NetNews Usenet Archive 1993 #3/NN_1993_3.iso/spool/alt/cyberpun/7165

Text File | 1993-01-24 | 8.9 KB | 167 lines

Newsgroups: alt.cyberpunk
Path: sparky!uunet!mcsun!sun4nl!ruuinf!plato.phil.ruu.nl!freek
From: [e-mail] (Freek Wiedijk)
Subject: Re: The Difference Engine
Message-ID: <[e-mail]>
Sender: [e-mail]
Nntp-Posting-Host: groucho.phil.ruu.nl
Organization: Department of Philosophy, University of Utrecht, The Netherlands
References: <[e-mail]> <[e-mail]> <[e-mail]>
Date: Sun, 24 Jan 1993 16:04:43 GMT
Lines: 154

From `Charles Babbage and his Calculating Engines', an exhibition
catalogue from the Science Museum in London, pages 10-24, reproduced
without permission:

------------------------------------------------------------------------
The development of Difference Engine No. 1, the largest of Babbage's
practical ventures, was under way in 1824.  This vast machine called for an
estimated 25,000 parts, would have weighed several tons, and measured
8 feet high, 7 feet long and 3 feet deep (2.4 x 2.1 x 0.9 m).  Babbage hired
Joseph Clement, a skilled toolmaker and draughtsman -- a prized and none
too common combination at the time -- to build the machine.  The years of
design, development and manufacture that followed were the most
gruelling and ultimately disappointing of Babbage's life.  Work stopped in
1833 after a dispute with Clement over compensation for moving his
workshops to premises alongside Babbage's house.

Accounts vary on how close to completion the engine was when work
stopped.  What seems clear is that a substantial number of essential parts
for the calculating mechanism had been made and assembly was a realistic
prospect.  On Babbage's instruction Clement assembled a small section of
the engine as a demonstration piece.  This `finished portion of the
unfinished engine', completed in 1832, represents about one seventh of the
calculating part of the full machine. The printing mechanism was not
assembled.

The portion of the Difference Engine completed in 1832 still works and is
one of the finest examples of precision engineering of the time.  It was of
practical use to Babbage in tabulating certain functions and also became a
conversation piece at his Saturday evening soir<e'>es where he used it to
elaborate on his theory of miracles.  Its successful completion has been used
as a formidable argument in support of Babbage's conviction the the full
sized machine was a practical prospect.

This portion of Difference Engine No. 1 is the first known automatic cal-
culating device and is one of the most celebrated icons in the prehistory of
computing.

By the end of 1834 Babbage had conceived of a more ambitious and
technically more demanding machine.  This was the Analytical Engine -- an
revolutionary machine upon which much of Babbage's fame as a computer
pioneer rests.  This engine called for more near-identical parts that even the
Difference Engine.  Daunted perhaps by the fate of the Difference Engine,
Babbage expected that if he built an Analytical Engine it would be at his
own expense.  He looked for an alternative to machining hundreds of repeat
parts and investigated pressure die-casting and stamping sheet metal as a
way of reducing costs.  A simplified portion of the engine was all that was
built in his lifetime.  This, and a piece built by his son after Babbage's death,
a few small experimental assemblies, and the designs, are the physical
legacy of a remarkable intellectual achievement of the nineteenth century.

The distinction of completing the first working difference engine goes to a
Swedish father and son team, Georg and Edvard Scheutz.  Inspired by
accounts of the Babbage engine, they completed, in 1843, a prototype
engine based on an account of Babbage's scheme but largely of their own
design.  This machine produced the first tables calculated and printed by
machinery.

In the Scheutzes~' machine the problem of replicating results without risk of
human error was solved in the way Babbage had inteded -- by impressing
the numbers on _papier m<a^>ch<e'>_ (flong) or soft metal strips.  These were used as
moulds for stereotype plates from which multiple copies of the results could
be printed.

