[GreenKeys] AN/FGC-30, and telegraph switching

[Jim Haynes]

I remember reading a report on the development of the FGC-30.  By the
time I read the report, about 1960, development was complete, so probably
the reason the report came to my attention was that it had just been
declassified.  In fact as I recall the title of the report was about the
AN/GGC-2, which is what it was called at the time of the trial 
installation at 5th Army HQ.

About the only thing I remember from the report is the funny part about
alternate routing.  This was perhaps the first automatic TTY system to
use alternate routing.  The problem was that if a station is down, two
different relay points might each think that the other is a good alternate
route to that station.  The result would be the message getting 
transmitted back and forth between the two relay points for as long as
the destination was unreachable.  So they had to add logic to prevent a
message from being sent to the relay point it had come from.

All of which tells us that the electromechanical switching systems had
just about hit the wall of complexity.  Computer technology switching
systems came along just in time to save the situation.  I believe
Western Union's last was Plan 59 for the FAA in Hawaii.  I remember a
complexity there was that they wanted the routing indicators for all
addressees to appear in all copies of a message.  But when a message
hit the switching center it was possible that some of the addressees
had already received it as it was being transmitted to the switching
center.  So the message format had to include a line for routing
indicators that did not need to receive copies from the center.  Imagine
trying to design relay logic that has to handle a list of routing
indicators that are to be ignored.

About the same time frame as the AN/FGC-30, Western Union had built Plan 
55 for the Air Force, and the Navy had gone to AT&T for the 82B1 system. 
Each of the three armed services doing it as different as possible from 
the other two.  So when AUTODIN came along you can understand why it was 
desirable to have it take over the message switching for all the services, 
even though its original purpose was just to replace a separate Air Force 
network of IBM punched card transceivers.

Teletype Corp. had a limited niche in the switching business.  Anything
likely to be of general use in the Bell System was on Bell Labs' turf.
Teletype was allowed to design and sell systems that were not like
anything a Bell customer would want.  One of these was ADIS - Automatic
Data Interchange System - for the FAA.  This was to collect and 
disseminate aviation weather data.  A very similar system AIDS was built
for New York Telephone Co. to handle their internal needs.  A second
system for the FAA was BDIS, which handled flight plan data and other
messages related to air traffic control.  BDIS doesn't stand for anything;
it's a pun in that ADIS handled what FAA called "Service A" and BDIS
was for "Service B".  After that the computer technology took over.
Collins had the C-8400 system and a strong presence in the aviation
industry.  IBM had the 7740; G.E. had the Datanet-30; and I.T.T. had
a product.  Several Univac computers were used for the purpose also.
And RCA built the computers for AUTODIN.