At 8:00 PM -0700 8/10/00, ned(_dot_)freed(_at_)innosoft(_dot_)com wrote:
> > Well, there is a big difference between WAP's breaking the e2e model
> > and i-mode. WAP does an application gateway and uses no Internet
> > protocols. At least, i-mode is using IP, TCP, HTTP, etc.
> Who cares what protocol a device runs as long as it delivers the
> application that satisfies its intended users? Most subscribers
> couldn't care less if i-mode used CLNP and TP4 instead of IP and TCP.
I agree that most users don't know or care what transport they use. However,
the choice of transport and network layer protocols has real consequences, and
some of those consequences are things that subscribers do care about.
Your own example illustrates this quite nicely. As it happens I've dealt with
operating Internet application protocols over CLNP/CONS and TP0/TP4, and the
transport-level gateways this sort of practice engenders are all too often a
trouble spot for the users of the service. And these were small scale
deployments. Large scale stuff would be even more problematic.
For the record, the protocol design problems of WAP have little to do
with anything that you or Ohta are talking about. A quick look at
the WAP protocols, shows this. The WAP designers have made protocol
design decisions that severely limit their ability to succeed.
My previous note was very much to the point. WAP most closely
resembles Videotext. In some sense, one could conclude that the
designers of WAP were assuming that it would fail. Hence, they
developed an architecture that can not exploit its success or exploit
new technology. Many of the errors they have made are the same ones,
we saw the PTTs make with Videotext. In fact, they are so similar
that without knowing the age of the designers I can't tell if it was
accidental or on purpose. If WAP is widely successful and/or as new
technology appears, a wholesale replacement of WAP will be required
that can be exploited by others to steal their customers.
The WAP architecture is fundamentally flawed ways in significant ways
that i-mode is not. I-mode is in a much better position to
capitalize on its success. And it has little if anything to do with
the knee jerks about NATs or "CLNP and TP4 instead of IP and TCP"
none of which had anything to do with Videotext. I suggest you read
the WAP specs and see for yourselves.