W3C home > Mailing lists > Public > www-talk@w3.org > May to June 1995

VALUES: Let them Adjust

From: Ken Meyering <immedia@netwest.com>
Date: Tue, 27 Jun 1995 14:34:22 -0700
Message-Id: <199506272136.OAA05837@saguaro.netwest.com>
To: "Ronald E. Daniel" <rdaniel@acl.lanl.gov>, bede@scotty.mitre.org, www-talk@www10.w3.org
Cc: immedia@netwest.com, nazgul@utopia.com, peterd@bunyip.com, marc@matahari.ckm.ucsf.edu, michael@junction.net, rating@junction.net, uri@bunyip.com
I'm opposed to any system that requires any effort on the part of the
content-providers to "label" or "classify" their material.

People should not be required to attach other peoples value judgements to
their own expressions.  This is just another way of saying, "BE INHIBITED",
"THINK LIKE US", "SAY IT OUR WAY".

What you think of my http server is *your* business, and is subjective.

There are already giant list-building/selling services that maintain
"orwellian overviews"... however, John Q. Fundamentalist isn't aware of them
yet.

In a "Free-Censorhip Market", where any number of list-providers can serve
up ratings, values, vectors, recommendations, etc., no effort is required on
the part of the content providers, and the public will "learn faster" by
looking at different lists and their relative opinions.

Eventually, I think, people will gradually desensitize to material that
society previously considered taboo.  What was once "shocking" will become
"just information".

When you consider child development, (and most of our personal histories),
or sensitivities are just a product of our experiences and identification,
and as we mature these sensitivities change, hopefully our tolerance
increases as well.

===

I'm interested in Expert Systems applied to Computer-based Training.  One of
the accepted requirements of an expert system is that it be able to
"Explain, in plain terms, it's reasoning".

Similarly, if a third-party rating service wishes to gain the respect of the
public, it should also be required to explain it's ratings.  

Some systems will simply GENERALIZE, for example exclude any domain which in
the past has posted a photograph considered objectionable to that's systems
values.  This would be the equivalent of an "unforgiving black-list".  Other
systems will incorporate time-domains, for example, they will allow ratings
to change with time..."forgiveness".  Some systems may actually have humans
that review an individual file.  

Initially, the sheer number of systems will be confusing, and the methods
those systems use will also be confusing.  But over time, the "whole system"
and all of it's "value servers" will become a mirror of a "thinking,
reasoning machine".  It will be able to give advice, solve problems, and
respond to natural language queries.

Critical to the success of the "intelligence" of the whole system is that it
be "all inclusive"...not requiring any information provider to "self-judge".


For those who think that such a system will be cumbersome, I suggest the
following:

   1. Most parents who currently own computers are not terrified by
      information, and therefore will "think for themselves" and not use
      third-party value filters.

   2. Those who are frightened by shocking pictures and words can choose
      to subscribe to third-party value servers.  If they are irritated by
      the slowness of the "mediated transaction", they can help subsidize
      higher bandwidth networks and distributed value services.


Many will suggest that self-rating protocols etc. are more efficient for the
clients.  Of course.  But then the clients are not "contributing", they are
simply passive consumers of information.  That is not what the interactive
mediums are about.  

As for third-party systems being complex and cumbersome.  Of course!  So are
morals.  If they choose to be hyper-cautious, then they probably won't mind
a slight slowing of their access.  Chalk-it up to "playing it safe".

The third-party "Free Censorship Market" will itself be an educational exercise.


Will someone please write a public domain Windows program and a
public-domain Macintosh program (prototype client-based filter, that CNN,
ABC, NBC, CBS, BBC can show to anyone who cares) that is capable of:

   "intercepting" URL socket requests by Browsers and News Readers
   demonstrating a mediating process (could be as simple as looking up
   the numerical address then checking for a match in a local list)
   proceeding to "allow" or "dissallow" the transaction.

Let this be an educational exercise

   Let this little mediating agent "check references"
   Let it be basic-like

Let it be invisible (background mode), or visible (pop-up mode)

Release the source code...

Let the whole exercise be an education to the public.

Ken Meyering
immedia@netwest.com
http://www.netwest.com/~immedia
Received on Tuesday, 27 June 1995 17:34:50 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 27 October 2010 18:14:17 GMT