Announce: The GroupLens Trial

Announce: The GroupLens Trial

[Brad Miller]

	  Are you tired of all the junk on Usenet these days?
      Be part of a project that could change the face of USENET!
	 Here's your chance to help make USENET useful again.

	 ----------------  What is GroupLens?  ----------------

In a nutshell, GroupLens is a collaborative filtering system that
helps you work together with other people to find the quality news
articles out of the huge volume of news articles generated every day.

To accomplish this the GroupLens system combines your opinions about
articles you have already read with the opinions of others who have
done likewise and gives you a personalized prediction for each unread
news article.  The prediction is on a scale from 1-5, and indicates to
you how likely you are to find the article useful.

I am Brad Miller, a graduate student in Computer Science at the
University of Minnesota.  I have been working on GroupLens as my
thesis project with professors John Riedl and Joe Konstan, and our
collaborators Paul Resnick of MIT, and David Maltz of CMU..  We have
been working on GroupLens for two years, and are ready to run a test
of the system involving 10,000 users.

For much more information, or to sign-up, please visit our web page at 

		http://www.cs.umn.edu/Research/GroupLens
 

   ----------------  Why should *I* use GroupLens?  ----------------

Because Usenet doesn't work any more!  Too many people posting too
much junk have clogged the net.  But, "put simply, Usenet is the
largest information utility in existence" [Harley Hahn,"A Student's
Guide to UNIX"]!  We each need help selecting the nuggets of useful
information from the muddy sluice of the Net.  GroupLens is a filter
that identifies the high quality articles that will be of interest to
*you*.

The more users we get the better GroupLens works! We have been testing
the system for several weeks with about 25 users, but are now ready to
add many more.  With 25 users we are getting good results, but in high
volume newsgroups we need more people.  The more users, the less we
each need to read to be sure we're discovering all the good stuff!


       ---------------- Hey!  It's easy, and fun  ----------------

We have tried to make GroupLens as easy to use and as unobtrusive as
possible.  In all cases the number of extra keystrokes you need to
type in order to rate news articles is zero or one!  In return you get
predictions from GroupLens that tell you how good an article is.

But what about PRIVACY?  GroupLens lets you choose your own pseudonym
for your ratings and predictions.  Noone -- not even the GroupLens
servers -- knows who "SpacemanSpiff" *really* is.

You can also subscribe to the GroupLens mailing list by sending mail
to grouplens-request@cs.umn.edu.  This list is for general discussion
about GroupLens, and may be helpful.


 ----------------  What is the Future of GroupLens? ----------------
   ----------------  What is the Future of Usenet? ----------------

We are currently running GroupLens on a few groups with just 10,000
lucky people, to collect data on the best collaborative filtering
methods for figuring out which news you like to read.  After a two to
three month test period, we plan to give away our software so
GroupLens can run for as many people, and on as many newsgroups as
possible.  GroupLens has the potential to shape the future of the
Internet!  This is your chance to help shape that future.


	   ----------------  To Join  ----------------

Get the GroupLens version of one of the following news readers [or modify
your favorite reader -- see footnote 1]:

  - gnus-5 under emacs or xemacs
  - tin
  - xrn

Join as a GroupLens member to read and rate news articles in one or
more of the following groups [or propose your favorite groups -- see
footnote 2]:

  - comp.lang.c++
  - comp.os.linux.*
  - mn.general
  - comp.groupware
  - rec.humor
  - rec.food.recipes
  - rec.arts.movies


(1) We tried to choose newsreaders with a large user base.  On the
other hand, if there is another newsreader that you think should be on
the list maybe you'd like to get to know the GroupLens client library!
The client library does most of the work needed to add GroupLens
support to any newsreader.

We are particularly interested in finding a windows based newsreader,
if you know of a good, popular newsreader for windows, and we can get
the source code, let us know!

(2) Due to space and resource limitations, we cannot support
everybody's favorite newsgroup.  On the other hand, we'd like to have
a small set of newsgroups with wide appeal.  Let us know if there is a
particular newsgroup for which you would especially like to use
GroupLens.

----------------------------------------------------------------------------
Brad Miller                    | e-mail: bmiller@cs.umn.edu
University of Minnesota        | phone:  (612) 626-8396     
Department of Computer Science | www:    http://www.cs.umn.edu/~bmiller
EE/CS 5-244                    | 
----------------------------------------------------------------------------