Pleasant@Rutgers (10/22/82)
HUMAN-NETS Digest Friday, 22 Oct 1982 Volume 5 : Issue 97
Today's Topics:
Administrivia - Dedicated Discussion,
Technology - Worldnet (7 msgs)
----------------------------------------------------------------------
Date: 22 Oct 1982 0019-EDT
From: Mel <Pleasant at RUTGERS>
Subject: Administrivia
Due to the large number of responses, this entire digest
will be dedicated to the WORLDnet discussion.
-Mel
------------------------------
Date: Thursday, 14 October 1982 22:35-EDT
From: AGRE at MIT-MC
Subject: WORLDNET! (long msg)
Hu-rahhh! Here are a couple of thoughts on getting worldnet
started, one positive and one negative:
(1) First positive. Here is how you can get an ersatz worldnet up
RIGHT NOW with minimal hardware investment. Some large network
company (like ITT or TPC or N others) should offer a service whereby
you call them up on the phone and give them a phone number and a
pile of bits and they ship the bits to the vicinity of that phone
number and call it up and give it the bits with appropriate header
information. Then you write software for some existing BBS's that
bundle users' messages to users in other cities and call up the
network company and ship out the bits for various cities in big
bundles whose internal structure the network company doesn't care
about. The BBS software at the other end gets a call from the
network company, sucks in the bits (all of which are for its users),
breaks them down into messages, and distributes the messages to the
intended recipients. This can be done RIGHT NOW with trivial
hardware investment (the network company has to make a minor
investment in interfaces to TPC). The network company bills the BBS
person by SnailMail and the BBS software generates SnailMail bills
for the users. There has got to be some BBS person out there
willing to talk some network company into setting up something like
this at least experimentally.
(2) Now negative. May InterNet burn eternally. Here's why. Any
small-scale commercial part of a worldnet that gets started is going
to need a core of seriously interested, tolerant, and technically
with-it network hobbyists to keep it alive financially for its first
few years. But very many such people don't have any special reason
to put their money into such a thing because they get such good
service from the ArpaNet, and for free yet. InterNet will only make
this worse by expanding the space of government and academic
networks that can serve as siphons of seriously interested network
hobbyists. Now this might be OK if the InterNet protocols were
capable of supporting anything like a proper WorldNet. (If we're
going to have DoD socialism in WorldNets, well, let's at least do a
minimally competent job of it technically!!!) But they're not, as
Jim pointed out. They just haven't haven't solved the problem of
addressing in a large space of small networks (like the one in the
Smiths' house). Even zip codes (that is, some hierarchical
geographically oriented coding system) would be better than the
crock they ended up with, which is just routing specs no matter what
they say. I don't know what action all of this implies for all of
us who are benefiting from this creeping socialism, but it sure
sucks. I could also be wrong. - phiL
------------------------------
Date: 15 Oct 1982 1530-MDT
From: Walt <Haas at UTAH-20>
Subject: Re: Worldnet!
Worldnet is not only coming, it's here and it works. Utah-20 is
directly interfaced to Telenet via my interface package. We have
two regular interactive users in Japan who link in via KDD and
Tymnet. We have had people use Utah-20 from France via, I believe,
Transpac, but this does not go on regularly.
The cost is unfortunately still rather high, but the technology
works fine. I suspect that the cost will come down as use goes up
and engineering costs are amortized.
Cheers -- Walt
PS. Usenet is a terrible model - horribly slow and unreliable.
It's a fun toy though -- W.
------------------------------
Date: 15 October 1982 18:37-EDT
From: Robert Elton Maas <REM at MIT-MC>
Subject: WORLDNET! --> Your goals compared to PCNET's and my goals
Mostly your goals are identical with the goals of PCNET. We want
anybody to be able to join the net just by getting equipment,
looking up (in a directory) the net address of somebody else, and
just starting to send messages and wait for replies. If somebody's
equipment breaks down, it shouldn't upset all the net routing
algorithms, only messages to that node should be seriously affected.
Also all uses (play, work, amateur research, commerce, school)
should co-exist rather than each require its own network
disconnected from networks supporting the other uses.
