doug@isishq.FIDONET.ORG (Doug Thompson) (11/10/88)
BS>From: bzs@encore.com (Barry Shein)
BS>I agree with Vincent's apprehensions.
BS>
BS>What would probably be revolutionary would be an inexpensive home
BS>computer who's forte is setting up home-brew usenet/e-mail links
BS>between households. Something like an Atari/ST or PC/Klone with a
BS>2400b modem and a 100MB disk with a "just pick a site name and a
BS>neighbor" software set-up. It doesn't have to be fast. It could
BS>probably be done for around $2000 list or less.
BS>I suppose FIDOnet was an
BS>attempt at that, perhaps someone from that culture could comment on
BS>its current status?
Whew! Yeah, I could comment on it. It exists. You want a uucp mailer in
your home? You can do it for well under $2000, and you can do it today,
and it's no more complex to learn than a very simple word processor
(good for novices etc.)
Fidonet itself may end up as a footnote in history, but out of fidonet
has come the software to do this on a very plain-jane PC clone.
I start salivating when I think of what a Sun or a NeXT or a turbo 386
could do with this idea. Hardware capability is going up and prices are
coming down. Things can only get better.
This message is entered on a PC, a PC called "isishq" in Usenet and
"221/162" in Fidonet, that serves as a hub for an international network of
student newspapers (low budget agencies with no computer savvy for the
most part). Transcontinental phone costs being what they are, we went
with trailblazers instead of 2400. If you think uucp is fast on a TB,
you should see fidonet mail protocols! It puts FAST into a CONNECT FAST!
<grin> Basically, 2400 is too slow for heavy-duty long-distance bulk
transfers, and uucp -g is too inefficient.
And yes, this is highly revolutionary I think.
Fidonet is suffering from the revolution, and provides an incredible
case study . . . but moves are afoot to convert addressing to the
internet standard and develop gateways in all directions.
In Fidonet (somewhat like usenet) ownership of resources is dispersed,
and net-wide decisions require a consensus among all those involved.
Such a consensus is hard to achieve. What tends to happen is that net
standards emerge as a result of someone writing software which everyone
else starts using. Along the way there is an awful lot of argument and
disputation. Like any large group of people, unanimity does not exist
and a number of very different points of view compete with each other.
Fidonet now involves 4,000 listed nodes, almot all MS-DOS systems, and
several times that many "points" (unlisted, private nodes). Many are
public access BBS systems with socres or hundreds of users. The
nodelist doubles in size every 12 months. A few dozen of these are also
uucp sites and run uucp protocol g under MS-DOS. With my set-up the
software that answers the phone detects whether it is a uucp poll, a
fidonet poll, or a human calling, and loads the appropriate software to
conduct the session. Feel free to call (519-747-1965) and browse,
you'll get some sense of what is happening.
Historically, it took a real "hacker" to get a fidonet system up and
running and required a lot of skill to link it all to uucp. But now
there are new programs which enable novices to conduct a meaningful mail
and news existence.
BS> Centralized service machines could then be built around this.
That is exactly what we are doing. Some listed nodes are servicing
hundreds of point systems. The structure is in the very early stages of
development, but a public-owned, cheap e-mail system is emerging.
BS>The important distinction from services like
BS>CompuServe is that a significant amount of the computing would go on
BS>in the households.
Right. The economies of using fully automated (uucp or FTSC) mail
connections instead of manual log-on are phenomenal! I've been costing
this out just today, and you get savings ranging from modest in some
circumstances to factors of several hundred in others. One of the main
savings is that the "host" can serve 300 automated "mailer" points with
the same hardware and system load required to serve 20 people logging on
manually. Overseas, the cost savings in phone bills are very
substantial.
The marginal cost of sending the average message from Canada to
Australia is about 4 cents. Within North America it is a fraction of a
cent (in a well organized system).
BS>
BS>What do other people think would be revolutionary in a personal
BS>computer?
