[sc34wg3] Newcomb on "Where WG3 should be going"

[Jack Park]

A tiny contribution:

Patrick, Steve, and Michel have stimulated a lot of deep thought. I'd
like to add an observation, not necessarily a complaint.  It is that I
have discovered that it is frequently useful to switch to a browser
where I can clear the cookies when I want to use Google. That's
because Google's marvelous agents think they know what I'm looking
for, which is, in fact, rarely the case. Most of the time, I don't
want to see search hits I've seen before.

That compounds a slightly different issue raised by search algorithms
that are, largely, beauty contests. They downplay novelty. In the
past, I found it useful to reach deep into the search hit list just to
see something new.

Were the web organized by topic maps, or what Steve calls subject
maps, I'd like to think I'd have a better chance of spotting novel
occurrences. I'd like to think I could set an agent to alert me to
changes in topics of interest.

My own work these days is evolving an open source subject map platform
I call TopicSpaces. My goal is to place it at the core of what Dino
Karabeg and I, together with many others, call a knowledge federation.
 We seek to conduct some of the experimentation Patrick mentioned.

Jack