Presence and IntComm WG

["ATD Munro, Electrical & Electronic Engineering" <>]

Hi All,

>   You all are way ahead of me.  As I see it this group's task is to inform
> the debate among stakeholders and perhaps the engineers who will implement
> PIC/ALS with technology reality, not to engage in the tussle itself.

I'm with Jeremy on this. We need to be clear where true conflict arises and 
where it does not. The discussion is interesting and informed (and 
informative). It seems to be going direct to technology with concrete ideas 
about the conflicts, and I think we risk coming up with solutions that miss 
the tussle even if they are exciting technologically.

I'll try to summarise my take on the social context of the privacy tussle.

We have emerging technology that can register presence and provide 
information about location and identity. Cellular networks have this 
capability too, so we aren't on our own, but we are able to experiment on a 
small scale while they have to do it system wide.

It has become a requirement in the US that a terminal (fixed phone, 
cellphone, etc.) can be located, ostensibly to improve the efficiency of 
emergency services. This is not an invasion of privacy in principle because 
the phone and its user aren't intrinsically linked: it is only the identity 
of the line or the handset that are associated with the location. So I 
don't see any potential for conflict so far.

Cellphone operators can also tie the subscriber to the phone identity in 
various ways (it varies from country to country) and this reduces anonymity 
and privacy. So a conflict is beginning to emerge. In Europe, I think it 
would be the norm to withold location information from a called party.

In the PIC case, I think the conflict is more tangible because, as far as I 
know, (please correct me if mistaken), the SIP user is identified by 
(roughly speaking) a "meaningful" name/address, which is carried in 
protocol between the clients/servers in the system. This seems to me to be 
a greater challenge to privacy and anonymity. If the protocol carried only 
IP addresses or machine identifiers (e.g. MAC addresses) then this conflict 
might be reduced, but at the expense of potential loss of reachability and 
increased signalling traffic. This refers back to one of Clark's examples, 
I think, in the loss of transparency in relating IP addresses to persistent 
objects.

This doesn't describe the tussle itself very precisely but I feel I'm 
getting close to the issue: we want reachability but we are moving towards 
a paradigm that exposes people's identity, possibly in a misleading way, 
more for expediency than for very good technical reasons.

By using identifiers such as, e.g.  as handle in 
IP telephony, are we actually opening up more outcomes, or closing down a 
larger number of opportunities? Doing it this way may be perceived as an 
aid to law enforcement or an opportunity for subversion, but we need to be 
clear about the outcomes that we want. These outcomes are the results of 
the tussle resolution.

Regards, sorry if this is a bit abstract,

Alistair

--
Prof. Alistair Munro, Toshiba Professor of Communications Networks
Dept. Electrical & Electronic Engineering, Bristol University
Merchant Venturers Building, Woodland Road, Clifton,
Bristol BS8 1UB
E-mail: 

Tel.: +44-7974-922442