----- Original Message -----
From: "Christian Hauser" <lobase@gmx.net>
To: <geopriv@mail.apps.ietf.org>
Sent: Friday, July 20, 2001 7:35 AM
Subject: Abstraction from existing technologies
> First of all, as this is an IETF-WG, we seem to focus too much on UMTS and
> other mobile (and already existing) technologies. Im my opinion, we should
> discuss issues regarding any imaginable platform, because future systems
will
> bring much more sophisticated features regarding location support than
> nowaday's. I think, considering only current technologies would restrict
our minds
> too much.
I agree. The privacy framework needs to have universal applicability. I
guess more discussion has been focused on UMTS becuase it is the most
obvious example.
> For that reason, I wonder if it would help to just assume one arbitrary
> party having the users' location information. Whether this is the location
> database of a UMTS provider, any other third party (e.g. a new Location
Service), a
> combination of those or something else is not very relevant in my opinion.
> This entity maintains all the location information, arising about the
users
> (e.g. the location information from the mobile network operator, from the
> users' GPS device, from any other locating infrastructure like Active
Badges or
> whatever is imaginable).
With the user roaming through hetrogeneous networks operated by different
service providers, how will trust relationships be defined between the user
and the service providers and between the service providers themselves. I
think the privacy of privacy policies is going to be another issue.
> This entity would know very accurate location information about the users.
> Because that information has to be prevented from being linked to the
users'
> identities (as I depicted in my mail from July, 12th), this Location
Service
> must be usable pseudonymously.
The problem is that where ever the mobile user goes, he has to be
authenticated first. For example, if the user moves from a UMTS network
outside to a WLAN network inside a building, he will need to be
authenticated by the WLAN operator, which will reveal his identity and
location to the WLAN operator.
> So, what I would propose is, that we try to find privacy issues and
> mechanisms / principles to deal with these issues, regardless of any
possible (or
> existing) implementation in order to find universal solutions being
applicable
> for whatever future will bring.
Agreed.
----------------------------------------------------------------------------
--------------------------
Shahid Shoaib +1-408-451-4740 Direct
DoCoMo USA Labs Inc. +1-408-573-1090 Fax
shahid@dcl.docomo-usa.com http://www.docomo-usa.com
Received on Tue Jul 24 14:08:42 2001
This archive was generated by hypermail 2.1.8 : Thu Jan 22 2004 - 12:32:22 EST