Python Productivity Gain?

[Scott David Daniels]

Peter Hansen wrote:

> Scott David Daniels wrote:
> 
>>>>A9 80 20 FF D2 40 anyone? :-)
> 
> 
>>Actually hex was: 0 1 2 3 4 5 6 7 8 9 F G J K Q W
> 
> 
> Why were those particular characters used?
Low order for bits of the flexowriter generated the appropriate
value.  I expect there was some other key which would also work,
but I never experimented to discover it.  I do remember there was
a "six-bit" button which messed everything up, I believe by
pulling in 6 bits rather than 4 from the keyboard.  Bu I could
be wrong -- I forget the details of the "i" instruction.  It was
a single-address machine, with 31 bits stored, a PC, accumulator,
and instruction register.  Multiply produced _either_ the high
32 bits of a multiplication ("M") or low 32 ("N" multiply).  Memory
was 32 bits wide, but always zeroed the low order bit (a "spacer"
bit on the drum).  You got pretty good at fixed point math because
the easiest way to think of the data was a a number in [-1.0 : 1)
range (then losing the low oder bit made more sense).

A little bit of googling shows me I was right about the 6-bit button.
We used 6 bits on paper tape for the key codes (there were shift
codes to go to upper or lower case).  The 4 bits are the 4 high-order
bits of the character.

   http://ed-thelen.org/comp-hist/lgp-30-man.html

Has a pretty good description.

-- 
-Scott David Daniels
Scott.Daniels at Acm.Org