the WWW                    \__________________________