From: "Perry E. Metzger" <perry@snark.imsi.com>
Date: Fri, 6 May 94 15:04:38 PDT
To: lile@netcom.com (Lile Elam)
Subject: Re: MBone cypherpunks session...
In-Reply-To: <199405062130.OAA09701@netcom.com>
Message-ID: <9405062201.AA02903@snark.imsi.com>
MIME-Version: 1.0
Content-Type: text/plain



Lile Elam says:
> >You completely misunderstand. The MBONE is a very scarce resource at
> >the moment. Setting up teleconferencing sessions that aren't, say,
> >multicasting a conference (like, say, a Cypherpunks meeting) but are
> >just being used to let people do what they do on mailing lists at
> >500,000 times the cost in network bandwidth isn't social.
> 
> I think you are not quite clear on how the MBone works. It uses the
> bases of broadcasting a session once which can be received by the
> many. Only a few of those many will get involved with the session
> in the since of retransmitting...

Could you post that in English? "the since of retransmitting" sounds
especially interesting.

I'm quite clear on how the MBone works.

Anyway, I'll summarise my opinion on this subject.

If no one is using it it takes up no bandwidth. Naturally, if no one
is using it having a session doesn't make much sense. If very few
people are using it phone calls are cheaper and better on the ears, so
having the session makes no sense. If large numbers are using it the
technology makes considerable sense provided that what one is
multicasting is something like an IETF or Cypherpunks meeting, but if
its just a few random folk chatting, the session isn't justified --
one is spending a huge amount of networkd bandwidth on something that
Netnews or IRC is far better suited for.

Perry