Re: [Geopriv] Geopriv L7 LCP: New Requirement

From: Hannes Tschofenig ^lt;Hannes.Tschofenig@gmx.net>
Date: Sat Feb 10 2007 - 15:53:13 EST

I agree that this is a fuzzy term.

I wonder whether we can make it a bit more precise.

Two conditions seem to be necessary for the mechanism to work:
* The OBO-Client needs to learn the identifier. For some identifiers
that already requires being quite close to the Target.
* The OBO-Client needs to be authorized to perform the
identifier-to-location information resolution step.

I think that the latter aspect is the restrictive part here.

Ciao
Hannes

Henning Schulzrinne wrote:
> Right. But 'having a relationship' with the network means very little.
> I have a relationship with Verizon DSL. Should I be able to get the
> addresses of all other Verizon DSL customers?
>
> On Feb 10, 2007, at 3:40 PM, Hannes Tschofenig wrote:
>
>> Hi Henning,
>>
>> some terminology first:
>> * Target -- RFC 3693
>> * Location Information Server (LIS) -- entity that knows the location
>> of the Target
>> * OBO-Client -- entity that requests the location of the Target
>> without being the Target itself.
>>
>> If I understood the requirement correctly then it is
>> * applicable only if the OBO-Client is in the same network as the
>> Target or has a relationship to it,
>> This aspect is necessary since (a) the security issues are
>> difficult to handle and (b) the identifier might not be resolvable
>> outside this "realm".
>> * uses identifiers like the VCI/VPI (and not the URI)
>>

_______________________________________________
Geopriv mailing list
Geopriv@ietf.org
https://www1.ietf.org/mailman/listinfo/geopriv
Received on Sat, 10 Feb 2007 21:53:13 +0100

This archive was generated by hypermail 2.1.8 : Sat Feb 10 2007 - 15:52:34 EST