Re: RE: about the Home Network Models WG item

Sounds good to have the sections labelled

Basic
Extended
Aggregated
Virtual

Would like to see what Pascal says though.

At 06:58 PM 1/5/2005 +0100, Alexandru Petrescu wrote:
>Pascal Thubert (pthubert) wrote:
>[...]
>>please present on this list at this time, and if you get enough support, 
>>I'll do the changes as required.
>
>I do not think in the simplest Home Network configuration the MR may use
>a Home Address derived from the prefix assigned on its moving network
>link, the MNP.  In the simplest cases, the MR Home Address is derived
>from the prefix valid on the Home Link.
>
>Or, section 4.1 Extended Home Network configuration has a last item saying:
>>Alternatively, a Mobile Router could also form a Home Address from
>>one of its prefixes and use it to register[...]
>
>I think this kind of item should belong to other more complex Home Network 
>configurations.
>
>Besides, I think that a useful informational document on the types of home 
>networks should describe several such networks - in the order of 
>simplicity - starting first from the most simple and basic and obvious and 
>only later to the most exotic.
>
>Or, while the current document does describe three such networks, there 
>does not seem to be an order from basic to complex.  The first described 
>is already "Extended" and the other two are "Aggregated" and 
>"Virtual".  The "Aggregated" is a little more complex than the "Extended" 
>in that the MR looks up its MNNs via ND on the Home Link, which is a 
>little exotic feature.  The "Virtual" model is the most exotic in that 
>there is no returning home procedure for example.
>
>So, I understand there is a form of scale on differentiation, from the 
>most basic up to the most complex, but the most basic is still called 
>"Extended" and allows MR HoA from MNP.
>
>Hence my proposal to call the basic model just "Basic Home Network" and 
>this model not to use MR Home Address from the MNP.  Alternatively, I 
>propose to introduce a 4th model, called "Basic Home Network Model", that 
>would come before the other three models, and in which MR would only 
>construct its Home Address from the prefix on the home link, that would 
>perform ND for that address, that would implement the NEMO Returning Home 
>procedure and for which address the HA performs proxy ND when MR not at 
>home, according to Mobile IPv6.
>
>What do you think Pascal?  What do the others think?
>
>Alex

Dan Shell
Government Service Unit
CISCO Systems
IP mobility/Wireless/Satellite
440 331 5663