[2138] in Commercialization & Privatization of the Internet
Re: Trickle Down? How About Bubble Up?
daemon@ATHENA.MIT.EDU (Steve Simmons)
Tue Jan 28 10:23:46 1992
From: scs@lokkur.dexter.mi.us (Steve Simmons)
To: bzs@world.std.com (Barry Shein)
Date: Tue, 28 Jan 92 9:19:44 EST
Cc: vaf@Valinor.Stanford.EDU, emv@msen.com, stef@nma.com, com-priv@uu.psi.com
In-Reply-To: <9201280158.AA16677@world.std.com>; from "Barry Shein" at Jan 27, 92 08:58:36 pm
>I also agree that people do not understand how labor-intensive it is
>to support many "little" customers and why that makes exactly what
>they want so difficult to deliver. 500 customers at $100/month
>represent the same revenue as 50 customers at $1000/month. But the
>overheads are dramatically different.
>
>In fact, the 500 customers usually require more per-connection
>hand-holding than each of the 50 customers do (mostly due to the
>nature of low-ball equipment and software, and its tendency to be not
>quite to spec.)
Two items here:
Barry's right about the higher per-user overhead, but one would have to
be an idiot to use that pricing (no, I'm not calling Barry an idiot).
A smart vendor would factor those overhead costs in to the price, and
come up with something like 500 customers at $250/month. Now you
have the same overall bandwidth, but far more cash to pay for those
overhead costs.
In a similar vein, one would require the end user to make certain
hardware and software investments: eg, you must buy one of a particular
set of modems, and you must use X PPP software from vendor Y. This is
no different from insisting on a Cisco router for high-speed
connections, except of course it's cheaper. Once one has done this,
the cost of overhead goes way down because you've drasticly minimized
the variety which you support. With enough customers, you even get
economies of scale in troubleshooting.
