Proprietary Rights in Computer Software: Individual and Policy Issues
Deborah G. Johnson
3. Is Our System of Copyright and Patent Protection for Computer Software Good?
To put the policy issue in a moral or value framework, let me begin by saying that while in earlier work (Johnson, 1985), I toyed with moral arguments supporting the ownership of software, namely a Lockean labor theory argument, I now believe that property rights do not have a moral basis in the sense that they would exist prior to a society of laws. That is, I believe that property rights are social or conventional or artificial. This is entirely consistent with copyright and patent law in this country for both these systems of law are utilitarian in character. Both systems aim to produce good consequences for society in the long run. Debate and discussion about what the law should be with regard to computer software should, then, be framed in utilitarian theory.
In the late 1970s and early 1980s, a good deal of concern was being expressed that neither copyright nor patent law would adequately protect computer software. A sizable literature described the extent and impact of software piracy and illegal copying, and expressed fear that software development would be significantly impeded because software companies would not be able to recover the costs of development, let alone, profit from their creations. The incentive to create would be significantly dampened.
In the late 1980s and early 1990s, more and more concern is being expressed that there is too much protection for computer software, that is, that too much has become proprietary. The concern now is that copyright and patent protection are being extended too far. They now get in the way of software development (Kahin, 1991).
This shift of concern goes to the heart of the aims of our intellectual property laws, for both copyright and patent law aim to create an environment in which invention is encouraged. This is done, on the one hand, by granting ownership in things such that the owner can put the new invention into the marketplace and profit when the invention is useful. On the other hand, invention is facilitated by insuring that the “building blocks” of the technological arts and sciences are not owned. Ownership of the building blocks would interfere with invention insofar as new inventors would have to seek permission to use these building blocks from private owners – owners who could refuse to grant permission (to avoid competition, out of personal whim, or for any reason whatsoever), or drive the price of invention prohibitively high.
When it comes to copyright, you can own the expression of an idea, but not the idea itself; when it comes to patents, you can receive a monopoly on the use of your invention as long as your invention is not, or does not preempt use of, a law of nature, abstract idea, mathematical formula, etc. The rationale for these restrictions is the same in both cases. To grant ownership of ideas, laws of nature, mathematical formulas, etc. would interfere with progress in the technological arts and sciences because others could not freely use these building blocks.
So, both patent and copyright law aim at facilitating invention; that is, both aim at producing good consequences in the technological arts and sciences. In order to do this, both systems of law must draw a very careful line between what can and cannot be owned. However, the line is particularly difficult to draw in the case of computer software for the distinctions traditionally used to draw the line, such as that between idea and expression and that between mathematical formula and application, get very blurry when worked through computer technology.
The shift in concern about patent protection on software from the early 1980s to the early 1990s can be traced to a shift in the policies and practices of the Patent Office and the Court of Customs and Patent Appeals (CCPA) after the Diamond v Diehr case (Samuelson, 1990). Up until Diamond v Diehr the Patent Office had been extremely reluctant to grant patents on computer-related claims, though its reluctance had been challenged by the CCPA. After Diamond v Diehr the Patent Office began granting patents and the CCPA found new reasons to grant more. While only a handful of software related patents had been granted before Diamond v Diehr, several thousand have probably been granted since (Kahin, 1990).
The new concerns about patent protection on software go to the heart of the patent system’s aim, for they suggest that because so much is owned, invention is now being inhibited. The subject matter limitation on what can be patented aims to insure that the building blocks of science and technology should not be owned so that continued development will flourish, yet complaints suggest that the building blocks may now be owned.
The situation is described roughly as follows: Because so many patents have been granted, before putting new software on the market, one must do an extensive and expensive patent search. If overlapping patents are found, licenses must be bought. Even if no overlapping patents are found, there is always the risk of late issuing patents. Patent searches are not guaranteed to identify all potential infringements because the Patent Office has a poor classification system for software. Hence, there is always the risk of lawsuit due to patent infringement. One may invest a great deal in developing a product, invest even more in a patent search, and then find at the last minute that the new product infringes on something already claimed. These factors make software development a risky business and constitute barriers to development of new software. In particular the costs and risks are barriers to small entrepreneurs.
I have argued elsewhere that computer algorithms should not be patentable and these criticisms lend support to that position. Depending on how the term “algorithm” is defined, these criticisms suggest that an even broader subject matter limitation for program-related patent claims should be implemented. Stallman proposes that we pass a law that excludes software from the domain of patents. Samuelson argues that the Patent Office and the CCPA have overextended the meaning of the Supreme Court’s decision in the Diamond v Diehr case.
The situation with regard to copyright is less clear and I am not going to spend as much time on it. Copyright protection has the legal advantage that the Copyright Act was amended in 1980 to explicitly specify that it applies to computer software. On the other hand, the meaning of copyright protection is unclear in the sense that it is not clear what aspect of a piece of software you own. The focus of attention has been on a number of aspects of software which are roughly classified as “look and feel.”
Copyright protection is easier to acquire in the sense that if you develop a new system on your own, even though the new software may duplicate something already copyrighted, you do not infringe as long as you were unfamiliar with the copyrighted program while you were developing your own. This is an advantage from the point of view of acquiring protection, but a disadvantage in that the protection you acquire is weak.
It is becoming increasingly apparent that the idea/expression distinction which is the conceptual heart of copyright law may not be adequate to handle computer software. The uncertainty of the application of copyright law is itself enough to get in the way of development in the field.
Note that these criticisms of patent and copyright protection are not directed at the fundamental character of the laws. The aims and strategies of copyright and patent law seem right on target in seeking to create an environment in which invention can flourish. In this respect they are not bad laws. But while their aims are right, they seem to lack the conceptual tools to handle the issues posed by computer technology. It appears that copyright law and patent law will have to be modified or abandoned for computer-related invention.
Go to: 4. Is Policy Change Called For?
Home > Research Resources > Software Ownership & Intellectual Property Rights > Proprietary Rights in Computer Software: Individual and Policy Issues
HOME |
IN THE NEWS | RESEARCH RESOURCES
TEACHING
RESOURCES | STUDENT RESOURCES | LINKS
The Research Center on Computing & Society
at Southern Connecticut
State University
501 Crescent Street • New Haven, CT 06515
Director: (203) 392-6790 •
e-mail: webmaster@computerethics.org
© 2000 – 2007 – Research Center on Computing & Society