[ih] AT&T, carterfone, the 103, and why didnt BBSs start earlier?

[Jack Haverty]

I agree that everything was expensive, but there were other factors too
that had an effect.

It was common at the time to interact with mainframes by decks of
punch-cards, then wait til your job ran (minutes to hours to overnight),
then collect the printout.  Hooking terminals to computers wasn't very
common, except for very special-purpose applications like reservation
systems.  The common "general purpose" computing approach was still
batch processing and punch cards in the 60s.  So, especially outside of
the R&D environment, comms devices (modems and front-end processors)
were bought as part of the mainframe system configuration designed to do
a specific application such as airline reservations.  Computer terminals
were thought of as I/O devices for some application, rather than as ways
to interact with the computer in a general way.

Also, modems were typically connected to leased lines, often
"multidrop", with huge monthly bills for those lines.  So not only was
the box expensive, the monthly charges were high too.

So, even when "modems were available", there wasn't much available on
the other end of the communications path except inside a single closed
application system.

Modems were however used in non-computer applications, which we computer
guys hardly ever think about.  E.G., there were HF radio receivers that
could be placed in a quonset hut somewhere at the end of a long wire,
interacting via RS232 over modems with a human operator far away.
Through the modem link, the human could manipulate all of the knobs and
switches as if he were located at the radio.  I suspect there were lots
of these kinds of "remote operation" modem uses totally independent of
computers. 

Even back in the early 70s, as computers were coming on to the Arpanet,
batch processing was still lingering on.  I remember writing a
program/service on the MIT ITS machine that allowed a user to
prepare/edit a virtual "deck of cards" for the UCLA CCN 360,
automatically submit it to the CCN machine, and then poll the mainframe
and retrieve the "printout" when it was ready.  That was how you could
use the 360 to create and run your own programs.

So, before dial-up modems became common, people had to get used to the
idea of interacting with computers in the new "terminal" way, to create
the demand for equipment that motivated companies to try to compete with
the established mainframe industry.  

Carterfone made competition possible; the mindset outside of the
early-adopters had to change to increase demand for equipment and make
competition sensible.  Timesharing was still a new idea outside of the
R&D crowd.  Dial-up services like TymNet/Tymshare helped change the
mindset and create the demand in the "real world".

It wasn't all that long before those days that experts in the industry
projected that the computing needs of the world would require a handful
or so of computers at most.  I forget the quote - was it T. J. Watson?

/Jack Haverty
Point Arena, CA


On Thu, 2009-08-13 at 21:17 +0100, Johnny RYAN wrote:
> Dear Noel
> 
> (good point Vint - thanks)
> 
> Yes, as you say, the Altair and the rest of the personal computers had
> not come until the mid 70s.
> What I'm curious about is a more abstract point - on the modem side
> rather than the computer side, why is Carterfone decision such a big
> deal if modems were already available for use on the normal phone
> line? Was it, as Vint says, price?
> 
> Johnny
> 
> --
> My Next Book... http://johnnyryan.wordpress.com/books/net-history-2010/
> 
> 
> 
> 
> On Thu, Aug 13, 2009 at 8:51 PM, Noel Chiappa<jnc at mercury.lcs.mit.edu> wrote:
> >    > From: Johnny RYAN <johnnyryan1 at gmail.com>
> >
> >    > This is my first posting to this list.
> >
> > Welcome!
> >
> >    > the Bell 103 modem in 1962 and Carterfone
> >    > If AT&T sold modems commercially since 1962 (the 103 modem), why was
> >    > the carterfone decision so important?
> >    > ... does anybody recall why these things could not have happened with
> >    > the Bell 103 from 1962 on? Was the 103 just intended for subscribers of
> >    > expensive leased lines such as corporations or universities?
> >
> > I think you're conflating two different things.
> >
> > Carterfone was important because it allowed other people to build stuff to
> > connect up to the network (originally only acoustically, like the old
> > acoustic-coupler modems), _but_ I don't think it has any relationship to the
> > thing you're asking about (which I take to be the generic 'computer
> > communication revolution').
> >
> >
> > The answer to your question about 'why no computer communication revolution
> > in the 60s' is, I am pretty sure, in the technology of the era (both hardware
> > and software).
> >
> > Remember that until things like the PDP-11 (1970 - although I suppose the
> > PDP-8, from 1965 on, also would count) there weren't a lot of small computers
> > to connect together. Personal computers were significantly later than that -
> > the Altair was 1975, and the Apple II (the first really plausible personal
> > PC) was 1977.
> >
> > Ditto for software - the first time-sharing OS's were in the early 1960's,
> > but there were only a very few early on, and they ran on a very few large
> > mainframe systems. It wasn't until circa 1970 that that operational mode
> > became relatively common. Even the simplest of computer communication stuff
> > (remote dumb terminal dialed into a time-sharing machine) thus had to wait
> > for that.
> >
> > Saying that, though, reminds me that there was a small amount of stuff
> > significantly earlier - you might want to look into the SABRE reservations
> > system, which dates back to 1957 or so (although the idea is a couple of
> > years older), for remote access in a more specialized system.
> >
> >
> > That's just my opinion, though - others may have a different take.
> >
> >        Noel
> >
> 
>