Re: TCP Mapping

[Eric Brunner-Williams in Portland Maine <brunner@nic-naa.net>]

Hong,

How is your extension supposed to work?

Exactly what do you mean by "EPP PUSH"?

Eric

------- Forwarded Message

Hong,

You wrote:

> Most (if not all) EPP implementations today use TCP and will stay that way
> for a long time.

I'm don't doubt the correctness of your statement concerning the NeuStar
registry and client. I don't do business with NeuStar, or follow it with
any particular interest, so other than what NeuStar contributors disclose
on this list, concerning its implementation or plan of record, I wouldn't
know (and don't care).

I do know that I'm not the sole implementor who has, or is working on, an
implementation of EPP using a BEEP channel profile. I'm surprised to learn
of NeuStar's long-term de-committment from EPP over BEEP -- this month in
particular.

You wrote:
> ... we want to make PUSH work using TCP. It is
> not a requirement, it is an extension with which we can use PUSH for EPP
> over TCP.

Your initial note to this list, and Scott's careful recitation, reflect
that your desire is to modify a transport draft that has no options, and
which itself is not optional in the protocol draft. 

Just how is the NeuStar extension supposed to work?

This way?

	o client implementations communicating with the NeuStar registry
	  MUST NOT ignore the P-BIT,
and
	o client implementations communicating with another registry
          MAY ignore the P-BIT?

Or this way?

	o client implementations communicating with any registry
          MUST NOT ignore the P-BIT?


You wrote:
> 1. Can EPP PUSH be used for EPP over TCP?

Could you suggest a definition for "EPP PUSH" please? I seem to have missed
it.

Eric

------- End of Forwarded Message