(ngtrans) NAT-PTv02 update

[Brian E Carpenter]

I suspect we agree on the immediate issue: this topic is outside
the scope of NAT-PT. 

  Brian

Pyda Srisuresh wrote:
> 
> >
> > > >- Brian Carpenter suggested that the group needs to consider how
> > > >this technique would combine with the various other transition
> > > >techniques, and API issues it raises.
> >
> > I did, and I'm still worried about that, but I think it's oustide
> > the scope of this particular document. It is more a question of
> > how applications layer code can be written in such a way that it
> > doesn't care whether it's running over IPv4 with or without traditional
> > NAPT, IPv6, either half of a dual stack, or a v4/v6 translator.
> >
> >    Brian
> >
> 
> I am not sure, I understand what you are saying.
> 
> NAT-PT does not have any issues with the API on V4-only or V6-only
> end nodes. As for dual stack nodes, once again, the API doesnt
> change because of NAT-PT enroute. However, I do agree, an
> application using the dual-stack API could (a) opt to go with the
> appropriate native protocol (V4 or v6) or, (b) opt to go with
> using a single stack if assured of NAT-PT services en-route,
> no matter who it is trying to connect to.
> 
> The API issues are outside the scope of this document.
> 
> cheers,
> suresh