Mohsen;
> Masataka> WAP and IP over NAT are equally bad.
>
> We have two sets of problems and layering helps here.
>
> At layer 3, we need to make things end-to-end.
>
> At layer 7, the WAP approach is simply the wrong approach.
>
I'm operating on all the layers.
> We need competition in the efficient appliction protocols
> space.
As TCP/UDP port space is, so far without NAT, large enough, we don't
have any real problems.
> As you pointed out more than a month ago:
>
> Masataka> To make the competition fair, the important questions are:
>
> Masataka> Is it fair if providers using iMODE or WAP are advertised
> Masataka> to be ISPs?
>
> Masataka> Is it fair if providers using NAT are advertised to be ISPs?
>
> Masataka> My answer to both questions is
>
> Masataka> No, while they may be Internet Service Access Providers and
> Masataka> NAT users may be IP Service Providers, they don't provide
> Masataka> Internet service and are no ISPs.
>
> Which in my thinking is equivalent of saying that WAP is at best an
> Internet gateway model. Which is consistent to my position in The WAP
> Trap paper ...
Thank you.
Masataka Ohta
