Information Technology Reference
In-Depth Information
was not an unrepresentative example.
34
As Wilkes later recalled in his
memoirs, early on in the life of the EDSAC, its programmers had “begun
to realize that it was not so easy to get a program right as had at one
time appeared.” It was with some shock and dismay that he himself
realized that “a good part of the remainder of my life was going to be
spent in fi nding errors in my own programs.”
35
The tedious process of
identifying and removing these errors, known as “debugging,” was time-
consuming, diffi cult, and intellectually unfulfi lling. As much as one-half
of the budget of a large programming project could be spent on testing
and debugging—activities that were perceived as being low-status and
unpleasant.
36
As will be described in subsequent chapters, improvements in com-
puter hardware along with the development of compilers and other
programming utilities would help alleviate some of the challenges associ-
ated with coding. But as many FORTRAN and COBOL programmers
would soon realize, the dull and mechanical aspects of software develop-
ment did not disappear with the advent of compilers and automatic
programming languages. Nor did the intellectual challenges associated
with analysis and design. Mistakes were inevitable, even from the most
profi cient of programmers. In one widely recited and tragic (and possibly
apocryphal) example, a minor transcription error in control software for
the Mariner I probe to Venus caused the spacecraft to veer off-course
four minutes after takeoff, forcing NASA to destroy it remotely. The
mistake that the programmer allegedly made was to replace the
FORTRAN statement
DO 3 I = 1,3
with
DO 3 I = 1.3
Instead of looping through a series of statements, as the code in the fi rst
version would have required, the latter form was interpreted by the
FORTRAN compiler as the assignment of a variable. That the loss of
the Mariner I could be caused by such a seemingly trivial error high-
lighted for many observers the central importance of employing only the
most skilled programmers.
37
This perception holds true regardless of
whether or not the Mariner I anecdote is factually accurate. During the
late 1950s and 1960s such stories of software-related disaster were a
staple of the popular press, and helped set the state for the emergence
of a full-blown software crisis in the late 1960s.
