[sci.logic] What were real machines which helped Turing?

steve@hubcap.clemson.edu ("Steve" Stevenson) (01/16/91)

In a class I'm teaching, I was trying to get the students into the historical
context in the 1920-1940 time frame. The Atansoff computer was probably
known to Alan Turing. Did this help in his formulation? Did he ever say where
the ideas came from? What other machines were there then that might have
contributed to his formulation?-- 
===============================================================================
Steve (really "D. E.") Stevenson           steve@hubcap.clemson.edu
Department of Computer Science,            (803)656-5880.mabell
Clemson University, Clemson, SC 29634-1906

dc@cs.qmw.ac.uk (Daniel Cohen) (01/16/91)

The background to Turing's work is well discussed
in the book: "Turing: the enigma of intelligence",
I think it's by Hodges ( e-mail me if you want a
proper reference; it's at home ).

The gist of the answer would be that the concept of
the Turing machine did not derive from any actual
computing devices, but from a generalisation of how
a human being does computation - essentially, by
looking at pieces of paper ( squares on a tape ),
and adjusting context accordingly ( changing state
). This approach came about because Turing was
interested in the famous Hilbert problem ( one of
23 ) concerned with the decidability of
mathematical questions.

Of course, that doesn't mean that Turing was
uninterested in the practicalities of mechanical
computation; he had an interest in analogue
computers and was later the leading light behind
the ACE. But I believe that the foundations for his
work were mathematical, not practical computers.

-- 
Daniel Cohen              Department of Computer Science 
Email: dc@cs.qmw.ac.uk    Queen Mary and Westfield College
Tel: +44 71 975 5249/4/5  Mile End Road, London E1 4NS, UK
Fax: +44 81 980 6533      *** Hit the North!!! ***

aa1@j.cc.purdue.edu (Saul Rosen) (01/24/91)

In article <1991Jan19.234454.26225@eecs.wsu.edu> pcooper@yoda.UUCP (Phil Cooper - CS495) writes:
>In article <1991Jan16.165251.3783@sctc.com> smith@sctc.com (Rick Smith) writes:
>>steve@hubcap.clemson.edu ("Steve" Stevenson) writes:
>>
>>> The Atansoff computer was probably known to Alan Turing.
>       ^^^^^^^^
>  I'm not 100% sure, but I believe the man's name is Atanasoff
>
>>
>>This is a fascinating statement, but your use of the word "probably"
>>implies that you lack hard evidence. Is there any evidence to support
>>this statement?
>>
Turing wrote his famous paper on computable numbers in 1935-36.  In it 
he introduced his conceptual computer that has come to be known as the
Turing machine. The Universal Turing Machine that he introduced in the
same paper can be considered to be a conceptual or even a mathematical
model of the real universal computers (e.g. the EDVAC) that appeared
about 10 years later. It is not clear whether Turing's ideas had any 
direct effect on the Edvac.  However, it is known that von Neumann knew
Turing while Turing was a graduate student at Princeton in 1937-39.

Atanasoff started to work on his machine, later known as the ABC computer
in 1937-38, and worked on a prototype in 1939.  I don't think it is at
all likely that Turing ever knew anything about the Atanasoff computer,
even after World War II when Turing designed the real hardware ACE computer.
The above dates show that Turing's work on computability preceded the work
of Atanasoff.  Note also that the Atanasoff computer was not a programmed
general purpose computer, and thus its concepts were in no way related
to the very general programming concepts introduced  by Turing in his 
1936 paper.

Saul Rosen