This is one of the most coherent summaries of the issue of paying for the
Internet I've found. Maybe there are others out there? Ciao, Lyo.
-------------------------------------------------------------------
---------------
Date: Thu, 29 Jul 1993 08:27:53 CDT
Reply-To: Public-Access Computer Systems Forum <PACS-L@UHUPVM1.BITNET>
Sender: Public-Access Computer Systems Forum <PACS-L@UHUPVM1.BITNET>
From: Public-Access Computer Systems Forum <LIBPACS@UHUPVM1.BITNET>
Subject: War Over the Future of the Internet
To: Multiple recipients of list PACS-L <PACS-L@UHUPVM1.BITNET>
2 Messages, 107 Lines
*-----
From: Jack Kessler <kessler@well.sf.ca.us>
Subject: Re: War Over the Future of the Internet
Jennifer Heise,
Your posting asks for some reassurance about Internet-use pricing by
the BabyBells. There is no reason why the BabyBells would abandon their
established mixed-approach pricing for providing Internet service:
per-minute charges for some users, flat rate for others. Internet use
is not different from voice telephone use in this respect, to a
BabyBell. I don't think the question is one of which type of fee of
these two will be charged, as both and others and various ingenious
combinations undoubtedly will rear their heads as they have with
telephone service.
The question is rather how much the charges will come to, and here I
can offer some reassurance. The unit charges, whether billed at a flat
rate or by the minute, will have to be low. No one understands the
advantages of economies of scale better than do the BabyBells: the
cheaper the call, the more calls will be made. They have built their
industry on it. I would expect Internet use charges from the BabyBells
similar to their charges now for telephone use: some free services, low
unit charges for normal use, flat rate charges for special high-volume
uses (perhaps university libraries, perhaps not -- how's your
negotiating power?). People screamed when they charged for voice
telephone services too, but it's worked out fairly well.
Jack Kessler
kessler@well.sf.ca.us
*-----
From: Ken Laws <LAWS@ai.sri.com>
Subject: Re: War Over the Future of the Internet
As to the net backbone being "sold off," there is really
nothing to sell. Neither NSF nor any other agency (other
than the military) owns backbone hardware. NSF rents
channel capacity from ANS, a subsidiary of Merit.
The privatization debate has many aspects. One is whether
NSF will continue to pay ANS for backbone services or will
pay regional networks more directly so that they can
purchase services from ANS or whomever they wish. There
are plenty of bandwidth providers out there who are eager
for the business, and who claim that NSF is competing with
them unfairly.
Another aspect is whether the government-supported backbone
will restrict traffic to non-commercial use (as it does now).
We need commercial networks, and they are being provided by
many companies. These commercial networks have been linking
together in various ways, including a consortium called CIX,
and have very nearly made the NSFNet backbone irrelevant
-- for commercial users. Academics still depend on NSFNet,
Bitnet, and Usenet. Many of them are fearful of commercial
use -- perhaps rightly. Others are eager for access to
commercial services. NSF seems to be "looking the other
way" at the moment, allowing commercial experiments to
flourish on the Internet. Ultimately, however, NSF is not
permitted to pay for support of such services unless they
somehow serve the research and education communities.
(Definition of "R&E" is still an open question. It generally
has not included libraries, but NSF and Congress clearly
want to support libraries under the various NREN plans.
R&E also excludes home use -- even to access the Library
of Congress -- and hobby use such as NetNews, but NSF
and perhaps Congress would like to support these as well.
Commercial providers claim that such uses are within their
domain, and they don't appreciate such government support.)
Another issue is whether charges should be metered or
flat-rate, universal or by customer type. This is already
in the hands of market forces, for the most part. New or
naive users generally prefer flat rates; sophisticated
customers want to pay only for services used. Service
providers prefer flat-rate billing, but have a basic problem:
flat rates are based on "pipe capacity" and on a statistical
model of bandwidth usage. That works well for naive
customers, who typically use only a fraction of the
capacity that they have available. Sophisticated
customers learn to fill their pipes to near 100% use,
thus taking advantage of the statistical pricing structure.
However this works out, it is in the hands of individual
service providers. There is no natural monopoly here
(other than the wires that enter your house), so competition
will soon give consumers what they want.
Another commercialization issue is whether you will have
to pay for services at the other end of the wire. Internet,
Bitnet, and Usenet have no mechanisms for billing or
funds transfer, so there has been no way for commercial
services to get started. This will be changing soon,
and "virtual stores" of all sorts will be established.
Whether they will displace the free services no available
on the net remains to be seen. My own belief is that
free and for-pay services will co-exist forever. I have
no doubt, though, that the better services will be drawn
into the commercial world. This is a threat to those
who want the best-possible free services, but is welcome
to those who want the best-possible services (whatever
the cost) and to those who want to earn a living from
helping people via the net.
-- Ken Laws
This archive was generated by hypermail 2b29 : 03/09/00 PST