Authors: James Essinger
Jacquards' Web (38 page)
At the same time, I am not sure that
30
years hence, I may put even so much value as
this
, upon human fame. Every year adds to the unlimited nature of my trust & hope in the Creator, & decreases my value for my relations with mankind
excepting as His minister
; & in
this
point of view those relations become yearly more interesting to me. Thro’ my present relations with
man
, I am doubtless to become fit for relations of another order hereafter; perhaps
directly
with the great Power Himself. Of course my view respecting every even
casual
social contact & intercourse, takes a corresponding colour; & will do so increasingly, if
that
view should become more confirmed.
My dear friend, if you knew what
sad & direful
experience I have had, in ways of which you cannot be aware, you would feel that
some
weight
is
due to my feelings about God & man.
As it is, you will only smile & say, “poor little thing; she knows nothing of life or wickedness!”
Such as my principles are, & the conditions (founded on 275
Jacquard’s Web
them), on which alone you may command my services, I have now stated them; to just such extent as I think is absolutely necessary for any comfortable understanding & cooperation between us, in a course of a systematized & continued intellectual labour. It is now for
you
to decide. Do not attempt to make out to yourself or me that our principles entirely accord.
They do
not,
nor
cannot
at present, (for people’s views as I said are not to be altered in a moment).
Will you come
here
for some days on Monday. I hope so.
Lord L— is very anxious to see & converse with you; & was vexed that the Rail called him away on Tuesday before he had heard from yourself your own views about the recent affair.
I sadly want your
Calculus of Functions
. So
Pray
get it for me.
I cannot understand the
Examples.
I have ventured inserting to one passage of Note G a small Foot-Note, which I am sure is
quite tenable
. I say in it that the engine is remarkedly well adapted to include the
whole Calculus
of Finite Differences
, & I allude to the computation of the
Bernoullian Numbers by means of the difference of Nothing
, as a beautiful example for its processes. I hope it
is
correctly the case.
This letter is sadly blotted & corrected. Never mind that however.
I wonder if you will choose to retain the lady-fairy in your service or not. Yours ever most sincerely.
A.A.L.
276
The operation of the Jacquard loom can best be understood by looking at a simplified diagram of its operation (see following page). The control device is fixed to the top of the loom. The lifting hooks are indicated in the diagram by
a
. These hooks pass in a perpendicular fashion through eyes in a number of horizontal needles equivalent to the number of the lifting hooks. The horizontal needles are shown in the diagram as stretching from
b
to
c
.
They lie in rows in a frame indicated in the diagram by
d
. In the diagram, in order to simplify the explanation, only eight hooks and eight needles are shown, but in the actual Jacquard loom there are as many as
400
of each, or fifty in each of the eight rows, allowing the machine to control the action of up to
400
warp threads, or even more.
The horizontal needles protrude through the frame at the points indicated by
c
. They are kept in that position by helical or spiral springs in the position indicated by
e
. These springs are placed in hollowed-out cavities in the frame. They are held fast inside the frame. Whenever pressure is applied to the tips of the needles at the points
c
, the needles will slide back into the frame, but the instant the pressure is removed, the elasticity of the springs will make the needles shoot outwards again. The extent to which the needles can protrude from the frame is checked by vertical pins located at
g
.
In order for the loom to operate, there has to be a system for governing the action of the individual needles that control the lifting hooks which in turn control the action of the loom’s warp 277
h
a
a
a
a
g
e
d
b
c
c
f
c
c
d
b
g
e
278
Appendix 3
The bar of the Jacquard loom against which the punched cards were pressed.
threads. The device that permits this action to be governed with an enormously high degree of accuracy is a special square bar in the shape of an elongated cube. The bar, shown in the figure above, is fitted with hundreds of tiny identical holes, each of which is the right size to accommodate the tips of the needles.
The operation of the bar in the Jacquard attachment is arranged so that the bar turns onto the next flat, perforated side for each pick of the shuttle. Every time the bar turns it momentarily locks against the ends of the horizontal needles.
The mechanism is designed so that as the bar turns onto its next side, one card from the chain of punched cards turns with it and is pressed firmly against the ends of the needles. This pressing action is part of a much more complex overall operation—
that is, the entire operation of the Jacquard loom—but each press is in itself complete, comprehensive, and provides a moment when the card is stationary as it pushes against the needles.
The cards themselves are partially perforated. The holes are made in the cards so that whenever a card is pressed against the bank of needles it only causes those needles to be raised which lift the warp thread for a pick of a particular design. Naturally, the number of perforations in each card is less than those on the sides of the revolving bar.
(
left
) How the Jacquard loom works.
Jacquard’s Web
Nature depicted by mathematical analysis: the
mise-en-carte
principle in action.
