RE: full integrity?

From: Randall Gellens ^lt;>
Date: Thu May 23 2002 - 17:20:00 EDT

At 5:09 PM -0400 5/22/02, Richard Shockey wrote:

>> I was trying to get an idea of how wg participants felt about
>> accuracy adjustments being the primary method of disclosure
>> control, as opposed to also allowing multiple locations.
> Well my answer to that question is simple ..out of scope.

Maybe it would help if I phrased the question this way: Is it
necessary for the geopriv object(s) to be able to carry multiple
locations for the target?

> Now you are imposing policy on the objects as well

I don't see how my question could be interpreted as imposing policy.

> I think the best course of action for geopriv is to remain policy
> neutral to how applications may or may not use the objects in
> question

I agree that the less we need to do with policy the better off we'll
be. We may need to pay some attention to how our objects are used,
if only to address the privacy and security issues, and to make sure
that our work is capable of being used.

> would have really enjoyed the IETF debates on wiretapping
> several years ago :-)

I recall those debates well; that issue has relevance to our work
here, but not to the question I have tried to raise. (The
wiretapping issue relates to how much attention we pay to regulatory
and legal constraints.)

> If the policy can be expressed or defined in such a manner that
> the location server gives multiple locations then who are we as
> designers to say it shouldn't ..

Can our object carry multiple locations?

> however I am going to suggest that there may eventually be
> legislation that demands location servers give accurate information
> to special classes (Fire Police) of requesters if they can be
> properly authenticated.

That shouldn't be our concern. There can be all sorts of constraints
on policy (legal, contractual, who knows what), but I don't think we
need to worry too much about it, any more than the IETF worries about
lawful wiretapping. We design our protocols to resist eavesdropping
or other attacks, and let the implementers worry about adherence to
legal or contractual constraints.
Received on Thu May 23 17:28:00 2002

This archive was generated by hypermail 2.1.8 : Thu Jan 22 2004 - 12:32:23 EST