[GreenKeys] Historical documents?
[email protected]
[email protected]
Wed, 20 Nov 2002 22:09:55 -0600 (CST)
I just gave it a quick skim through, but it seems plausible enough.
Note that we have some positive feedback here in that he uses Greenkeys
as a source, so if we tell lies on Greenkeys they will get repeated back
to us.
It's hard for me to read the code tables the way he presents them, using
decimal values of binary numbers; I would find it easier to read things
in the style of Teletype code cards where the binary numeric
interpretation is played down.
I was unaware of different makes of equipment running the bits in reverse
order, but then I'm unaware of lots of things. Maybe it's true. I know
that once we got into ASCII even if there was agreement on the way the
holes would be punched into the tape there was a lot of arguing over
which bit should be transmitted first. Teletype was transmitting the
least-significant bit, in the binary sense, first; IBM thought they should
go the other way.
He could have said something about the various BCD codes the various
computer companies used before ASCII and EBCDIC. I believe it was the
wide number of variations in those that led the military to want to
have a standard Fieldata to make interoperability possible. It wouldn't
do for the government to declare that IBM was the de facto standard and
everybody else would have to get in line.
As I recall Fieldata deliberately left half the codes undefined, so that
system implementers would have plenty of codes to use for control
purposes. This is widely regarded as a terrible mistake, because it led
to a proliferation of incompatible codes. So the designers of ASCII were
at pains to have every code nailed down to a specific meaning.
COMLOGNET is what morphed into AFDATACOM and then into AUTODIN.
He references the two Fred Smith papers from Western Union Technical
Review; those should be highly reliable. The other references I posted
yesterday are also worthwhile.
A reference I didn't post - and I don't have it at my fingertips - is an
IBM document on "Binary Synchronous Communications". This shows an
example using EBCDIC or ASCII of how all the various control codes can be
used in real systems that include error detection and correction by
retransmission.