From: “Perry E. Metzger” <perry@imsi.com>
To: rishab@dxm.ernet.in
Message Hash: 04cce4d780efc396c419ae6179a94f6977c9870c08b341515106759a8252b540
Message ID: <9409211913.AA12090@snark.imsi.com>
Reply To: <gate.0e91sc1w165w@dxm.ernet.in>
UTC Datetime: 1994-09-21 19:15:16 UTC
Raw Date: Wed, 21 Sep 94 12:15:16 PDT
From: "Perry E. Metzger" <perry@imsi.com>
Date: Wed, 21 Sep 94 12:15:16 PDT
To: rishab@dxm.ernet.in
Subject: Re: Reserved DC-Net IP addresses - NOT!
In-Reply-To: <gate.0e91sc1w165w@dxm.ernet.in>
Message-ID: <9409211913.AA12090@snark.imsi.com>
MIME-Version: 1.0
Content-Type: text/plain
rishab@dxm.ernet.in says:
> hughes@ah.com (Eric Hughes):
> > Speaking of long-term integration on the internet, might it not be a
> > good idea to get some IP address range assigned for dc-net use?
>
> I don't agree. Apart from the IP address crunch that already poses a
> problem for the IETF, I don't see why DC-Nets should be limited to
> specially allocated addresses. DC-Nets should be implemented a layer
> over protocols such as IP, or as extensions to them. It should be
> possible for _anyone_, regardless of IP address (officially
> allocated DC-Net number) to connect with _any_ others on a DCN, with
> some software or other interfacing.
1) Were DC nets to be considered an important service, Rishab's
comment would be correct -- one can put things on top of other
protocols much more easily.
2) To be interesting, DC nets require physically secure transmission
between neighbors. Given that the internet is not physically secure,
I'm not sure DC nets are going to run on the bare net
any time soon. Presumably one time pads could be used to
implement DC nets in order to maintain "unconditional" security or
public keys could be used to maintain "good" security.
Perry
Return to September 1994
Return to “rishab@dxm.ernet.in”