It is not strictly necessary to this argument to accept 1977 and 2003 as the two watershed years
in order to understand that early on, software practice emerged into an era of scientific verification
of its results. And later computer science itself became an alibi for the obvious damage
caused by the software professional.

At the first watershed the desirable effects of new scientific
discoveries were easily measured and verified. Manual labour and the
movement of physical objects were no longer necessary to conduct simple
information-based activities. Activities such as record-keeping,
type-setting, and later mapping and messaging, could be conducted
without expensive movement of physical materials or specialised
distribution networks. Companies no longer retained vast typing pools or
armies of filing clerks performing unrewarding rote work. Human
knowledge could be shared freely on open networks. Governments and
corporate publishers lost their monopoly on the distribution of
information. Political dissidents could no longer be kept silent by
physical isolation, or kept under observation by crude wire-tapping.
Arts and entertainments did not need to be realised by specially crafted
fixed-function devices, and could be made more interactive and
stimulating. The positive contribution of modern computer science to
individual utility during that time can hardly be questioned.

But then computer science began to approach the second watershed. Every
year it reported a new breakthrough. Practitioners of new specialties
evangelised their new techniques which provided benefits in a few cases.

The practice of programming became centered on the performance of
professional staffs. Trust in "the next big thing" obliterated good
sense and traditional wisdom on software and its construction. The
irresponsible use of abstraction spread from professionals to the
general public. The second watershed was approached when the marginal
utility of further software declined, at least insofar as it can be
expressed in terms of the satisfaction of the largest number of people.

The second watershed was superseded when the marginal disutility
increased as further monopoly by the software technology establishment
became an indicator of more suffering for larger numbers of people.
After the passage of this second watershed, computer science still
claimed continued progress, as measured by the new landmarks
technologists set for themselves and then reached: both predictable
discoveries and costs. For instance, a few human endeavours faced
problems large enough to justify the creation and application of
supercomputing and "big data" techniques. On the other hand, the total
social cost exacted by big technology ceased to be measurable in
conventional terms. Society can have no quantitative standards by which
to add up the negative value of illusion, social control, drudgery,
isolation, impotence and frustration produced by all-pervading software.

The characteristic reaction to the growing frustration was further
technological and bureaucratic escalation. While evidence shows that
more of the same leads to utter defeat, nothing less than more and more
seems worthwhile in a society infected by the growth mania. The
desperate plea is not only for more software and more computers, but
also for more technology and more research. It has become fashionable to
say that where computer science and technology have created problems, it
is only more scientific understanding and better technology that can
carry us past them. The cure for bad management is more management. The
cure for traffic jams is more roads. The attempt to overwhelm present
problems by the introduction of more science is the ultimate attempt to
solve a crisis by escalation.

[Based on passages from Illich's "Tools for Conviviality".]