Subject:
Re: Baudot code direct to computers?
From:
John Varela
Date:
Tue, 26 Aug 2008 18:46:21 -0400
Newsgroups:
alt.folklore.computers

On Tue, 26 Aug 2008 14:52:05 -0400, hancock4@bbs.cpcn.com wrote
(in article 
<137c32bb-cf51-4bab-b1cb-787d7ad10b9c@x41g2000hsb.googlegroups.com>):

> > Were there computers that supported direct Baudot connections, either
> > as one way (e.g. broadcasting messages) or two way (on line inquiry)?

Yes.

Circa 1960 IBM built a Low Data Rate Input (LDRI) frame for the 
AN/FSQ-7 (XD-1) at Lincoln Lab for use by MITRE in air traffic control 
experiments.  ASR 28s were used for (what was then considered to be) 
fast data entry of flight plans.  KSR 28s were used for interactive 
I/O.  Photo of the lab, circa 1962, at

  http://tinyurl.com/6cj3zk

That's me second from the left, pretending to be an air traffic 
controller.  The KSRs can be plainly seen.  They used 6-bit code, 
however.

In the late 1960s the IBM 9020 computers went operational in the air 
traffic control centers.  Flight plans reached the centers in 5-bit 
Baudot from the existing teletype network.  I forget whether that was 
on-line and if so whether it was interactive, but I'm pretty sure that 
was the plan.  I'm unsure how successful it was, given the need for 
local personnel to correct input errors.  Selectrics, not TTYs, were 
used for flight plan data entry in the control centers, with mag tape 
for repetitive flights such as airlines.  Special purpose devices were 
used at the air traffic control sectors.

I *think* 5-bit Baudot was also used for flight plans and flight plan 
updates in both directions between the US and Canada in the late '60s 
or the very early '70s.  The messages to Canada would have been 
generated by the 9020s.  As I recall (you do realize this was all a 
long time ago), international flight plans were not input directly to 
the computer because they contained too many errors, but Canada might 
have been an exception. 

-- John Varela