From: "E. ALLEN SMITH" <EALLENSMITH@ocelot.Rutgers.EDU>
Date: Wed, 8 May 1996 15:16:28 +0800
To: fair@clock.org
Subject: Re: Is the network layer geodesic?
Message-ID: <01I4FP905KBM8Y59D8@mbcl.rutgers.edu>
MIME-Version: 1.0
Content-Type: text/plain


From:	IN%"fair@clock.org"  "Erik E. Fair"  7-MAY-1996 02:06:23.16

>The Internet is amorphous. It ain't a star, exactly, but it still not too
>far from that. However, to get away from this situation into the rich and
>more fully amorphous connectivity we used to take for granted in the UUCP
>network, we're going to have to see a lot more cooperation on the part of
>the small ISPs in agreeing to talk *directly* to each other to exchange
>traffic, and more small exchange points, instead of the small number of
>large ones.

	One thing that may make a difference in this is that the major
providers are getting overloaded. We're currently having problems here (in
terms of slowdowns, failures to connect, et al) because MCI's trunk lines
are majorly overloaded, and that's who _all_ the local ISPs go through. I am
currently attempting to persuade them to at least get contacts with another
ISP (and to serve as a router) in order to speed up local stuff.
	-Allen