====> In view of yesterday's Windows 98 decision this review of
technical and proprietary software issues, by official France's
head geek, offers useful perspective. It originally appeared in
the January '98 issue of Le Monde Diplomatique.
valis
FREEWARE AND THE ISSUE OF INTELLECTUAL PROPERTY
Free software for all
________________________________________________________________________
The Internet and the World Wide Web emerged (in 1969 and 1989 respectively)
as extensions of a publicly available research network (both military and
civilian). Their success is explained largely by a generosity on the part of
the people who created them, and a spirit of sharing. These technologies,
freely accessible at the outset, have been partially taken over by a handful
of large companies. But this process of merger and concentration has not
curbed the creativity and generosity of radical programmers. Often developed
as a sideline to their "real work", a whole galaxy of software has been
created. This form of production sets itself deliberately outside the logic
of the market place. Paradoxically, it is the only force really threatening
the domination of Microsoft, because a clever use of copyright laws has put
it irreversibly in the public domain.
by BERNARD LANG *
________________________________________________________________________
By virtue of their immaterial nature, computer programmes operate
counter to normal commercial logic. Leaving aside development and
conception costs, the marginal costs of their production and
distribution are virtually nil. And the development of the Internet
has meant that this is even more the case. Major software
publishing houses are suddenly faced with the competitive challenge
of micro-firms emerging at the global level (such as the "start-up"
firms in Silicon Valley), and of programmers offering their
creations free to consumers. The companies are attempting to
preserve their revenues and their monopoly grip on the market (1)
by establishing controls over the duplication and standards of
digital material.
This was the area dealt with by the Berne conference of December
1996, which attempted to create a framework of legislation covering
intellectual property rights (2). This area has also generated
intense technological activity geared to inventing procedures for
marking and tracking electronic commodities, and mechanisms capable
of preventing the reproduction of things which are, in their
essence, reproducible. The energy that was previously devoted to
creating those commodities is now being put into finding ways of
preventing their duplication. This is detrimental to the efficient
use of information technology (3) and is threatening the durability
of its contents.
Of course, some would argue that this artificial re- establishment
of scarcity - which one might see as a deliberate and calculated
destruction of resources - is justified by companies' need for
revenue in order to develop new technology, create new products and
improve existing products, as well as by the costs of job creation.
But such arguments should be set alongside the fabulous profits of
the firms in question: in 1997 Microsoft made a profit of $3.5
billion on a turnover of $11.4 billion, and directly employed only
22,300 workers. Furthermore, the extraordinary growth of the
communications and information technology sectors have been driven
by the Internet and the Web, which owe nothing, or very little, to
the market economy (4).
For the software industry, the mechanisms of open competition are
ambiguous. We have a situation in which commercial industrial
software is distributed in forms that are directly usable by a
computer (executable code), but without any of the information
(source code and documentation) which would permit the user to
modify it, adapt it to other machines and uses, make it more
reliable, or correct the ever-present programming errors.
Furthermore, users' licences expressly forbid any such
modifications. This deprives client companies and organisations of
control over the maintenance of their software, its time-durability
and the possibility of adapting it to their individual needs. These
are often crucial factors in the business world, particularly at
the strategic level, when companies are integrating this software
into their products or services.
The world scale of the market for software, the specific properties
of immaterial commodities, and especially the legal or
technological control of "standards" (particularly as regards the
functional interfaces of software and the ways in which information
is represented) are leading ineluctably to a concentration of
monopoly power. Not only are client companies put into a state of
dependency, but they also no longer have alternatives.
Since the suppliers have few competitors, they are even less
motivated to satisfy their clients' specific needs. It is possible
for a whole sector of technology to fall under the control of a
single firm (or a small number of firms). Users in the fields of
education and research are particularly concerned at this
single-sourcing of available technology, and the resulting control
of the flows of information which are so vital to researchers.
The ecology of ideas and technologies obeys the same laws as that
of living beings. Single evolutionary solutions present a number of
dangers. To have a small number of producing companies
correspondingly diminishes the quantity, and especially the
variety, of research and, therefore, also diminishes technological
progress. The competitive element of technological evolution, which
is essential in order to avoid technological dead ends, is either
weakened or disappears. The absence of diversity makes the fabric
of technology more vulnerable to attack, with the threat of
computer viruses being only one danger among many.
A recurrent theme of liberal thinking is that there is no
alternative to the market economy. In the case of software, nothing
could be further from the truth. Another path is already being
traced. One can understand the big firms being nervous about recent
developments, but it is hard to explain the almost total media
black-out, since this is an economic phenomenon which is as massive
as it is new.
This search for a different way of doing things was undertaken in
the early 1980s by Richard Stallman, at that time a researcher at
the Massachusetts Institute of Techology (MIT), and was
subsequently embodied in the creation of the Free Software
Foundation (5) and a number of associated companies. The initial
intention was to create free software ("freeware") which, like
ideas, would be available to all, in line with the philosophy of
Pasteur, Jefferson et al. In order to avoid people laying economic
claim to this free software, Stallman turned the notion of
copyright on its head by popularising a new kind of licence, known
as the "general public licence", which protects a given piece of
software from technical or legal attempts to restrict its
utilisation, diffusion and modification (6).
In tandem with the spread of such licences, there has been a
sizeable and varied production of free software. The necessary
specifications and background information have been made available,
so that people can adapt or improve the software as they see fit,
and redistribute it, with or without payment, and without any
control over this redistribution by third parties. True to the
tenets of economic liberalism, this free competition has had an
extremely positive effect on the quantity and quality of the
software being produced. But the influence of the money economy is
much reduced.
