[1899] in Commercialization & Privatization of the Internet
Re: Understanding Combits
daemon@ATHENA.MIT.EDU (Dory Leifer)
Mon Jan 6 22:05:27 1992
To: wright@saturn.hsi.com (Gary Wright)
Cc: Jordan Becker <becker@ans.net>, cook@tmn.com, com-priv@psi.com
In-Reply-To: Your message of Mon, 06 Jan 92 23:54:04 +0000.
Date: Mon, 06 Jan 92 22:03:33 -0500
From: Dory Leifer <del@terminator.cc.umich.edu>
> So if I declare my network as commercial, then I have to pay for people
> to send me email, forward news to me, anonymous ftp, finger, etc.
> whether I have asked for their traffic or not. I don't suppose I'll get a "n
et
> bill" as detailed as the somewhat analogous phone bill in order to
> verify I'm getting charged appropriately. I wouldn't want to
> audit such a document.
>
This is a really hairly issue that bears its ugly head with telco SMDS
services as well. On connectionless networks, there are no solicited
packets like on X.25 or the PSTN. If you charge for both ingress and
egress, then users can receive huge bills even if they don't send
anything (quite a tool for malicious use!). If you charge only at
the ingress, then commercial users would not be charged for packets
they received and could be overcharged for packets that they sent
that did not get delivered. Instead of counting packets maybe it is
possible to consider implementing access classes like what has been
proposed by Bellcore for SMDS. One could even imagine that the T3
between the regional and a backbone carrier could be partitioned
into at atleast two logical virtual circuits each specifiying a
particular access class (ie 2 virtual commerical DS1's and the rest
for R&E)
Can anyone comment on proposals for charging between SMDS IECs and
LECs?
Dory Leifer
The University of Michigan
