Voluntarism and the fruits of collaboration.
Technology and Culture 42 (2001), 710–736.
Examines the early history of the ShARE
user group mentioned in this column.
A material history of bits. Journal of the
American Society for Information Science
and Technology 62, 6 (2011), 1042–1057.
Begins with an interesting critique of the
idea of information as an immaterial digital
substance, as sometimes favored by
media theorists, then surveys key modular
design techniques used to produce the
computer systems that support digital
representations of information.
The history of the history of software. IEEE
Annals of the History of Computing 29, 4
(2007), 40–51. Explores the development of
software history over its first few decades,
identifying trends and key works.
Software in the 1960s as concept, service,
and product. IEEE Annals of the History of
Computing 24, 5 (2002), 5–13. Examines
the origins of the software concept
and reconstructs the challenges and
opportunities early software packages
provided to data processing users.
Hashagen, U., Keil-Slawik, R.,
and Norberg, A.L., Eds.
Mapping the History of Computing:
Software Issues. Springer-Verlag, new
York (2002). Papers and discussion from
leading scholars chosen to reflect different
approaches to software history. Includes
several insightful papers from museum
specialists on the challenges of collecting
and displaying software.
The hard work of software history. RBM:
A Journal of Rare Books, Manuscripts,
and Cultural Heritage 2 (2001), 41–161.
Overview of the challenges of software
history, including those posed to curators.
What makes the history of software hard.
IEEE Annals of the History of Computing
30, 3 (2008), 8–18. A thoughtful attempt
to position software history within an
approach to what Mahoney called the
“histories of computing(s)” structured
around user communities and practices
rather than hardware.
Williams, R. and Pollock, N.
Software and Organizations: The Biography
of the Enterprise-Wide System or How SAP
Conquered the World. Routledge, London,
2009. Applies perspectives from science
studies to enterprise software packages,
thus making a case for examining their
“biographies” to understand them.
Copyright held by owner/author(s).
complement to hardware, already well
known as a colloquial term for computer equipment.b It was sometimes
used to describe everything else the
computer manufacturer bundled with
the computer hardware—perhaps
including services, documentation,
and other intangibles. In that sense
it has its roots in packaging practice.
Programs became software when they
were packaged, and not everything in
the package was code. For much of the
1960s the most commonly accepted
definition of software therefore included only systems programs rather
than applications, which were usually produced or heavily customized
with the organizations where they
were used. For example, a glossy 1962
pullout inserted into Datamation, a
leading computer industry magazine,
promoted Honeywell’s expertise in
software. This was defined as “the automatic programming aids that simplify the task of telling the computer
‘hardware’ to do its job.” According to
Honeywell the “three basic categories
of software,” were assembly systems,
compilers, and operating systems.
When computer manufacturers
eventually began to “unbundle” their
software offerings, that is, charge
separately for them, this was part of a
broader trend toward packaging code
as a good in its own right—literally as
a “ware” for sale. Over the 1970s the
mainframe packaged software industry developed from a curiosity to a significant market. This growth relied on
a legal framework in which the rights
of producers are protected, on the acceptance of banks and investors of
software as a viable business, on the
willingness of accountants to value
packages as assets on a software company’s balance sheet, on the willingness of customers to purchase something that may contain flaws they are
unable to fix, and on the creation of a
set of shared cultural understandings
such as the difference between a bug
fix (free) and an upgrade (usually paid
for). None of these things were initially
obvious, and each involved a process of
b Shapiro, F.R. Origin of the term software:
Evidence from the JSTOR electronic archive.
IEEE Annals of the History of Computing 22,
2 (2000) located an initial usage in 1958 by
mathematician John W. Tukey to describe
automatic programming aids.
collective learning and negotiation between producers and suppliers during
which a variety of practices were experimented with to figure out a viable new
way of packaging software.
This is a column rather than a
book, and there is insufficient space
here to explore the many other chapters in the life of the software package
such as the first high-quality manu-facturer-supported packages (IBM’s
FORTRAN seems to have been a model), commercial software libraries,
the shrink-wrapped model developed
by the personal computer industry,
online app stores, and subscription
services. The shape and size of the
package varied, and the bundle of
code, documentation, services, support, and tacit knowledge assembled
to make an enterprise product like
SAP ERP into a salable commodity
are clearly quite different from those
packaged as Angry Birds. Still, airmail envelopes and modern inter-modal shipping containers are both
packaging technologies functioning
on very different scales.
The point remains that the history
of software is much more than just
the history of code. Despite its apparent immateriality, software has always
been tied to a platform and has always
been physically embodied in one form
or another. What turned programs into
software was the work of packaging
needed to transport them effectively
from one group of users to another. To
understand software we cannot just
look at the bits. We need to examine
the whole package.
Thomas haigh ( firstname.lastname@example.org) is an associate
professor of information studies at the university of
wisconsin, milwaukee, and chair of the SIgCIS group for
historians of computing.
just the history