The most visible product of this freeware culture is an operating
system - the software necessary to the functioning of every
computer, providing a basic set of operative functions (file
handling, posting, text capture, connection to networks etc.) -
known as Linux. This was developed initially in 1991 on the basis
of work done by a Finnish student, Linus Torvalds. Since then it
has grown, benefiting from cutting-edge contributions from a
supportive army of experts worldwide, linked via the Internet. The
development of Linux has been self-organising, like a huge
enterprise without walls, without shareholders, without wages,
without advertising and without revenue. To date the number of
Linux installations is estimated at between five and six million,
with increasing evidence of applications in industry, too. The
system's market share compares favourably with Apple's, but its
growth rate is higher.
Various studies have shown that this software is in all respects
competitive with commercial products. This is also confirmed by the
extent of its penetration and infiltration in economic activity.
The most significant example is undoubtedly the Internet, which
would disappear entirely if this software were eliminated (7).
Technological and economic dependence on supplier companies is
eliminated, or at least greatly diminished. The durability of the
products, their evolution and adaptability, as well as the help
available to users, are better guaranteed by the presence, activity
and stability of a large mass of users and programmers than by the
unforeseeable strategies of the big software publishing houses. The
free availability of all the elements involved in the product's
development makes it possible to buy in any guarantees and
complementary services that may be required.
The contradictions of liberal economics
In technological terms, freeware is a credible and tested solution.
Furthermore, it encourages new areas of economic activity, and thus
creates jobs. It does this by developing a range of services, by
encouraging the development of complementary or competitive
commercial products, and above all by fertilising technology-based
businesses by providing resources that are free, independent,
tested and of high quality (8). The development of freeware tends
to replace a commercial publishing activity (which was centralised
and monopolistic, and the protection of which stifled economic and
technological development in the short term, creating relatively
few jobs) with a commercially-based servicing sector, which creates
more jobs (often decentralised), and which is more competitive.
Furthermore, given its greater technological fluidity, it
encourages the creation of new companies. In the South, the
availability of cost-free and freely available resources opens the
way for independence in the furtherance of technological
development.
The development of software is similar in nature to the development
of mathematical theory (9). As we know, science in general and
mathematics in particular sit very uncomfortably with the secrecy
and barriers which are the daily bread of commercial industrial
development. Good specifications (definitions) and reliable
products (explanations, demonstrations) are created only slowly,
and this happens through an open social process involving
evaluation, comparison and collaboration. It is not at all
surprising that Linux, although the younger product, is a better
system than Windows NT, the leading software product of the biggest
software publishing house on earth. At this point, it would be fair
to ask whether the classic industrial environment is the most
appropriate for developing the technologies of the immaterial
sector.
Consumer goods (films, music, novels) and production goods
(software, scientific articles etc.) have very different economic
and social roles. It is absurd to subject them to identical
legislation and protective mechanisms. Economic liberalism is
caught in a contradiction. It justifies weakening the sovereignty
of individual states and the elimination of all kinds of barriers,
including those designed to protect individuals, by the need for
much greater economic fluidity. But, at the same time, it operates
- via the abuse of copyrights and patents, non-respect of
industrial standards, proprietary control of interfaces, industrial
secrecy and the creation of monopolies - to create barriers that
are extremely harmful to economic and technological progress and to
the creation of useful jobs.
______________________________________________________________
* Research director at the Institut national de la recherche en
informatique et en automatique (INRIA).
(1) See Ralph Nader and James Love, "Microsoft, monopole du
prochain siècle?", (available in English, "What to do about
Microsoft?"), Le Monde diplomatique, November 1997.
(2) Pamela Samuelson, "On authors' rights in Cyberspace", First
Monday, Copenhagen, vol. 1, no. 4, October 1996.
http://www.firstmonday.dk. See also Philipe Quéau, "Offensive
insidieuse contre le droit du public à l'information", (available
in English, "Covert campaign against freedom of information"), Le
Monde diplomatique, February 1997.
(3) Richard Stallman, "Copywrong", Wired, San Francisco,
July-August 1993. http://www.wired.com/wired/1.3/
(4) Hervé Le Crosnier, "L'économie de l'information dans le
contexte des nouvelles technologies", Study Day on "Public domain
information in the light of the Internet and digital technology",
Paris, 18 June 1997. http://www.info.unicaen.fr/Herve/pub97/juin/
(5) http://www.gnu.org/fsf/
(6) FSF: GNU General Public Licence (GPL).
http://www.gnu.ai.mit.edu/copyleft/gpl.html
(7) Keith W. Porterfield, "Information Wants to be Valuable",
NetAction Notes, no. 26, 3 September 1997, San Francisco,
California. http://www.netaction.org/articles/freesoft.html
(8) Freeware is used in a number of applications where reliability
is a critical factor: the control of experiments launched aboard
the space shuttle (NASA), industrial robotics (Lectra-Systèmes,
France), operational control of passenger lifts (Fujitec, Japan),
command systems in the US military, fuel distribution
(Schlumberger, USA), etc. The magazine Linux Journal, published
from Seattle, Washington, gives an account of these industrial
applications. http://www.linuxjournal.com
(9) William A. Howard, "The formulae-as-types notion of
construction", in "To Haskell Brooks Curry, Essays on Combinatory
Logic, Lamda Calculus and Formalism", Jonathan Paul Seldin and
James Roger Hindley, Academic Press, 1980.
Translated by Ed Emery
