Re: [mif] definition of a MIF host

[Scott Brim <Scott.Brim at gmail.com>]

Excerpts from Nicolas Montavont on Thu, Jul 30, 2009 01:58:35PM +0200:
> In the same sense, I think that the bullet "Communications using
> these IP addresses may happen simultaneous and independently."
> should not be in the definition, but again, in some explanations of
> scenarios.

Everything you say is fine with me except this.  This sentence does
not refer to actual simultaneous/independent use, which could go in
scenarios.  Rather it says "may happen", i.e. the endpoint has
arranged to be _capable_ of using multiple addresses simultaneously
and independently -- through enabling and configuring interfaces.
When that is done, the MIF problem arises.  If addresses are only used
sequentially, there need not be a MIF problem.  So this sentence
describes an essential part of the problem, and I think it should be
retained.

Scott