I think that we
should provide some mechanisms to foster integrity and authentication of
source of location. I think that such mechanisms should be widely
deployed. I believe there are use cases where such information is
useful. I think ONE of those use cases is emergency calls. I think
it is not possible to provide assurances that are equivalent to current
wireline location for emergency calls. I think location integrity and
source authentication is work in geopriv, and not
ecrit.
Brian
From: James
Winterbottom [mailto:winterb at nortel.com]
Sent: Wednesday, August 03, 2005 6:31
AM
To: Brian Rosen; 'Andrew
Newton'
Cc:
ecrit at ietf.org
Subject: RE:
[Ecrit] some observations from yesterday wg
meeting
Brian,
Let me
make sure that I have a clear understanding of your point here.
I as a recipient of location
information, that provide a service in good faith, have no right or come back,
if the location provided to me is incorrect?
Please
explain to me how this is a step in a positive direction, and how this will
bolster adoption of IP telephony?
I think
that people will accept "near-equivalency", but that they will not accept out
and out lack of equivalent service.
Cheers
James
-----Original Message-----
From: Brian Rosen [mailto:br at brianrosen.net]
Sent: Wednesday,
3 August 2005 8:10 PM
To: Winterbottom, James [WOLL:5500:EXCH]; 'Andrew
Newton'
Cc:
ecrit at ietf.org
Subject: RE: [Ecrit] some observations from yesterday
wg meeting
Chairs?
I was not
a fan of the narrow focus of the charter. It is what it
is.
Each
change in technology changes the situation for emergency call systems. It
takes away things, and provides new things. When mobile phones came,
they removed accurate location and substituted inaccurate location in a
totally different, and difficult-for-psap-to-use form. They even took
away call back number in the case of un-initialized phones. They gave the
possibility of tracking a moving caller, and they gave the ability to get
calls immediately from the scene of common events like car crashes. On
balance, it's a good thing.
VoIP is
the same; some things are being taken away, some things are being added.
It's different, get over it.
Brian
________________________________________
From: James Winterbottom [mailto:winterb at nortel.com]
Sent: Wednesday,
August 03, 2005 5:46 AM
To: Brian Rosen; 'Andrew Newton'
Cc:
ecrit at ietf.org
Subject: RE: [Ecrit] some observations from yesterday
wg meeting
Bah!!!
The problem must
exist in someone's charter, and GeoPriv don't seem to want it either!
It is not easy,
without having relatively substantial local foot print, to spoof emergency
calls with in the USA today. In a IP world, that is
simply not true any longer with the currently proposed mechanisms of user
created location documents, no matter where the source data came from. That is
consumers of location (location recipients LRs) have absolutely NO means of
assuring that this correct. The same is absolutely not true for 99.999% of
PSTN calls.
Location
is today in the cellular and wireline world is the responsibility of the
access provider, and this will need to remain the responsibility of the access
provider in the future until telekenic-osmosis becomes a reality. ECRIT or
GeoPriv pick it up until then, but stop passing the
buck!!!
Cheers
James
-----Original Message-----
From: ecrit-bounces at ietf.org [mailto:ecrit-bounces at ietf.org] On
Behalf Of Brian Rosen
Sent: Wednesday, 3 August 2005 7:21 PM
To: 'Andrew
Newton'
Cc:
ecrit at ietf.org
Subject: RE: [Ecrit] some observations from yesterday
wg meeting
But that
problem is not in our charter. If an attacker learned a PSAP's SIP URI
from any mechanism, they can mount an attack. True.
Not in
our charter.
Determining that a location represents the location of
the caller is within charter.
I believe identity of caller is NOT a requirement in
general, but I think it's clearly not a requirement in ecrit.
Brian
-----Original
Message-----
From: Andrew Newton [mailto:andy at hxr.us]
Sent: Wednesday, August 03, 2005 5:07 AM
To: Brian Rosen
Cc: 'James Seng';
ecrit at ietf.org
Subject: Re: [Ecrit] some observations from yesterday
wg meeting
On Aug 3,
2005, at 3:23 AM, Brian Rosen wrote:
> This discussion is more or less out of charter,
as the work in ecrit
> addresses the first point above.
I don't think
so. I think James has made a fairly important point:
the notion that
L2 location must be signed can simply be circumvented
by an attacker
contacting the PSAP's SIP server directly.
-andy
_______________________________________________
Ecrit mailing
list
Ecrit at ietf.org
https://www1.ietf.org/mailman/listinfo/ecrit