[GreenKeys] interesting paper on code development

[gil smith]

Hi folks:

In my quest to find some answers on ITA2/ustty/telex codes, I ran across a 
very interesting paper:
http://www.transbay.net/~enf/ascii/ascii.pdf

The author, Eric Fischer, did a lot of research, as the 199 references will 
attest.

I will paraphrase a bit:

Fischer discusses the early Hughes, Baudot, Murray, and Morkrum systems, 
including diagrams of various related codes and keyboard layouts.

Apparently, Murray sold US patent rights to Western Union in 1912, at which 
point US and British Murray equipment took different development paths.  By 
1915, Western Union, and Western Electric, were using a very similar code 
which merged features of the Murray and Morkrum codes.  During this time, 
the original Baudot piano-key system was still in use (in France, and 
perhaps England and other areas).  This is all before any standards were in 
place, and there were quite a few incompatible systems in use.

By 1924, Germany was advocating an international code, the idea getting 
support from Murray and others.  In 1926, the CCIT met for the first time 
in Berlin (CCIT was not called CCITT until a merger in 1957).  There were 
debates about whether to design a new code, or stay close to the original 
baudot code, whether to use FIGS/LTRS vs. FIG-SPACE/LTR-SPACE, etc.  In a 
1929 CCIT meeting, they finally proposed ITA2 as a code that generally 
combined the letters set from Baudot, with the numbers set from Murray, 
reserving four figs chars for national (country-specific) use.  They also 
proposed ITA1 as a modified (original) Baudot code.  I believe that neither 
proposed code was officially released at this point.

In a 1931 meeting of the CCIT, there apparently were further changes to the 
proposed codes.  It is unclear to me when the actual ITA1 and ITA2 codes 
were officially standardized, since Eric's reference to final code adoption 
is dated 1938 -- I thought it was earlier than this (perhaps the 1931 
meeting?).  Does anyone have an actual date of adoption?

As we all know, the ITA1 code died off when the early Baudot machines went 
out of service, leaving ITA2 and USTTY.  Eric's paper does not mention the 
USTTY variant, and I am still unsure whether USTTY evolved from ITA2, or 
simply standardized in parallel.  It seems increasing likely to me that it 
evolved in parallel.  A lot more happened in the 20s and 30s -- twx in the 
US, telex internationally....  I'd like to nail down a timeline for some of 
these events.

Something close to USTTY must have been at least in place by 1922 when the 
Model 12 was introduced -- at least two of you greenkeys guys have M12s, so 
what code set is used on those machines?  What about the early GPE perfs 
(hard keycap)  that go back to around 1913?  Anyone have one?

Then there is the Model 11.  I only have pics of one, from the Netherlands:
http://www.baudot.net/tty/M11.htm
Note that this machine has separate FIGURES and LETTERS spacebars -- this 
is neither ITA2 nor USTTY.  Eric mentions that the FIGS-SPACE/LTRS-SPACE 
codes were introduced into non-US Murray equipment after 1912, so it looks 
like this M11 is a Murray code machine.  Sadly, the original Netherlands 
site is gone, and the whereabouts of this ultra-rare M11 are unknown.

Further down this M11 page is a patent drawing of the M11 showing a single 
spacebar -- this drawing is likely a US version.  Obviously there were code 
changes in the 1921 to 1927 lifetime of the M11.

And related to all of this is the point at which start-stop was introduced.

I am not mentioning the bulk of Eric's paper, which is devoted to later 
code developments leading to ascii.  Also very good reading.

Still a lot of questions on the early days though.  Like who were Gauss & 
Weber, who supposedly actually designed the original Baudot code?

So, as best as I can tell so far:

        ITA2/Telex    USTTY
--------------------------------------------------------
D      wru (1)          $
F         (1)            !
G      (1)            &
H      (1)            #/stop
J      bell            '
S      '              bell
V      =              ;
Z      +              "

Note 1:  ITA2 defined four "national -use" chars, which were
                not to be used for international communications.
                I presume that these were not used in Telex service.
                Figs-D was generally WRU -- I don't know if this one was 
used by Telex.
                Figs-F/G/H seem to have no "typical" chars, in keeping with
                their original country-specific definition.


gil



Vaux Electronics, Inc.
480-354-5556
(fax: 480-354-5558)
www.vauxelectronics.com