Re: on the point of mobility & multihoming

[Brian E Carpenter <brc@zurich.ibm.com>]

Masataka,

Masataka Ohta wrote:
> 
> Kanchei Loa wrote:
> 
> >>Thus, in my proposal, details of possible mobility solution is
> >>dropped, though minimal explanation on why a bit is reserved is
> >>given.
> >>
> >>                                                      Masataka Ohta
> >
> >
> > In my opinion, you just hit an important issue of multi6 architecture, which
> > this WG is supposed to work on first.
> 
> I'm afraid you miss my points.
> 
> > If we take Dave Croker's scenario
> > (suggested by Geoff Huston) and generalize it, the problem space is the same
> > as depicted in Geoff's slide (page 3) of "An Architectural View of Multi6
> > proposals"
> 
> Dave and Geoff completely misunderstand the relationships
> between M6 and MIP.
> 
> Worse, they don't recognize that shim layer approaches are
> making the the IP layer connection oriented by let it maintain
> states of connections.

Please do not over-simplify and please do not personalize this
discussion. We all understand, I assume, that site
multihoming implies state. The exact nature of the state, and how
it is maintained and deleted, depends on the layer (or wedge)
that maintains it. 

The architectural analysis is not supposed to *define* the solution,
i.e. choose a particular model for the multihoming state. It is
supposed to analyze the options available and the arguments for and
against them. 
 
> Geoff's presentation is nothing more than rhetorical combinations
> of layers without analyzing architectural implications.

Please do not insult our colleague. This was a first presentation
and we know that a lot more work remains.

> Ignore their documents.

Please do not make such rude statements on this mailing list. 

Regards
   Brian Carpenter
   co-chair