BS>
The auto became revolutionary with Henry Ford, not because his car was
anything so amazing, but because its price tag was. When they become a
mass phenonemon, they will have revolutionary impact.
It's already happening now with computers.
You want a uucp site on your home computer? Gimme a call, I'll send you
the software :-). Don't expect the sophistication of Framework III, this
is home-grown amatuer code in large part (especially the stuff I wrote)
but like the Model T, it works, and heck, the price is right!
But seriously, I've written a series of programs which will install
themselves on your home computer and set you up as a "leaf node" or point
off my system. This gives you mail and news capabilities with a few
limitations. The software is beta test and draws heavily on a lot of
existing shareware fidonet and DOSuucp software. But it does work, it is
very simple and convenient. There is a lot of room for improvement, and
in time it will be improved, as the Model T was improved. Of course it
can be configured to connect to *any* FTSC capable mail host and I could
bend it to do uucp directly if I put my mind to it. Right now the uucp
part is handled at the host, and is fairly complex.
If I spent more time cobbling code and less time indulging my addiction
in net-news it would probably be better and I'd probably be richer. But
then if it weren't for the net-news addiction I'd have never thought of
this :-).
I wrote this software for my *mother* so I could exchange e-mail with
her. She gets intimidated by electric typewriters and manual
transmissions, but it's simple enough for her to use. It would not have
been possible without the code generated by hundreds of FidoNet
developers, at least not for another five years.
=Doug
--
Doug Thompson - via FidoNet node 1:221/162
UUCP: ...!watmath!isishq!doug
Internet: doug@isishq.FIDONET.ORGbzs@PINOCCHIO.BERKELEY.EDU (Barry Shein) (11/11/88)
From Doug Thompson (replying to my unfortunate initials) > BS> > BS>What do other people think would be revolutionary in a personal > BS>computer? > BS> > >The auto became revolutionary with Henry Ford, not because his car was >anything so amazing, but because its price tag was. When they become a >mass phenonemon, they will have revolutionary impact. > >It's already happening now with computers. Some say that the critical thing in making the auto revolutionary was committing to building roads for the autos to drive on. W/o smooth roads horse-drawn buggies seem a win over cars. Now many are saying networks are to computers what roads are to cars (including the need to invest in infrastructure at a societal scale.) In the end this "computer" thing may be renamed "communicator" by many. Well, the car pioneers called 'em "horseless carraiges", like the word "computer", very literal minded. -Barry Shein, ||Encore||
daveb@gonzo.UUCP (Dave Brower) (11/13/88)
In <8811102118.AA12763@pinocchio.UUCP> bzs@PINOCCHIO.BERKELEY.EDU (Barry Shein) writes: >Some say that the critical thing in making the auto revolutionary was >committing to building roads for the autos to drive on. W/o smooth >roads horse-drawn buggies seem a win over cars. > >Now many are saying networks are to computers what roads are to cars >(including the need to invest in infrastructure at a societal scale.) Unfortunately, the roads are being spelled "OSI", which poetic license says is like getting railroad tracks when you wanted pavement. -dB
dm@bbn.com (Dave Mankins) (11/14/88)
In article <8811102118.AA12763@pinocchio.UUCP> bzs@PINOCCHIO.BERKELEY.EDU (Barry Shein) writes: > >Some say that the critical thing in making the auto revolutionary was >committing to building roads for the autos to drive on. W/o smooth >roads horse-drawn buggies seem a win over cars. Actually, before the automobile came the bicycle. The bicycle gave you the freedom of a horse-drawn carriage on a laborer's salary. In the late 1880s and following decades the US went bicycle crazy. One of the results of this madness was a network of smooth roads near towns. Bicycles literally paved the way for automobiles. [Bikes were expensive then --- they could cost as much as half of a workman's yearly wage. I think they also paved the way for installment buying. They made bloomers respectable, too.] -- david mankins/dm@bbn.com
snoopy@sopwith.UUCP (Snoopy T. Beagle) (11/14/88)
