RE: [Geopriv] coming to terms on location by reference

So this is a statement in support of location by reference followed up
by a statement that it should only be the Rosen-preferred form of
reference.

To decouple the arguments then, I most certainly agree with the former.
For the latter, see all the arguments in the SIP conveyance discussions
to not restrain the form of the reference with prejudice. In particular,
the i2 architecture was not defined on the basis of *requiring* presence
to permit emergency calling to work. Talk about introducing unnecessary
elements in the emergency call process - despite claims to the opposite.
The LIS is not a presence server but the VPC needs to be able to query
it.

Cheers,
Martin

> -----Original Message-----
> From: Brian Rosen [mailto:br@brianrosen.net]
> Sent: Wednesday, 13 September 2006 3:08 AM
> To: 'Andrew Newton'; geopriv@ietf.org
> Subject: RE: [Geopriv] coming to terms on location by reference
>
> Thanks for trying to settle this issue which has nagged us for too
long.
>
> In my opinion, location by reference was defined when we accepted the
> notion
> that location could be included in a presence notification. If you do
a
> SIP
> SUBSCRIBE operation on a presence URL, you may get back a NOTIFY with
a
> PIDF-LO.
>
> If that is not location-by-reference, I'd like to know why not.
>
> I don't think the efficiency argument is particularly compelling.
It's
> true, but not worth much.
>
> I do worry about the 3rd party aspects of location by reference in the
> emergency case. We have tried pretty hard to make all of the
mechanisms
> for
> emergency fail safe, primarily by caching information in advance.
> Location
> by reference, generally speaking, spoils that effort. If an endpoint
gets
> a
> reference instead of a value, short of doing the dereference itself,
it
> can't do a LoST mapping and can't cache its actual location. When the
> emergency comes, if the dereferencer does not have connectivity to the
> reference host, it can't get location.
>
> There is the positive argument that a location reference does have
some
> efficiency when you have a moving target. Really, this is the
presence
> server argument, and it pretty much rings true, especially if there is
> more
> than one watcher. I'm pretty happy that location is part of presence.
>
> The real question in my mind is, is presence SUFFICIENT? Do we need
any
> other protocols? I think not.
>
> Brian
>
>
>
> > -----Original Message-----
> > From: Andrew Newton [mailto:andy@hxr.us]
> > Sent: Tuesday, September 12, 2006 11:18 AM
> > To: geopriv@ietf.org
> > Subject: [Geopriv] coming to terms on location by reference
> >
> >
> > As John Schnizlein has noted in a message
> > (http://www.ecotroph.net/~anewton/hypermail/geopriv/0607/2717.html),
we
> > have
> > not truly come to consensus on location by reference within this
working
> > group. The GEOPRIV working group has been presented with many
> individual
> > submissions for both architectural considerations and concrete
> solutions.
> > Additionally, this working group has discussed location-by-reference
> with
> > regard to its component uses in location configuration protocols,
> location
> > dereference protocols, and location conveyance protocols. As a
working
> > group, we are much further along in our understanding of
> > location-by-reference than we were just 1 year ago. And as it
stands,
> > there
> > are many who feel it is valuable, many who feel it is not, and many
who
> > have
> > not given their opinions. But it is now time for us to determine
how we
> > should move forward on location-by-reference, either accepting it as
a
> > whole, accepting it in limited form, or rejecting it entirely.
> >
> > To that end, I offer the following inexhaustive list of observations
on
> > this
> > topic:
> >
> > 1) Location-by-reference offers better efficiency where location of
a
> > target
> > is not determined by the target and when the target wishes to convey
its
> > location. It is more efficient because the target does not have to
> first
> > retrieve the location in order to convey it.
> >
> > 2) Location measurement and determination (sighting) of a target by
> > another
> > element is of limited scope as self-determination methods such as
GPS
> may
> > offer greater accuracy.
> >
> > 3) Authentication of the dereferencer may have unintended
consequences.
> > For
> > example, a referent host may require authentication that could cause
the
> > failure of emergency call routing. Therefore, location-by-reference
may
> > not
> > appropriate or could even be harmful for certain use cases.
> >
> > To further this discussion so that we may come to resolution on this
> > topic,
> > I welcome comments regarding the above as well as further
observations.
> >
> > -andy
> >
> >
> > _______________________________________________
> > Geopriv mailing list
> > Geopriv@ietf.org
> > https://www1.ietf.org/mailman/listinfo/geopriv
>
>
> _______________________________________________
> Geopriv mailing list
> Geopriv@ietf.org
> https://www1.ietf.org/mailman/listinfo/geopriv

------------------------------------------------------------------------------------------------
This message is for the designated recipient only and may
contain privileged, proprietary, or otherwise private information.
If you have received it in error, please notify the sender
immediately and delete the original. Any unauthorized use of
this email is prohibited.
------------------------------------------------------------------------------------------------
[mf2]

_______________________________________________
Geopriv mailing list
Geopriv@ietf.org
https://www1.ietf.org/mailman/listinfo/geopriv
Received on Tue, 12 Sep 2006 17:14:31 -0500