[teknoids] Cool Tool Monday
Thomas R. Bruce
trb2 at cornell.edu
Tue Dec 12 04:05:16 EST 2006
Jonathan Franklin wrote:
> Perhaps soon case research will be done this way - see www.precydent.com
This is interesting, and looks promising but not all the way there yet.
I tried playing around with "fair use" as a search. After about 5
seconds of intensive study, I *think* it does what I might have
predicted, which is unduly favor cases in the "obligatory cite" category
-- that is those not of much relevance to the matter at hand, but which
everyone writing on the topic feels obligated to nod to. I'd like to
see a few others try similar stuff, just to see if you have the same
impression. There's been some interesting work on citation networks in
scientific literature that somehow teases out obligatory cites and
weights them differently.
The really interesting question is whether co-citation clusters revealed
by this kind of approach actually map to cases that are conceptually
(not factually) related. An interesting question, eg., would be whether
the array of clusters at some level of resolution maps to (eg.) a set of
West key numbers. In a rough way.
They sure do need a new UI designer, though. And I want to know what
"Web 3.0" technology is. I'm still stuck around 1.74 somewhere.
Thomas R. Bruce (trb2 at cornell.edu)
Director, Legal Information Institute
Cornell Law School
More information about the Teknoids