In article <8811102118.AA12763@pinocchio.UUCP> bzs@PINOCCHIO.BERKELEY.EDU (Barry Shein) writes: |Some say that the critical thing in making the auto revolutionary was |committing to building roads for the autos to drive on. W/o smooth |roads horse-drawn buggies seem a win over cars. Roads happened because of bicycles, not cars. |Now many are saying networks are to computers what roads are to cars |(including the need to invest in infrastructure at a societal scale.) The network is already there. It is called the phone system. Bringing the digital out to the customer will help a lot. Eventually, there may be some kind of integrated voice/data/video/music/etc. system, probably on optical fiber. Two-way video or high-speed data gets expensive real fast though, compared with ~3400 Hz voice lines. _____ /_____\ Snoopy /_______\ |___| tektronix!tekecs!sopwith!snoopy |___| sun!nosun!illian!sopwith!snoopy
doug@isishq.FIDONET.ORG (Doug Thompson) (11/14/88)
BS>From: bzs@PINOCCHIO.BERKELEY.EDU (Barry Shein)
BS>Some say that the critical thing in making the auto revolutionary was
BS>committing to building roads for the autos to drive on. W/o smooth
BS>roads horse-drawn buggies seem a win over cars.
BS>
BS>Now many are saying networks are to computers what roads are to cars
BS>(including the need to invest in infrastructure at a societal scale.)
The more I think about it, the more I think that the telephone system,
with its network of cables, operators and switching stations is a more
apt metaphor. At first you had a bunch of small local phone systems that
gradually became linked into a world-wide telecommunications network -
that is just barely adequate for limited computer communication.
=Doug
--
Doug Thompson - via FidoNet node 1:221/162
UUCP: ...!watmath!isishq!doug
Internet: doug@isishq.FIDONET.ORGreggie@pdn.UUCP (George W. Leach) (11/16/88)
In article <8811102118.AA12763@pinocchio.UUCP> bzs@PINOCCHIO.BERKELEY.EDU (Barry Shein) writes: >Some say that the critical thing in making the auto revolutionary was >committing to building roads for the autos to drive on. W/o smooth >roads horse-drawn buggies seem a win over cars. Someone in another posting pointed out the chicken and the egg situation with roads and the number of cars, so I guess we could debate about this for a while. My thoughts are that there was a recognition on the part of the general public of the utility of the automobile. The assembly line made the auto affordable. The building of roads made the auto even more practicle. However, it was the recognition by the average Joe that the auto would make life easier that was the key to the wide acceptance of it. >Now many are saying networks are to computers what roads are to cars >(including the need to invest in infrastructure at a societal scale.) I sort of look at current networks as akin to the early development of the railroad. Each line potentially has a different guage of track making it impossible for all trains to travel on it. OSI can be viewed as an attempt to standardize the guage of track (in a simplistic view). I still feel that the average person has *NO* need for a personal computer. I don't have one at home, nor do I need one. However, should the networks and information sources become available at a reasonable cost, then *I* would be more inclined to get one. But I am not the average person (who is!). How much does the average person utilize the local public library? I would imagine not much. And I don't think it is because of a lack of material being available. I think it is more a lack of concern. Most people are more interested in entertainment, ala the TV. This is where we will see an explosion in the usage of computers, networks and information providers. The potential for furthering our ability to gather informatiion will be there, but I am afraid this will not be the common usage of the beast. Didn't they have grand views of how TV would help shape society back in the late 40's and early 50's? Well it certainly has, but not as they intended :-( -- George W. Leach Paradyne Corporation ..!uunet!pdn!reggie Mail stop LF-207 Phone: (813) 530-2376 P.O. Box 2826 Largo, FL USA 34649-2826
anderson@secd.cs.umd.edu (Gary Anderson) (11/17/88)
>|Some say that the critical thing in making the auto revolutionary was >|committing to building roads for the autos to drive on. W/o smooth >|roads horse-drawn buggies seem a win over cars. > >|Now many are saying networks are to computers what roads are to cars >|(including the need to invest in infrastructure at a societal scale.) > I have no answers, only questions. There is a potential tension between providing easy access to "authorized colleagues" for sharing data and programs and providing easy to access to "potentially hostile and uncolleagual illmannered pranksters". Security versus openness. How much security? Who pays? How much access? Who pays? Is it reasonable to use revenues collected from poor families who have no computers in order to finance a computer network for relatively well off academics and business persons? How can we address the redistributional aspects of providing this important public good?
