Computer Networks and the resistance ?
Steven D. Majewski
( in particular, one from one of the folks at DEMOS )
referring to using the computer network for organising the
resistance.
I have seen very little in the press here ( Wash.Post, NYTimes )
on this : one article in the WashPost-Style section that mentioned
computer networks, but was mostly about the flurry of FAX's between
the Yeltsin gov. & the outside world.
Can someone fill us in on the story ?
How widespread is the network in the USSR ?
I gather that DEMOS is connected via phone line to Finland.
Are the Universities in USSR connected ?
( And are any of my Gromicko cousins out there ? )
======== "If you have a hammer, find a nail" - George Bush,'91 =========
Steven D. Majewski University of Virginia Physiology Dept.
sd...@Virginia.EDU Box 449 Health Sciences Center
Voice: (804)-982-0831 1600 Jefferson Park Avenue
FAX: (804)-982-1616 Charlottesville, VA 22908
Laurence I. Press
>
>How widespread is the network in the USSR ?
>I gather that DEMOS is connected via phone line to Finland.
>
become an international network :-)). They have about 14 full-time
backbone machines and about 400 user organizations ranging from
universities to commodity exchanges and news services. They are
an important part of the Soviet communication infrastructure, in
part because the alternatives like the telephone system are not
very good. And, as you gather, they connect to Finland about once
per hour I believe -- exchanging news and mail. They also connect to
Norway and Hungary, but less frequently. During the coup, they
were an important source of information and morale, at great personal
risk.
Larry
vlad
Thank you, Vadim!
Vlad
Disclaimer: my employer may disagree
Nick Szabo
>Demos runs a UUCP network that spans the Soviet Union. (Soon it may
>become an international network :-)). They have about 14 full-time
>backbone machines and about 400 user organizations ranging from
>universities to commodity exchanges and news services. They are
>an important part of the Soviet communication infrastructure, in
>part because the alternatives like the telephone system are not
>very good.
>...During the coup, they
>were an important source of information and morale, at great personal
>risk.
I'd like to express my utmost admiration to those operating this network
during the coup. It is once again demonstrated that democracy can only
grow with the free, decentralized flow of information, and can die
without it. It is no coincidence that the TV and radio stations, and the
telephone network, were strategic areas of control and the subject of
raids throughout the coup.
UUCP often operates through the telephone system, which you call an
"alternative". What physical connections is DEMOS using, and why cannot
they be shut down as the telephone system was shut down in many areas
during the coup?
--
sz...@techbook.COM ...!{tektronix!nosun,uunet}techbook!szabo
Public Access UNIX at (503) 644-8135 (1200/2400) Voice: +1 503 646-8257
Public Access User --- Not affiliated with TECHbooks
Douglas W. Jones,201H MLH,3193350740,3193382879
by sz...@techbook.com (Nick Szabo):
>
> UUCP often operates through the telephone system, which you call an
> "alternative". What physical connections is DEMOS using, and why cannot
> they be shut down as the telephone system was shut down in many areas
> during the coup?
I speak with no specific knowledge of how DEMOS is connected, but rather,
with a general knowledge of how computer networks are built. The key point
is that computer networks can use any of a huge number of communications
technologies, and many networks use a fairly random mix of technologies.
Typically, the first step in building a computer network is to use the
dial telephone system. Low traffic links with low reliability requirements
may continue to use the telephone system forever, but if traffic or
reliability are problems, managers of computer networks have many
alternatives:
1) Leased telephone lines -- these bypass the telephone exchanges and
provide permanent connections. The most common way to shut down a
telephone system is by turning off the exchanges; this doesn't interrupt
leased lines. Cutting cables, or shutting off power to the amplifiers
and repeaters does interrupt leased lines, but the military also uses
large numbers of leased lines, so they would be unlikely to do this.
Short of such an attack, the only way to interrupt leased line service
is to know precisely which wires to cut out of the millions of wires
that make up a telephone system. This takes cooperation from large
numbers of telephone company employees, and it might take a day or two
to trace through the records to find the right wires.
I'm fairly sure that most of the links in the RELCOM backbone are
leased lines, probably voice grade lines with 9600 baud modems.
2) Private wires outside the telephone system. For short haul traffic,
many networks have found that it is easier to string their own wire
than to rent wire from the phone company. To interrupt service on
these wires, you have to know that they exist and then find them.
Privately owned wire is common on campuses around the world, and
it is common in many third-world cities where the central telephone
system is unreliable. I wouldn't be surprised to find RELCOM hosts
connected by similar private wire within some of the larger Russian
cities.
3) Microwave links. Small point-to-point microwave antennas are getting
less and less expensive as the decades pass. The University of Iowa
has a microwave link to a plant 20 miles away where we teach some remote
courses by video link. One microwave link can easily carry a handfull
of TV channels and a one megabaud network link. Such links are hard to
shut down with anything short of a direct attack on the transcievers at
each end.
I wouldn't be surprised if the physics labs in Moscow and Dubna already
had such links, and I wouldn't be surprised if RELCOM is piggybacking
on them.
Finally, if all real-time links between two network segments are down,
it's not too hard to copy the disk queue of outgoing messages onto a
floppy disk and physically carry (or smuggle) the disk to the other
end. If two networks usually place a phone call once daily to exchange
messages, and if the machines are 6 hours apart by car, the users won't
even notice much of a delay if someone drives back and forth between the
networks with a floppy disk to carry the traffic.
Doug Jones
jo...@cs.uiowa.edu