The first Scheutz engine was built using simple hand tools and a primitive
lathe.  In comparison with the sophistication of Babbage's and Clement's
equipment, the means available to Edvard Scheutz were decidedly crude.
The successful demonstration of the Scheutz engine with its comparatively
rough construction raises questions as to whether the high degree of
precision to which Clement worked was in all instances necessary.

The engine was rediscovered in 1979 by a young Swedish historian,
Michael Lindgren, who traced the forgotten machine to the Nordiska
Museet, Stockholm, where it had been deposited in 1881.


    `Glory and Failure'

Charles Babbage failed to realise any complete engine in physical form.
The reasons for this remain the subject of analysis and energetic debate.

After Clement's last payment in 1834 the cost to the government of the
Difference Engine project was #17,470.  This was a massive sum: the steam
locomotive, _John Bull_, completed by Robert Stephenson and Co. for
shipment to America in 1831, cost #784 7s.  The failure to complete an
engine was the central trauma in Babbage's scientific life and something he
brooded over repeatedly.

The Scheutzes completed three difference engines -- a prototype and two
`production' machines.  The second two machines were sold, one to the
Dudley Observatory in Albany, New York State, and the other to the
General Register Office in London.  The Albany machine was little used.
The Register Office machine was used to produce the _English life table_
published in 1864.  Both these machines were more troublesome to use than
expected.

The Scheutz engines worked after a fashion but were not commercially
successful.  Georg Scheutz was honoured by his country for his
achievements.  He died in 1873 practically bankrupt.  Five years later his
son, Edvard died bankrupt and in debt.

There were other attempts to build difference engines in the nineteenth
century -- Alfred Deacon in London, Martin Wiberg in Sweden and
Barnard Grant in America.  Deacon's machine was never completed.  The
Wiberg machine, completed in about 1860, was used to produce a large
volume of tables published in Swedish, German, French and English.  The
history of the Wiberg calculator was financially fraught.  Grant's engine
was exhibited at the Centennial Exhibition in Philadelphia in 1876.
It attracted attention which was shortlived.

The nineteenth century movement to mechanise and automate mechanical
calculation had failed.  The pioneers of the movement, Babbage and the
Scheutzes, are honoured with the distinction of their pioneering efforts.  All
suffered personally and financially as a result.

Babbage was not the first to suggest a printing calculator nor was he the
first to consider the method of differences as a suitable principle upon which
to base a calculating machine.  This distinction belongs to Johann M<u">ller
(1746-1830) an engineer and master builder.

In a letter written in 1784 M<u">ller refers to a machine he considered building
which would calculate and print any arithmetic progression -- `all one has to
do is to turn a handle'.  This letter contains the earliest know description of
a printing calculator.  In another letter written in the same year he describes
plans for a machine for calculating squares and cubes by a series of
differences.  M<u">ller had conceived of a difference engine 36 years before
Babbage.  Details were published in 1786 in a book, key sections of which
were translated for Babbage by John Herschel.  The date of this translation
is uncertain and the question of whether some of Babbage's ideas on
difference engines came form M<u">ller remains open.

The reasons for Babbage's failure continue to exercise historians.  Factors
cited include Babbage's allegedly difficult personality, unconvincing
progress, personal enmities, a dispute with Joseph Clement, problematic
funding, unsympathetic entrepreneurial climate, several changes of
government, cultural conflict between pure and applied science, the fact
that experts were divided on whether there was any real need, and lack of
government vision when withdrawing support.

The view most often repeated in histories of computing is that Babbage
failed because of limitations in Victorian machine tool technology.

To investigate this the Science Museum has constructed Babbage's
Difference Engine No. 2.  Babbage designed this machine between 1847
and 1849.  It is a simpler and more elegant version of the earlier design and
benefits from work he had done on the Analytical Engine -- a much more
complex and therefore more demanding machine to design.  He offered the
Designs for his new engine to the Government in 1852 but no attempt was
made to build it.
------------------------------------------------------------------------

Freek
--
Third theory of Phenomenal Dynamics:  The difference between
a symbol and an object is quantitative, not qualitative.