bzs@ENCORE.COM (Barry Shein) (11/18/88)
From: secd.cs.umd.edu!anderson@mimsy.umd.edu (Gary Anderson) >Is it reasonable to use revenues collected from >poor families who have no >computers in order to finance a computer network >for relatively well off academics and business persons? > >How can we address the redistributional aspects of providing this >important public good? The product currently being "redistributed" is not the network, the network is a vehicle for education and research. It's the result of that education and research which are supposed to be redistributed. We buy test tubes for laboratories but don't wring our hands that the poor aren't getting their fair share of test tubes, no? And, as things develop, there are plenty of direct applications of such networking for everyone (eg. public schools and libraries.) I don't think it takes a lot of imagination to see how the sharing of information can be democratized. As volume goes up price should come down, it doesn't take much to get on a network (in theory) if you really want to, it could be done with leased equipment for $50 or so per month today, if you really can't afford that then it's probably the least of your problems. It's mainly the capacities and services which need to be developed. Read Stu Levy's book "Hackers" and the story about Lee Felsenstein who put a terminal in a public mall area for anyone to leave messages on, interesting experiment. -Barry Shein, ||Encore||
bzs@ENCORE.COM (Barry Shein) (11/18/88)
George W. Leach brings up the question of why would anyone want these networks and PC's in their homes? People seem to like TV, big hit, but as far as info goes they don't do it in other spheres... I agree, but that's the whole point, new kinds of services will appear and they may very well resemble passive entertainment more than the current systems. For example, the ability to use video-quality windows in a shopping program could easily (in my humble but correct opinion) replace most mail order services. I've already seen systems like this and they're coming along (not networked yet, CD ROMs will be a competing technology although immediacy is easier over a network.) -Barry Shein, ||Encore||
anderson@secd.cs.umd.edu (Gary Anderson) (11/18/88)
In article <8811171648.AA12727@multimax.encore.com> bzs@ENCORE.COM (Barry Shein) writes: > >From: secd.cs.umd.edu!anderson@mimsy.umd.edu (Gary Anderson) >>Is it reasonable to use revenues collected from >>poor families who have no >>computers in order to finance a computer network >>for relatively well off academics and business persons? >> >>How can we address the redistributional aspects of providing this >>important public good? > >The product currently being "redistributed" is not the network, the >network is a vehicle for education and research. It's the result of If there is a network, some of us are paying for it. If the network is in use someone is benefiting from its existence. It may be fair to say that on the Internet most of the traffic reflects conscientious scholars sharing important current research results, and that the benefits of this activity for society in general will more than offset the cost. That's my perception/hope although I don't know that anyone has done any serious analysis to support this contention. I am however skeptical that the traffic on a "Fidonet" would have benefits which extend far beyond the user community which is able to connect and interact. I do not intend to imply that a non research network is not also valuable society in general and its users in particular; I merely want to raise the question of cost and benefits when with current technology and prices a truly "populist" network is not feasible. If its not populist in orientation, and its purpose is to provide useful services to an easily identifiable group on networkers, then I suspect that those who benefit most will gladly pay the most for the network. >that education and research which are supposed to be redistributed. We >buy test tubes for laboratories but don't wring our hands that the >poor aren't getting their fair share of test tubes, no? I agree that whenever possible, it is always better to measure and value both outputs and inputs associated with a given enterprise when trying to assess the success of a project. > >And, as things develop, there are plenty of direct applications of >such networking for everyone (eg. public schools and libraries.) I Here here. I agree, I hope these applications will develop, but again