RE: [Geopriv] Location in SIP and "retransmission-allowed"

From: Brian Rosen ^lt;br@brianrosen.net>
Date: Fri May 04 2007 - 13:52:55 EDT

> >
> > A SIP UA sends its location to a proxy server, which takes
> > the LI out and sends it to LoST.
>
> What makes the LO 'its location'.
Who is lawyering now? The point of the exercise is to route the call based
on the location of the device. When the proxy sends the LI (not the LO) to
LoST, it knows it's the location of a device (it has the LO). The LoST
server may not know what it is, of course.

>
> What the proxy is sending
> > is the LI of a device.
>
> Not if you remove the pieces that makes the LO 'its location'.
Yes it is, it's the LI of the device. That is the definition of LI, LO
without identification. The proxy knows it is doing this.

>
> To me, that is exactly what the
> > current 4119 language says you can't do if retransmission-allowed=no.
>
> Agree if the proxy server were to include device/entity identifiers in the
> transmission to LoST.
But that is not what 4119 says. It says LI or LO. It is quite specific
about it. I can understand arguing to change that language. I don't
understand arguing that it doesn't apply.

>
> >
> > I'm NOT really arguing one way or another about whether this
> > should be allowed or not. I'm trying to figure out if you
> > want a normative change to 4119. I keep reading your
> > messages as being "NO, I don't think we need to normatively
> > change 4119". I'm trying to figure out how you do that.
>
> Obviously some are trying to explain common sense and others are trying to
> lawyer 4119.
>
> Which brings me back to my original question for the lawyer:
>
> What makes a LO a PIDF-LO?
Adherence to 4119

>
> or
>
> Why is a LoST query/response not covered by 4119?
Isn't this something roughly like removing the author's name and copyright
on a document, and then publishing it?

Once you remove the identity AND THE RULESET from a PIDF-LO, you can do
anything you want with it, isn't that what you are saying?

I do understand the point you are making: without identity (and ruleset) a
location is not private, it's just like having a map and finding a place on
the map. I don't think you can so easily say if I remove the identity and
ruleset from a location I can do anything I want with it.

Let me cite an example. Suppose the LI had interior room information in it,
information not generally known by the public. Do you attach ANY privacy
concerns to that? Your argument is no, once I take the identity and ruleset
out, I can treat it as completely public information. I think that is
wrong.

_______________________________________________
Geopriv mailing list
Geopriv@ietf.org
https://www1.ietf.org/mailman/listinfo/geopriv
Received on Fri, 4 May 2007 13:52:55 -0400

This archive was generated by hypermail 2.1.8 : Fri May 04 2007 - 13:53:04 EDT