Re: [mmox] reverse http RFC

On Sun, Mar 8, 2009 at 7:25 AM, Lawson English <lenglish5@cox.net> wrote:

Hey grats to Zero Linden (Mark Lentczner) and Donovan Preston  for getting
> this out. Will make virtual worlds communication much more robust in the
> long run, I think:
>
>
> http://www.ietf.org/internet-drafts/draft-lentczner-rhttp-00.txt
>
>
>
I agree.  This would be a very good thing for the downlink path of
client-server systems, which includes those where the "server" is a P2P peer
--- it provides great symmetry, and avoids the proliferation of random hacks
to overcome the problems associated with incoming connections to clients.
Reverse HTTP is not a hack, because everything that applied previously to
(forward) HTTP then becomes available in the reverse direction, so it's
beautifully elegant and powerful.  That's good progress.

OGP is currently tied to the specific technology of long poll (in 3
different paragraphs of the current draft), which is inappropriately
proscriptive.  This is one area that needs to be widenened for future draft
iterations, so that techniques like Reverse HTTP can be employed, as well as
streamed response and other more efficient mechanisms for state transfer.

Morgaine.

On Sun, Mar 8, 2009 at 7:25 AM, Lawson English <lenglish5@cox.net> wrote:

> Hey grats to Zero Linden (Mark Lentczner) and Donovan Preston  for getting
> this out. Will make virtual worlds communication much more robust in the
> long run, I think:
>
>
> http://www.ietf.org/internet-drafts/draft-lentczner-rhttp-00.txt
>
>
> Lawson
> _______________________________________________
> mmox mailing list
> mmox@ietf.org
> https://www.ietf.org/mailman/listinfo/mmox
>