who will pay? Indeed who will find it profitable to develop them? >don't think it takes a lot of imagination to see how the sharing of >information can be democratized. As volume goes up price should come >down, it doesn't take much to get on a network (in theory) if you >really want to, it could be done with leased equipment for $50 or so >per month today, if you really can't afford that then it's probably >the least of your problems. There are a great number of people who could not afford $50 per month. And you are certainly correct that they often have other problems to worry about. There are some people who would allocate less of society's scarce resources to building infra-structure or even research and more to addressing some of their problems. Given that Congress or state legislatures would probably beecome involved in funding any non-private network, they will have to consider such tradeoffs. Consequently, if you want a publicly funded inexpensive network, you may as well start using your "imagination" now "to see how sharing information can be democratized". >which need to be developed. > >Read Stu Levy's book "Hackers" and the story about Lee Felsenstein who >put a terminal in a public mall area for anyone to leave messages on, >interesting experiment. Thanks for the reference, I will take a look. Gary
rjc@aipna.ed.ac.uk (Richard Caley) (11/21/88)
In article <14578@mimsy.UUCP> anderson@secd.cs.umd.edu (Gary Anderson) writes: >How much security? Who pays? >How much access? Who pays? For security, I can't see any reason for anything other than "user pays". Just as with physical security. If you want more security you pay more for your locks. A generally available network would require some decent breaking and entering laws though. >Is it reasonable to use revenues collected from >poor families who have no >computers in order to finance a computer network >for relatively well off academics and business persons? The same could be asked about roads. It turns out that it is in everyones interest to have decent roads since that makes it cheeper to transport goods. The point is that it is not business people who would benefit so much as businesses and that shoud force prices down. Of corse it is open to question whether that would offset the cost of the infrastructure. -- rjc@uk.ac.ed.aipna AKA rjc%uk.ac.ed.aipna@nss.cs.ucl.ac.uk "We must retain the ability to strike deep into the heart of Edinburgh" - MoD
doug@isishq.FIDONET.ORG (Doug Thompson) (11/30/88)
RC>From: rjc@aipna.ed.ac.uk (Richard Caley) In article <14578@mimsy.UUCP> anderson@secd.cs.umd.edu (Gary RC>Anderson) writes: RC>>How much security? Who pays? RC>>How much access? Who pays? RC> RC>>Is it reasonable to use revenues collected from RC>>poor families who have no RC>>computers in order to finance a computer network RC>>for relatively well off academics and business persons? RC> RC>The same could be asked about roads. It turns out that it is in RC>everyones interest to have decent roads since that makes it RC>cheeper to transport goods. The point is that it is not business RC>people who would benefit so much as businesses and that shoud RC>force prices down. Of corse it is open to question whether that RC>would offset the cost of the infrastructure. Well, roads and the post office present reasonable examples. A society, rich and poor alike, benefits from having such things. Yet users, through fuel taxes, vehicle taxes, postage stamps, etc., pay the bulk of the operating costs, based on amount used. Further, the cost of computers is steadily coming down, and a home computer is now in the price range of a major home appliance. Most poor people do manage to acquire refrigerators, television sets, etc., so I'm not so sure that we will long be in a situation where one has to be rich in order to participate or make use of computer networks directly. Indeed, except for the fact that you need an expensive capital item to do computer communication, the real costs of moving text data today are much less than the real costs of moving text data through the post office. Thus, this should end up being a boon to poor people by reducing the marginal cost of communication. As for the capital goods, the basic computer and modem, I can see lots of reasons for a state to subsidize access for poor families. In the end, it's a whole lot more useful to all of us if everyone is connected. =Doug -- Doug Thompson - via FidoNet node 1:221/162 UUCP: ...!watmath!isishq!doug Internet: doug@isishq.FIDONET.ORG