[2777] in Commercialization & Privatization of the Internet
The future of the Internet, a few further comments
daemon@ATHENA.MIT.EDU (drw@BOURBAKI.MIT.EDU)
Mon Jun 1 16:04:20 1992
Date: Mon, 1 Jun 92 16:02:25 EDT
From: drw@BOURBAKI.MIT.EDU
To: com-priv@psi.com
In-Reply-To: <9205311949.AA12008@aland.bbn.com> "craig@aland.bbn.com"
From: Craig Partridge <craig@aland.bbn.com>
> 3. The Internet will be sold as a service for a fee, rather than being
> delivered by a club-like consortium effort between a small number of
> widely scattered users.
Access already is sold as a service for a fee. What's the point here?
The point is that people don't *think* about the Internet as a service
that they are buying from a vendor. The differences aren't enormous,
but they are there. For instance, people are much more paranoid about
sending sensitive data over the Internet than they are about sending
it over an ordinary phone connection. This isn't because the telco is
intrinsiclly more secure, but because they believe that the telco is a
"repuitable vendor" with relatively little incentive to spy on
customers.
Another aspect that shows up is that some of the regionals still
restrict their connections to "research and education" customers.
Even on those, commercial customers probably outnumber educational
customers, but the commercial customers lie and say that their usage
is for "research purposes". Once you look upon the regional as a
vendor, you say "I want them to have as many customers as possible, to
give the vendor economies of scale, so that he can charge me as low a
price as possible."
> B. Serious thought ought to be given to the development, interaction,
> and economics of commercial Internet providers.
Ask not, lest ye shall receive. More seriously, I think this is one
of the key questions. A concern is public utility commissions don't
understand the beast and probably could kill it through any one of a
number of bad decisions. So becoming a public utility is probably *not*
a virtue. (Think of this idea another way -- what if the US Postal Commission
was suddenly given authority over the rates of the various package
carriers such as UPS, Fed Ex, etc. -- would this be a good thing?
Would it improve service?)
Better to ask so that what we receive isn't decided by default.
Seriously, in a market with many different classes of customers with
varying needs, different vendors with different charging schemes
interact in *interesting* ways. (Consider what happened once
competition was introduced into long-distange telephone service.) We
should have some idea what those interactions are before we get hit
with the consequences. Perhaps we can reduce the amount of trouble
that will be endured as the market stabilizes.
> C. Should ubiquitous service be a goal? When the merchant sends a
> credit card number to be verified, why not do it by Internet?
I believe they already use private TCP/IP and/or X.25 networks. I know
that several of the major credit card networks were built by prominent
internetworking vendors. As for whether the credit card companies prefer
a private network or a public one, that's presumably their choice, yes?
Just as one can have virtual private networks for your phone service.
I was looking on credit card authorizations as an example use of the
Internet by a very small customer. As far as I know, merchants with
electronic authorization still make a standard phone call to some sort
of point-of-presence of the credit card processing organization. It
would be an ideal situation for Internet usage -- just send out one
UDP packet and wait for the answering packet! Unfortunately, you
can't order Internet jacks yet from the phone company.
Dale
Dale Worley Dept. of Math., MIT drw@math.mit.edu
--
The best way to demand something unreasonable is to call it a "freedom".
