Re: EFF, Educom & HR1757

[gnu () toad com <gnu () toad com>]

Pres Smith said, to nren-discuss:
> it authorizes giving away of a taxpayer investment of billions of
> dollars over a 20+year period to industry with only a trickle of
> of support for general education, k12 and public access.

Which particular taxpayer investment of billions of dollars is being
given away, and to whom?  The Internet technology (TCP/IP, all the
higher level protocols, routing technology, more than a thousand
RFC's) was all given away as it was done.  Public money was spent, and
the public reaped the results.  I didn't hear any screaming when DARPA
funded UC Berkeley to write a good implementation of TCP/IP and give
it away so that universities and industry could adopt it and produce
highly functional workstations and minicomputers.

If you are talking about the physical infrastructure, very little of
that was bought by NSF.  All the phone lines are leased, not owned.
The entire 56K ARPAnet has been dismantled and thrown away.  The
entire T1 NSFnet has been dismantled and thrown away.  The only thing
that remains is the T3 NSFnet, and that only costs $10M/year in
inflated ANS prices.  Total subsidy there is <<$100M.  And I'm not
sure that failing to fund it in a subsequent year counts as "giving it
away".  When I lease a T1 line for a few years, and then stop funding
it because I don't need it any more, am I "giving it away" to the
phone company?

>                                       ... with only a trickle of
> of support for general education, k12 and public access.

The existing NSFnet certainly doesn't provide more than a trickle for
general education or k12, and none at all for public access.  All the
public access to the Internet that has ever existed, was provided by
commercial or nonprofit companies -- like Portal, Netcom, the Well,
the World, Panix (for individual user access), and Alternet, PSI,
Sprint, and others for publicly available TCP/IP hookups to private
machines (individual and corporate).  NSF has been `standing in the
schoolhouse door' -- making it harder to integrate public access with
their `research and education' network -- not the other way around.

I'm cofounder of a nonprofit community cooperative providing Internet
services in the SF Bay Area -- the Little Garden network.  Our local
regional, Barrnet, blanched when we told them we were actively hooking
peoples' home machines up to the Internet at $100/month; they can't do
it for three times the price.  But we get nothing but stonewalling
from NSF.  Like, we tried to hook up to the NSF network to pass our
noncommercial traffic, but it turns out they have abrogated
responsibility for the NSFnet routers to the local fiefdoms, which
won't let anyone else connect up.  Whenever there's a policy question,
there's a whole hierarchy of organizations to point fingers at -- NSF,
MERIT, ANS, the regionals -- and nobody's responsible for anything.
We publicly asked Steve Wolff three years ago what the real rules for
what traffic can move over NSFnet are (e.g. can I send email via
NSFnet to someone at another commercial company?  How can I even tell
whether my message traverses NSFnet?).  His response was to point me
to their mealymouthed `policy', which doesn't answer any of these
questions.  Meanwhile, Wolff has personally excercised censorship over
NSFnet sites that were completely legal according to the published
policy, threatening to cut off both MIT and the entire country of
Finland, from the NSFnet -- because he thought reactionary Congressmen
might not want to see a particular kind of sexual free speech
happening over `their' network.  If this is NSF's idea of professional
service, academic freedom, and public access, we get a better deal from
Prodigy.  At least the censorship there is formal, published policy,
as is the price and terms.

Mr. Smith, what, exactly, is your complaint with the process of
removing the government from the direct purchase of network services
for education and research?

        John Gilmore