Although high-speed operation is desirable, the net should support
low-speed operation whenever that is cheaper than high-speed
operation, as it currently is (300 baud Oregon to Florida costs only
two modems ($400 each) plus long distance charges, whereas megabaud
Oregon to Florida costs about $50,000 at each end for the
satellite-microwave equipment). A single network design should
support all speeds of equipment rather than requiring different
speeds to be on different and disjoint networks (at the least,
gateways for the major services, email, ftp, telnet) should exist
even where differing equipment requires differing low-level
protocols.
The main problem I see with USENET is that they've adopted ARPANET's
convention of English names for hosts that are assigned at random,
instead of something like PCNET's node identifiers that convey
latitude and longitude as well as phone number. This causes many
headaches with routing of messages, when a simple
geographic-proximity heuristic or even just a direct
phone-number-caller would work better for messages sent between
random points (such as from a random HUMAN-NETS reader back to the
author of a random HUMAN-NETS message). After all, why should a
message from REM at MIT-MC to keithl at tekcad have to take the
route MIT-MC -ARPA-> UCB-C70 -UUCP-> menlo70 -UUCP-> sytek -UUCP->
zehntel -UUCP-> teklabs -UUCP-> tekcrd -UUCP-> tekcad instead of
just going MIT-MC -ARPA-> UCB-C70 -ARPA-> tekcak ? Why should it be
hard to discover that a shorter route such as MIT-MC -ARPA->
Udel-Relay -UUCP-> tektronix exists? Why should one gateway be able
to dial direct but not another, just because one has secret info
such as the phone number of the recipient that the other doesn't
have? For that matter, why should the sender of the message have to
specify the whole route in the first place? (What if one of the
links in that route drops out of the worldnet?)
[At this point MIT-MC crashed for several hours and I lost the rest
of what I had typed, retaining only the above which had been saved
in a file before the crash. The rest of the message, about ten lines
about how PCNET has tried for over 5 years to build a WorldNet using
volunteer labor and still doesn't have even a 3-node network
working, how funding the creation of WorldNet is a real question,
was lost.]
------------------------------
Date: 15 Oct 82 22:35:29 EDT (Fri)
From: Velu Sinha <velu.umcp-cs@UDel-Relay>
Subject: worldnet (LONG)
Re: Guyton's message of 10/13/82
A scientist from the Tata Institute of Fundamental Research in
Bombay, INDIA has just finished writing a paper on the theme of
worldnet. This person was here (in the US) this past month and
visited uWisconsin (CSNET && MMDF) MIT, and has been in contact with
ARPA and ALOHA people. His idea is mainly centered around India, but
the scheme that he has proposed would work anywhere. His scheme is
to put 5-6 satellites in orbit around the earth. Each satellite has
TDRSS like capability (One satellite is able to talk to another).
Every neighborhood/institution which wants to participate has to
build/buy a satellite dish. Using packet technology the message will
go from the neighborhood ground station to the satellite if the
message has to travel more than 100 miles (for messages with less
than a 100 mile destination ground radio channels are used). At the
satellite the ''path'' name will be decoded and the message will
either be relayed down to earth (if the message has to go to a place
in the range of this satellite) or the message will be sent to the
NEXT (messages can only travel a distance of ''one satellite'')
satellite, there the satellite will decide weather to re-relay it or
to transmit it down to earth...
This will allow a ''world net''. The current Indianet is still in
the planning stages and hopes for money are quite high. They plan
to have an all India prototype by the end of this decade.
Ideas, comments?
- Velu
------------------------------
Date: 17 Oct 1982 2104-EDT
From: ZALESKI at RU-GREEN (Michael Zaleski)
Subject: Why not AT&T for WorldNet?
In a recent message to Human-Nets, one reader expressed a wish for a
"World-Net", to tie all sorts of computers worldwide together. In
this message, the author stated a belief that it should not be owned
by one company and specifically stated that "Ma Bell" should not be
the owner. I must honestly say that I find this attitude toward the
phone company hard to understand. Compared to ANY other phone
system in the world, the U.S. has THE best.
Phone service in many third world countries is at a level that
Americans would find totally unacceptable. Even in France, (a
country that despite its indiscriminate sales of weapons and
technology must be called civilized) the wait for a phone is
measured in months.
(Incidentally, AT&T only serves about 40 percent of the land area in
America, although that area has 80 percent of the U.S. population.
In 1976, there were about 1600 independent phone companies. From
all accounts I hear, and from my own experience, these phone
companies provide the worst service in America. Of course, that is
still tremendously better than foreign countries, primarily because
these companies work closely with AT&T.)
