Beginnings of a revolution - The Computing Universe: A Journey through a Revolution - page 9

this can be done. How can one expect a machine to do all this multitudinous
variety of things? The answer is that we should consider the machine as doing
something quite simple, namely carrying out orders given to it in a standard
form which it is able to understand. 7

This is not the last computing project to underestimate the difficulties asso-
ciated with the “paper work” or, as we would now say, “programming the
machine”!
In 1946, at the instigation of the new dean of the Moore School, Howard
Pender, the Army Ordnance Department, and the U.S. Office of Naval Research
sponsored a summer school on stored-program computing at the Moore School
( Fig. 1.13 ). There were thirty to forty invitation-only participants mainly from
American companies, universities, and government agencies. Alone among the
wartime allies, Britain was invited to participate in the summer school. The
Moore School Lectures on Computing took place over eight weeks in July and
August, and besides Eckert and Mauchly, Aiken and von Neumann made guest
appearances as lecturers. The first part of the course was mainly concerned
with numerical mathematics and details of the ENIAC. It was only near the end
of the course that security clearance was obtained that enabled the instruc-
tors to show the participants some details of the EDVAC design. Wilkes had
received an invitation from Dean Pender and, despite funding and visa prob-
lems, decided it was worth going since he thought he was “not going to lose
very much in consequence of having arrived late.” 8 After attending the last
two weeks of the school, Wilkes had time to visit Harvard and MIT before he
left the United States. At Harvard he saw Howard Aiken's Mark I and II elec-
tromechanical computers, and at MIT he saw a new version of Bush's differ-
ential analyzer. He left the United States more convinced than ever that the
future was not going to follow such “dinosaurs” but instead follow the route
laid out by the EDVAC report for stored-program computers. On his return to
Cambridge in England, Wilkes started a project to build the Electronic Delay
Storage Automatic Calculator - usually shortened to EDSAC, in conscious hom-
age to its EDVAC heritage.
The EDSAC computer became operational in 1949. In these early days of
computing, a major problem was the development of suitable memory devices
to store the binary data. Eckert had had the idea of using tubes filled up with
mercury to store sound waves traveling back and forth to represent the bits of
data, and Wilkes was able to successfully build such mercury delay line memory
for the EDSAC. A variant on Wilkes's design for the EDSAC was developed into
a commercial computer called Lyons Electronic Office, or LEO. It was success-
fully used for running business calculations for the network of Lyon's Corner
Houses and Tea Shops. Wilkes later introduced the idea of microprogramming,
which enabled complicated operations to be implemented in software rather
than hardware. This idea significantly reduced the hardware complexity and
became one of the key principles of computer design.
Meanwhile, back in the United States, Eckert and Mauchly had resigned
from the Moore School after an argument over patent rights with the uni-
versity and were struggling to get funding to build a commercial computer.

Fig. 1.13 The Moore School of Electrical
Engineering at the University of
Pennsylvania, where the ENIAC
was born.