From: Simon McClenahan (SMcClenahan@ATTBI.com)
Date: Tue Sep 24 2002 - 10:33:23 MDT
This whole discussion about the technical and social dynamics of
collaborative filtering has been discussed at great lengths on the Slashdot
forums and the meta section of http://kuro5hin.org . No doubt there are
other discussion board systems, but to my knowledge these two are the most
popular that involve peer review of submitted comments. Blogging is supposed
to be a phenomena now, the next Internet killer app, at least if you believe
the propaganda that the blogging community writes about themselves.
The point I'm trying to make is that the discussion so far on collaborative
filtering, peer review, voting, anonymity, etc. have been discussed,
implemented and discussed again by at least the Slashcode and Scoop
developers, and the blogging community. I hardly think this topic is worthy
of SL4 discussion, unless you are talking about a real Earthweb
implementation (unfortunately I haven't read that book yet).
The latest KurzweilAI.net newsletter pointed to a Wired article
http://wired.com/wired/archive/10.10/start.html?pg=2 titled "Slaves to Our
Machines". The goal for most of us is to enable computer-assisted human
intelligence, not necessarily the current trend of human-assisted computer
intelligence. My view is of complementary assisted intelligence, especially
if an artificial system gains some sort of consciousness, why should humans
have to take on the role of slave or slave-driver? Can't we all just get
along? There I go with my anthropomorphism again, sorry.
Back to the collaborative filtering issue, the intelligence of that system
still lies within the humans. A tool that rearranges it and renders
different views, no matter how sophisticated, is still a dumb
(unintelligent) application whose value is only a factor of the intelligence
of the humans entering the information into it. These tools already exist,
more or less. What does not exist is an intelligent system that can parse
human thoughts (encoded in natural language text) and detect ambiguities,
incorrect reasoning, objective versus subjective statements, thought
patterns, etc. That would be a super job for a GAI.
IIRC, WebMind had an application that parsed the stock message boards to
determine the general consensus on whether people thought a stock was a good
buy or sell. A system that can detect arbitrary ideas from many human
sources seems to me like a general pattern recognition problem rather than a
specific pattern recognition such as "buy" or "sell".
----- Original Message -----
From: "Mitchell Porter" <firstname.lastname@example.org>
Sent: Monday, September 23, 2002 6:05 AM
Subject: anonymous collaborative filtering
> Run Slashcode configured so that no one can create an account.
> That way, everyone is an Anonymous Coward.
This archive was generated by hypermail 2.1.5 : Sun May 19 2013 - 04:00:53 MDT