What happens is that when a needle confronts an unperforated area of the card, the needle will not be released and the warp thread it governs will remain down. On the other hand, if a needle confronts a hole, the needle tip will penetrate the hole, ending up in one of the holes on the square bar, and the warp thread the needle governs will be raised.
The perforations in the card are made in order to ensure that the required pattern is woven by the weaver working below. In order for this to happen it is necessary that each card controls the warp threads for each pick of the weft. As a result, where the pattern is large or very varied, the number of cards is correspondingly large. A chain of punched cards in a Jacquard loom frequently reached
2000
in number, and sometimes even exceeded this.
280
Appendix 3
The production of the cards for the composition of different patterns on the Jacquard loom was always a distinct and separate business from the actual weaving. The first step was for the design to be first drawn onto paper on a much bigger scale than that of the finished design. The paper used for this purpose is divided by lines into very small squares: exactly like graph paper.
The Jacquard loom: the ancestor of the computer.
281
Jacquard’s Web
Once the design had been placed onto the ruled paper, the craftsman took a special frame containing more than enough warp threads and cross threads to allow the design to be transferred to it. This often necessitated as many as
500
warp threads and a similar number of cross-threads. The warp threads and cross-threads on the frame amounted to a representation of the ruled paper on a horizontally and vertically threaded frame. A craftsman then proceeded to insert a little piece of thread into all the intersections on the frame that corresponded with the composition of the pattern on the ruled paper.
Each insertion produced just one tiny element of a design which might contain many thousands of such elements. In effect, the craftsmen were constructing a computer graphic on a pixel-by-pixel basis. The French term for this initial mapping work was
mise en carte
: one translation of which is ‘put into a card’. That, of course, is a precise description of the essential nature—and extraordinary ingenuity—of the process.
282
Researching and writing
Jacquard’s Web
has been possible because many people were prepared to help me, even when doing so made heavy inroads into their own schedules.
I am especially grateful to Tim Bergin, editor-in-chief of
Annals of the History of Computing
. Tim read an earlier draft of this book and made many useful suggestions. I am similarly indebted to Professor Martin Campbell-Kelly, head of computer science at the University of Warwick, for his numerous incisive comments.
Inevitably, a book such as this involves taking a considered position on controversial historical issues that remain very much a matter for debate. Neither Tim nor Martin agreed with all of my conclusions, but their suggestions and comments were infused with a spirit of generosity and encouragement which was a delight to receive.
My sincere gratitude to Doron Swade for having read the proofs of this book and making some very useful suggestions.
Alan Woodward, a director of the information technology consultancy Charteris plc and a member of the British Computer Society, also read an earlier draft of the book and provided useful guidance.
David Craig, lecturer in history at Durham University, was extremely helpful in helping me to place the invention of the Jacquard loom in perspective with the progress of the British and French Industrial Revolutions.
My gratitude also to the Lyons historian who has devoted so much time and effort to the task of researching the life of Joseph-Marie Jacquard from original sources and disinterring the truth from legend, half-truth, and falsehood. He asked me not to mention him by name and I have respected his wishes.
283
Jacquard’s Web
I am grateful to the late Georges Mattelon, one of only two dozen remaining silk-weavers still using Jacquard hand-looms in Lyons, for showing me round his workshop in the city and giving me the opportunity to use an original Jacquard loom.
In addition, my thanks are due to Rolf Ziegler, who was such a fascinating guide at the House of IBM in Sindelfingen, near Stuttgart, where he introduced me to the most comprehensive collection of Hollerith punched-card machines in the world.
Karl-Otto Reimers arranged for me to meet Rolf and to see the collection at Sindelfingen. I was privileged to be shown Hollerith machines in operation at the Heinz Nixdorf Museum in Pader-born, Germany, and I am grateful to Michael Mikolajczak for making this possible.
Thanks also to the staff of the Science Museum Library, London, for access to Charles Babbage’s ‘Scribbling Books’ and to his social diary for
1844
; to the staff of the Modern Manuscripts Reading Room at the British Library, London; and to Audrey Giraud and Pascale Le Cacheux, of the Museums of the History of Weaving and Decorative Arts in Lyons, for their unfailing patience and kind assistance.
The most important original source for the lives of Lord Byron, Lady Byron, and Ada Lovelace is the Lovelace-Byron Collection in the Bodleian Library, Oxford. My thanks to Colin Harris, Rob Wilkes, and Nicola Cennan for their assistance with locating material. I also thank the Earl of Lytton for granting me access to this Collection and for permission to reproduce material from it and Lady Lytton for her helpful advice. Thanks are also due to the Earl of Lytton’s literary executors, Laurence Pollinger Limited.
The recently discovered letters from Ada Lovelace to Charles Babbage are held at the Northumberland Record Office.
I am grateful to Sarah Cooley of the Record Office for help in this respect and to the Society of Antiquaries of Newcastle upon Tyne for its permission to view and quote from these precious letters.