Re: Authentication/Authorization Issues: Re: [Geopriv] coming totermson location by reference

Hi Barbara

to simplify the description of the scenarios we indeed assumed that the
access network consists of link layer provider and IP provider. We just
assumed that these are the same for sake of simplicity. But you are
right that we would have to provide a more detailed description, for
example, in a document.

Ciao
Hannes

Stark, Barbara schrieb:
>> The DSL network has a relationship with the target;
>> the target is its subscriber.
>
> I thought the customer premises network used by the target is operated
> by a subscriber of the ISP, and the ISP is the subscriber of the access
> network provider. At least, that's how it works in our network. I do
> know of some cases where the access provider and ISP are the same, but
> I'd like to see a little more care in discussing who has responsibility
> for what, and who has a relationship with whom. In some HELD
> architectural discussions (such as in NENA), we had the target querying
> the ISP LIS, who in turn queried the access provider LIS. I'm curious as
> to what sort of architecture this discussion is leading towards.
>
> In the common wholesaling scenario, the access provider frequently does
> not have any sort of business relationship, at all, with the target.
> Often, either contractually or through regulatory rules, the access
> provider isn't supposed to provide any services (other than a physical
> pipe) directly to devices on the premises network. This is the ISP's
> subscriber, and NOT the access provider's. I'm curious how LIS discovery
> would work, if we were expecting the target to discover an access
> provider LIS, instead of an ISP LIS, since the ISP operates the DNS
> servers, and is responsible for all IP layer routing (including ports or
> any redirection rules).
>
> Also, the ISP often doesn't know whether or not it has a business
> relationship with the target. The ISP knows the account being used to
> access the ISP, but doesn't know what relationship the target has to
> that account. If a cafe is operating a simple free hotspot, the cafe is
> the ISP's subscriber, and any target devices may or may not be customers
> of that cafe. The ISP doesn't know the target. It doesn't know the
> target's identity (other than its IP address -- or is that what you mean
> by identity?) or its privacy policies. Also, I'm the ISP's subscriber in
> my household, but the ISP has no idea whether a target device is being
> operated by me or someone else. All the ISP knows is the IP address that
> it uses to communicate with that target. Multiple targets can use the
> same IP address, and my reading of things is that not all targets (using
> the same IP address) will have the same privacy policies. Or are we now
> moving privacy policy away from belonging to a target, to belonging to
> an IP address?
>
> Because of this, I think it's a really bad idea to state that all LISs
> MUST also be presence servers and support privacy policies. I think it's
> a good idea for a protocol to support transmission of privacy policy,
> but don't believe it should be a required function of a LIS.
> Barbara
>
>
>
> Barbara
>
> *****
>
> The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential, proprietary, and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from all computers. GA622
>
>
>

_______________________________________________
Geopriv mailing list
Geopriv@ietf.org
https://www1.ietf.org/mailman/listinfo/geopriv
Received on Wed, 13 Sep 2006 22:29:48 +0200