So, it clearly can't be the quality of "Ma Bell's" service that
bothered the aforementioned author. Perhaps AT&T is considered
suspect because the phone rates are too high? After all, don't
companies like MCI and Sprint provide cheaper long distance rates?
They do, but my experience with MCI showed me they also provide:
- Poorer quality connections.
- An extra nuisance at dialing time.
- An extra bill every month.
- A system where it is very easy to guess and use another
(random) person's account number.
- A less flexible system. (I can't use my MCI number if
I'm away from my home area, but I can bill calls to my
home phone from anywhere.)
- Insufficient savings to justify the above annoyances.
I am particularly concerned about the security aspect, since the
future of the phone company will probably have everyone using
systems like MCI - and will probably have all kinds of problems with
people using random account numbers.
Popular belief also has it that Bell's 1200 baud data transmission
standard is inferior to Vadic's, because of some sort of resonance
problem. My experience with Bell's 1200 baud is that it works fine
both locally and long distance (New Jersey to sunny California). My
experience with local calls using a Vadic has also been equally
positive.
Overall, I feel that our phone system is one of the things we should
be most happy with. If anything, the federal government should have
gotten rid of the 1600 little companies and established one gigantic
regulated phone company. Telecommunications is sufficiently
indispensible in our daily lives that a quality system is a need,
not a luxury service to be provided in a hodge-podge manner by a sea
of independent companies. It is my understanding that the breakup
of AT&T will preclude further advances in telephone service and
flexibility. I think this is the kind of loss that will result from
the breakup of AT&T, brought about in part by the anti-big
anti-phone-company feeling that some people have.
(Also, in response to a different query - probably meant humorously
- No, the phone company does not ring phones when system utilization
is low to attempt to stimulate usage.)
-- Michael Zaleski, mhtsa!mzal@UCBVAX or "Zaleski@GREEN"@Rutgers
Bell Labs, Murray Hill, NJ
------------------------------
Date: 15 Oct 1982 09:52:29-PDT
From: twc.hp-labs at UDel-Relay
Subject: Re: WORLDNET!
There is a relatively unknown Sci-Fi novella which depicts just such
a WORLDNET as you mentioned (and maybe more). As far as I know, it
is only published as part of Binary Star #5, by Dell. The name of
the story is 'True Names' by Vernor Vinge; in the book it is paired
with a not-bad novella called 'Nightflyers' by George R.R. Martin.
This book is still in print (I think, because the Oregon State
Bookstore just got more copies) so shouldn't be too hard to find.
It is a pretty exciting visualization of the capabilities possible
with such a net.
Tw Cook - HP Personal Computer Division - Corvallis, OR
twc.hplabs@Udel-Relay
------------------------------
Date: 19 Oct 1982 1442-EDT
From: Greg Skinner <Uc.Gds at MIT-EECS at MIT-MC>
Subject: Worldnet response
It's a nice idea to dream about. However, the legal hassles
alone (forget the implementation) would probably prevent such a
thing from being developed in the next ten years or so. What with
issues such as security, protection, etc., a totally distributed
network consisting of local users running PCs would be extremely
difficult to make safe, usable, etc. for all users without there
being some centralized agency who is in charge of policy.
For example, how would users be named in such a network. By
name (given names)? Too many conflicts. By address? The addresses
would most likely be akin to telephone numbers. (Hmm... could be a
prelude to Visi-phones)
Then, you want users to be able to add themselves to the
Worldnet databases without intervention. A good idea in theory, but
in case the users make errors in their applications the network must
be smart enough to resolve those mistakes, lest great mixups occur
when users try contacting and sending messages to each other. There
should also be some sort of terminal compatibility requirement (most
nets recognize a large variation of terminals, but yet in still a
finite number not equal to ALL terminals in existence).
Even still, you will always have the destructive hackers who
will try to destroy the net from wherever they are transmitting.
The net must (!!) be protected from destruction as much as possible,
or it will be almost impossible to keep it up for long periods at a
time.
Still, it's a fun idea to think about. Maybe it should be
tried on a smaller scale first (a distributed network of students
with PCs at a university, perhaps a small city or large community).
Who knows, with a PC in almost every home in a few years, maybe
it'll be possible and desirable.
--gregbo
------------------------------
End of HUMAN-NETS Digest
************************