[Cryptography] 'The intelligence coup of the century'

Bill Frantz frantz at pwpconsult.com
Fri Feb 14 16:43:26 EST 2020


On 2/13/20 at 12:08 AM, hbaker1 at pipeline.com (Henry Baker) wrote:

>I seem to recall that the Easties were fond of copying DEC equipment,
>while the Russkies were fond of copying IBM equipment.
>
>Just before 2000, there were a lot of 70year-old IBM programmers fixing
>really old 1960's/70's Cobol programs in the U.S.; many of them were
>*ex-Soviets* who learned on 1960-era IBM 360 'mainframe' copies.  For
>some reason, the Soviets stuck with the 1960's IBM DOS/TOS SW, even
>after 370-era software became available.

I assume by TOS and DOS, you are referring to early IBM 360 
operating systems, Tape Operating System and Disk Operating 
System. These systems were very similar to the previous 
generation (7090, 1401) input/output control systems. They were 
also largely compatible, except TOS did not support disks.

Cobol code for these systems was fairly easy to port to OS/360 
which was easy to port to OS/370.

The big cause of the Y2K problem was the designers' desire to 
minimize file size and disk usage. Many systems were designed 
with a 2 digit field for the year, and my wife worked on one 
system that used a single digit. It ran into trouble in the next decade.


>Some of this Cobol SW is undoubtedly still running in emulation in
>DMV's and also handling vote tabulation all across the U.S. :-)
>
>Perhaps DEC never had a Y2K problem, so I don't recall Eastie programmers
>working on Y2K issues.

The Y2K problem primarily affected business systems. I think DEC 
was more concentrated in the time sharing, scientific, and 
industrial control areas, where dates were a smaller part of the systems.

>
>BTW, IBM mainframes had powerful radio side-channels: in the early
>1960's, one of the programmers that worked with me used to put an
>ordinary portable radio on top of the IBM CPU so that he could
>"listen" to his program go through its paces.  Any change in the
>audio would indicate a problem with the software.
>
>I would guess that this radio technique would have been sufficient
>to enable a person listening to pick out the individual bits in a
>modular exponentiation, as those computers weren't all that fast.

I think you would need to process the signals to read the 
individual bits being processed. Also, in those days, people 
were much less concerned about passive listening to the 
processor's internal operations, and security in general.

One person at Reed Collage wrote a program, called MuTran, for 
the IBM 1620, which translated its input into music which it 
played using the AM radio signals generated by the computer.

One of the standard IBM demos included having the 1403 printer 
on the 1401 play, "Anchors Away" by controlling the timing of 
the hammers hitting the type chain using the specific characters printed.
>
>We didn't have a name for it, but in retrospect we should have
>called it 'ADB' -- 'Audio Debugging Bridge'.
>
>I never tested the *range* of these radio signals, but I suspect
>that they could easily have been heard outside the building
>where the computer was located.

When I used the technique on the 1620, I had to put the 
transistor AM radio close to the console. I'm sure using a 
directional antenna, one could receive the computer from outside 
the building.

Cheers - Bill


-----------------------------------------------------------------------
Bill Frantz        | When an old person dies, a   | Periwinkle
(408)348-7900      | library burns. - Joe McGawon | 150 
Rivermead Rd #235
www.pwpconsult.com | Irish Ethnographer           | 
Peterborough, NH 03458



More information about the cryptography mailing list