1996-05-11 - Re: Mandatory Voluntary Self-Ratings

Header Data

From: “Vladimir Z. Nuri” <vznuri@netcom.com>
To: “Joseph M. Reagle Jr.” <reagle@mit.edu>
Message Hash: 7bb6bb4822ea3408b05509673c1a70d70df13acb866202c372d53a755f1fd1e3
Message ID: <199605110408.VAA08516@netcom13.netcom.com>
Reply To: <9605092113.AA13795@rpcp.mit.edu>
UTC Datetime: 1996-05-11 11:21:44 UTC
Raw Date: Sat, 11 May 1996 19:21:44 +0800

Raw message

From: "Vladimir Z. Nuri" <vznuri@netcom.com>
Date: Sat, 11 May 1996 19:21:44 +0800
To: "Joseph M. Reagle Jr." <reagle@mit.edu>
Subject: Re: Mandatory Voluntary Self-Ratings
In-Reply-To: <9605092113.AA13795@rpcp.mit.edu>
Message-ID: <199605110408.VAA08516@netcom13.netcom.com>
MIME-Version: 1.0
Content-Type: text/plain



JR
>        In fact, 3rd party services may have problems with large and dynamic
>WEB sites (in which case they just might rate it high, and rate the whole
>directory.) (I was thinking about this with regards to incorporating rating
>systems into WEB site managements tools and apps...) If MICS and signatures
>do become prevalent, an easy way I can defeat ratings I don't like (or to
>keep from others rating me) is to repeatedly change my content in some
>simple way, throwing off their MICS.

the idea with the rating system is that the rating signs the signature
of the page, which is itself digitally hashed or something. in other
words, the rating is on the "state" of a page at some time. the system
would at least be able to detect a change in the state of a page,
and inform the user that a rating may no longer be valid due to obsolescence.
but you are correct that page changes are probably more problematic
for market ratings than self-ratings.

it is true that BOTH self-ratings and market ratings have major problems
associated with them. the question is, which has the fewest for a given
situation? if page designers are going to maliciously misuse rating
systems, then the market type system is superior. the market system
does suffer from the problem that it is less decentralized. however
it is possible that some rating services might be able
to economically justify entire armies of rating teams.

it is clear some key questions about ratings are as follows:

1. what pressure or coercion, if any, will be placed on page 
designers and by whom for certain self-ratings?

2. will self-ratings be deliberately misused by people protesting the
system? will it be a problem?

3. will page revisions make market ratings unviable?

all of these will become more apparent as implementatoins proliferate
more widely.

again, PICS supports both in theory, so I have no objections to PICS
and am fact have been supporting it here.

I suggest that we let the market decide which works better-- market
ratings or self-ratings. I suspect they will both coexist in the future.
trying to a priori argue which will be more problematic seems a bit
naive to me. market ratings might make more sense on more formal pages,
such as reference material that is likely to be steady over time.
self-ratings would be a good fallback if no other information is
available.

as far as page changes, I don't think the web has a good mechanism
for handling changes in its contents right now. improved methods
of handling this kind of thing in the future may make the rating
problem less difficult. for example, if there was a "systemized"
way that a web page could point to a new address it has relocated
to, so that everyone that runs their "checker" programs and hits
the old page would update the link, etc., this could be incorporated
into the rating system to handle one common kind of change.

another possibility is for people to put in information into their
pages about expected "shelf-life"-- this would help ratings agencies
avoid rating places that are not stable.







Thread