W3C home > Mailing lists > Public > www-rdf-interest@w3.org > March 2000

RE: Subclass of Thing/Resource

From: Bill dehOra <Wdehora@cromwellmedia.co.uk>
Date: Tue, 7 Mar 2000 15:19:41 -0000
Message-ID: <AA4C152BA2F9D211B9DD0008C79F760A5CA484@odin.cromwellmedia.co.uk>
To: "'Greg FitzPatrick'" <gf@medianet.org>, Marja-Riitta Koivunen <marja@w3.org>, Dan Brickley <Daniel.Brickley@bristol.ac.uk>, Guha <guha@epinions-inc.com>
Cc: www-rdf-interest <www-rdf-interest@w3.org>

:as represented by a trusted 3rd party agent
:	meeting your legal obligations as a citizen/member of something
:	exposing yourself to law enforcement (contentious)
:as represented by a trusted 3rd party agent
:	in buying and selling goods and ideas.
:		in transactions notarized by 3rd party agents
:			maintaining tax revenues (by consent 
:as represented by a trusted 3rd party agent
:	in communicating with
:		family
:		friends
:		strangers
:		interest groups
:		employers
:		doctors & lawyers
:		etc.
:whereas each possible communicator has varying access to you

But they don't have varying access to *me*. They have varying access to
layers of software that surrogate *for* me (that why it's called an agent),
assuming no layer has not been coopted. That is very different. Each added
level of indirection is a security risk.

:As a shopper it is in your advantage to let the seller know as 
:much about yourself as possible.  

That's debatable in scenarios involving commodities, and highly debatable
where the consumer has a clear picture of what they want, or where
bargaining is possible. It might be relevant for the sale of customised

:You would be more willing to give this 
:information if you were assured that your 
:identity was never to be divulged and 
:you would be more secure in your privacy 
:knowing that your agent and not the seller
:was allowed privy to your true identity.

How might we be 'assured' of security in such matters? Consequentially how
can we be assured of anonymity in such transactions? While it may be
inconvenient that people's data is dislocated, is limits damage to them in
the event of identity theft. 

The idea that centralised digital identities are resources that are not
going to be abused or cracked is frankly naive, and I cannot imagine an
assurance sufficient in this scenario, unless security is built in from the
very beginning, which is certainly not the case with rdf or the semantic

Received on Tuesday, 7 March 2000 10:20:18 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:07:29 UTC