[11096] in Commercialization & Privatization of the Internet
just testing whether the mailing list still works
daemon@ATHENA.MIT.EDU (Hans-Werner Braun)
Mon Mar 21 23:42:58 1994
From: hwb@upeksa.sdsc.edu (Hans-Werner Braun)
To: com-priv@psi.com
Date: Mon, 21 Mar 94 8:36:40 PST
May be this had been said before, but I think many people today seem to
be barking up the wrong tree. Five or ten years ago the packet
switching on the Internet, and improvements to same, was a really hot
thing, and driving varieties of groups to work in that area. Sometimes
today hearing quite some shrillness about the underlying fabric often
makes me think twice about whether the person has soe kind of personal
agenda in mind. Packet switching just isn't that hot any more, just
like the power and water and postal mail distribution. Its requires
lots of qualified work, no doubt, and someone has to do it, but a
utility isn't that hot to play with. A national information
infrastructure will eventually just have to assume that there is an
underlying fabric, and really focus on information provisioning, rather
than phone line (or even ATM circuit) provisioning. Just look at the
"and the company that will bring it to you is ATT" commercials; they
don't say one thing about how A is connected to B and at what speed and
via which NAP, just the perception and provision of an information and
application-data access system to end users That also means information
"for the people," not T1 and T3 and OC3 homeruns of an NSP to a site.
Look, ask yourself, assuming you believe that the Internet (information
access) is really important, if not critical, to you and your family,
how much are you willing to pay for it, assuming a really ubiquitous
and reliable/predictable network (not only to companies, agencies, and
universities, but the Smith and Doe families as well)? I asked around
a little, and a common answer seems to be "about as much as for cable
TV." The top figure I heard (and those were from people who really have
children and perceive a benefit for them) was $50/month. Now, a flat
rate of $35 per month and a $8.50 access charge per hour just won't cut
it. I bought my kids a modem a few weeks ago and showed them a little
about the Internet. They immediately got hooked on email and Mosaic.
$8.50/hr plus $35/month? No way that will work! If measured service,
say, 3 hrs/day times 30 days makes 90 hours a month. Lets say 100.
$50-$35 = $15 for the difference. $15/100 makes 15 cents per hour (a
quarter cent per minute) if you want measured service. No way, you say,
the phone line access costs are too high for that to be feasable. Well,
tough. What are the technology choices? Calling area-local terminal
servers (like Prodigy)? Cable TV plants? Perhaps even satellite
access? Flat rate ISDN?
What's my point? My point is that the underlying switching substrate
of the data component of the United States national telecommunications
infrastructure is about as relevant to an end user as the telephony
component. It is needed, and to those people it is irrelevant what it
looks like internally, as long as predictable services are being
delivered in an affordable fashion. Just like dial tone on phones.
The Real Trick, however, in the future will be the information
provisioning itself. Who cares about telnet, and ftp? Where are data
base services? Discussion groups? MUD games? More entertainment? Adult
stuff? May be it can even carry something as rare and precious as
research and education related stuff? Just like in a bookstore, where
R&E is there, but a detail compared to all the entertainment based
literature. I suspect that once the market has been developed, the
services will focus much more on the applications, with the
connectivity just being a detail to be provided from a utility company.
Some have learned the lesson already and provide services besides just
T1 access lines. For example, my favorite FTP server is FTP.UU.NET,
simply for the fact that it is very well maintained and almost always
available (rarely refuses connections). But running a service like
that costs real effort.
Now, what about a deal of a flat $40 service per month for personal Internet
access from wherever I am in the country (just like I can get cheap
Prodigy access from many places, for example) and then for $20/month more
for a total of $60/month bundle in access for a decent and always up
to date multimedia based encyclopedia (makes it easier to argue to
one's wife why paying this is really not for own benefit, but the
kid's; just like when buying a CD drive for your PC at home), some
multiuser entertainment host with a number of well maintained
multimedia based games (advanced MUDs), and perhaps something
financial or so (may be stock quotes and some host that runs this 10
megabyte application to help you balance your checkbook; with perhaps
some direct access to TRW or so). So, with an application-bundled
service, may be you would shell out a little more even than $50 per
month?
