From: Rich Graves <llurch@networking.stanford.edu>
Date: Sun, 28 Jan 1996 10:57:41 +0800
To: Bill Frantz <frantz@netcom.com>
Subject: Re: This post is rated LTC for `Low Technical Content'
In-Reply-To: <199601280200.SAA24044@netcom6.netcom.com>
Message-ID: <Pine.ULT.3.91.960127182733.27229L-100000@Networking.Stanford.EDU>
MIME-Version: 1.0
Content-Type: text/plain


On Sat, 27 Jan 1996, Bill Frantz wrote:

> At  7:55 PM 1/27/96 -0500, JMKELSEY@delphi.com wrote:
> >The best solution has always seemed to me to be one of these three:
> >
> >a.   Tags appended to notes/posts, from various reviewers, digitally
> >signed and otherwise coded to allow intelligent filtering, or
> >
> >b.   Electronic distributions of reviewers' evaluations tagged to
> >notes in some simple way.  (I.e. give each note or post a unique ID
> >which appears in the message.)  Then, a smart newsreader/mail
> >program sorts the notes accordingly, or
> >
> >c.   The reviewer reads the group/list, and rates posts according to
> >some useful criteria.  He then resends it out to his users, filtered
> >as desired.  (CP-LITE seems like a very early version of this.)
> 
> d. The "V-Chip" device makes a network query to the selected rating service
> to ask for a rating.  What happen when the rating service is unreachable is
> just one of the many parameters that the parent needs to set.  (If designed
> right, no parent could use it, but its availability would still stop the
> adult censorship croud in congress.)

This just gets ridiculous. It adds a lot of overhead without necessarily 
giving you good information.

On the Net, there is no longer any real difference between underground 
and mainstream data. It's all just as easy to get. You can't block it.

You're thinking like engineers. This isn't an engineering problem; it's a 
social and artistic problem.

Actually, it's two problems: how to censor people, and how to find stuff 
you're interested in.

Censorship only works if it's dictated and enforced. Ratings don't cut it.

Arbitrary scales can't judge stuff you're interested in. Art cannot be 
reduced to numberical criteria. 

A while ago, and maybe it's still going on, the MIT media lab had an 
interesting music rating service. The way it worked was, you submitted 
stuff you liked, lots of other people submitted stuff they liked, and the 
computer generated a list of stuff that you might like based on apparent 
matches among different people's tastes.

The model to emulate is a computer dating service, not a library.

-rich