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

Dawson, Martin wrote:
> 3) Authentication failure at the VPC, the emergency gateway, or indeed
> at the call server may mean an emergency call cannot be made at all -
> not even with default routing (which is at least possible without
> location). Citing scenarios involving error conditions, failures, or
> compromises through exploits are not valid arguments against any system.
> If it were one could argue that the entire IP protocol suite and the
> Internet should be disposed of as unacceptably risky.
>
> As mentioned in a previous post of mine - the non sequitur one -
> location by reference is of vital importance to the i2 architecture
> defined by NENA. It is the only mechanism which will permit location
> updates to occur for mobile IP access networks. If it is not just
> useful, but has an absolutely defined application right now, why would
> the IETF somehow attempt to forbid it?
>
> To my previous post, we can also add the benefit that the application
> provided with a location reference can ask for location in the desired
> form - e.g. in jurisdicational/emergency civic, rather than postal
> civic, or in geodetic. As opposed to the client device needing to divine
> this information or provide every possible form of location in the
> PIDF-LO to every application.

What about combining by-value with by-reference? Then the receiver:
1. gets a reference to obtain location updates and the preferred
location format, and
2. if 1. fails, there is still a location-by-value

Of course this adds complexity to the client but IMO there is a benefit.

regards
Klaus

_______________________________________________
Geopriv mailing list
Geopriv@ietf.org
https://www1.ietf.org/mailman/listinfo/geopriv
Received on Wed, 13 Sep 2006 17:21:25 +0200