[ This article originally appeared on
Nature’s web site. The article is part of a larger series on
Nature, and the introduction
to the series provides some useful background on the issues Mr.
Stallman addresses. -ed. ]
It should be a truism that the scientific literature exists to
disseminate scientific knowledge, and that scientific journals
exist to facilitate the process. It therefore follows that rules
for use of the scientific literature should be designed to help
achieve that goal.
The rules we have now, known as copyright, were established in
the age of the printing press, an inherently centralized method of
mass-production copying. In a print environment, copyright on
journal articles restricted only journal publishers – requiring
them to obtain permission to publish an article – and would-be
plagiarists. It helped journals to operate and disseminate
knowledge, without interfering with the useful work of scientists
or students, either as writers or readers of articles. These rules
fit that system well.
The modern technology for scientific publishing, however, is the
World Wide Web. What rules would best ensure the maximum
dissemination of scientific articles, and knowledge, on the Web?
Articles should be distributed in non-proprietary formats, with
open access for all. And everyone should have the right to `mirror’
articles; that is, to republish them verbatim with proper
attribution.
These rules should apply to past as well as future articles,
when they are distributed in electronic form. But there is no
crucial need to change the present copyright system as it applies
to paper publication of journals because the problem is not in that
domain.
Unfortunately, it seems that not everyone agrees with the
truisms that began this article. Many journal publishers appear to
believe that the purpose of scientific literature is to enable them
to publish journals so as to collect subscriptions from scientists
and students. Such thinking is known as `confusion of the means
with the ends’. Their approach has been to restrict access even to
read the scientific literature to those who can and will pay for
it.
They use copyright law, which is still in force despite its
inappropriateness for computer networks, as an excuse to stop
scientists from choosing new rules.
For the sake of scientific cooperation and humanity’s future, we
must reject that approach at its root – not merely the obstructive
systems that have been instituted, but the mistaken priorities that
inspired them.
Journal publishers sometimes claim that on-line access requires
expensive high-powered server machines, and that they must charge
access fees to pay for these servers. This `problem’ is a
consequence of its own `solution’. Give everyone the freedom to
mirror, and libraries around the world will set up mirror sites to
meet the demand. This decentralized solution will reduce network
bandwidth needs and provide faster access, all the while protecting
the scholarly record against accidental loss.
Publishers also argue that paying the editors requires charging
for access. Let us accept the assumption that editors must be paid;
this tail need not wag the dog. The cost of editing for a typical
paper is between 1% and 3% of the cost of funding the research to
produce it. Such a small percentage of the cost can hardly justify
obstructing the use of the results. Instead, the cost of editing
could be recovered, for example, through page charges to the
authors, who can pass these on to the research sponsors. The
sponsors should not mind, given that they currently pay for
publication in a more cumbersome way through overhead fees for the
university library’s subscription to the journal. By changing the
economic model to charge editing costs to the research sponsors, we
can eliminate the apparent need to restrict access. The occasional
author who is not affiliated with an institution or company, and
who has no research sponsor, could be exempted from page charges,
with costs levied on institution-based authors.
Another justification for access fees to online publications is
to fund conversion of the print archives of a journal into on-line
form. That work needs to be done, but we should seek alternative
ways of funding it that do not involve obstructing access to the
result.
The work itself will not be any more difficult, or cost any
more. It is self-defeating to digitize the archives and waste the
results by restricting access. The US Constitution says that
copyright exists “to promote the progress of science”. When
copyright impedes the progress of science, science must push
copyright out of the way.
Richard Stallman is the founder of the GNU project, launched
in 1984 to develop the free operating system GNU (an acronym for
`GNU’s Not Unix’), and thereby give computer users the freedom that
most of them have lost. GNU is free software: everyone is free to
copy it and redistribute it, as well as to make changes either
large or small. The GNU/Linux system, combining the GNU system and
the Linux kernel, has an estimated 17 to 20 million users. Stallman
was awarded a MacArthur Foundation fellowship in 1990.
Copyright 2001 Richard Stallman.
Verbatim copying and redistribution of this entire article are
permitted in any medium provided the copyright notice and this
notice are preserved.