Chris Lale
This article highlights some consequences of proposed software patenting in Europe. It uses an analogy with musical harmony to illustrate some of the issues. The article explores the European Commission's proposals to allow fundamental truths, such as algorithms, to be considered as inventions rather than truths.
Revision 1.0, 2nd January 2003, initial draft, Chris Lale.
Revision 1.1, 3rd January 2003, some additions, corrections and new sections added, Chris Lale.
Revision 1.2, 12th December 2007, updated links, Chris Lale.
You can find the latest version of this document at http://www.untrammelled.co.uk.
Copyright © 2003 Chris Lale (chrislale AT untrammelled DOT co DOT uk ). Permission is granted to copy, distribute and/or modify this document with no Invariant Sections, with no Front-Cover texts and with no Back-Cover Texts under the terms of the GNU Free Documentation License, version 1.1 or any later version, published by the Free Software Foundation. A copy of the license can be found at http://www.fsf.org/copyleft/fdl.html .
"Copyright gives the creators of a wide range of material ... economic rights enabling them to control use of their material in a number of ways, such as by making copies, issuing copies to the public... However, copyright does not protect ideas..." [1]
Computer programs can be protected by copyright. However, the European Commission now proposes to allow patenting of the technical ideas and principles behind a program [2]. This raises the issue of whether to allow the patenting of fundamental truths. At present, Article 52 of the "European patent convention" does not allow this. It states that "discoveries, scientific theories and mathematical methods ... shall not be regarded as inventions [3].
"A patent gives an inventor the right for a limited period to stop others from making, using or selling an invention without the permission of the inventor." [4]
There are four basic requirements for patentability:
The requirement for "invention" is particularly relevant.
The rest of this article explores the issues raised by the propsal to allow patenting software. It does this by means of a musical analogy. In the analogy, music is the equivalent of software and musical instruments are the equivalent of hardware. Chord progressions are the analogue of algorithms. Of course, both music and computer software may be protected by copyright. This article examines only the patenting issue.
Some of the analogy's examples relate to lengths of time in excess of twenty years (the lifetime of a typical patent). However, the pace of development of modern software is far in excess of the pace of development of western music. For the purposes of this illustration, it does not seem unreasonable to allow twenty years in the development of software to be analogous to a much greater period in the development of music.
Software (eg a computer program) is written in code. The code written by the programmer is called the source code. The source code may be translated into other forms so that it can be understood by a piece of hardware (eg a computer).
The same thing happens in music. Musical notation, the equivalent of source code, consists of notes (crotchets, quavers etc) written on a stave of five parallel lines. Musical instruments are the equivalent of computer hardware. Sometimes, musical notation is translated into other forms (eg punched cards for mechanical pianolas and steam organs, pins on a revolving drum for musical boxes, etc).
A programmer does not write random code. Code is structured and makes use of algorithms. An algorithm is a method or a series of logical steps used to solve a problem.
Algorithms form a body of knowledge, built up over years, available to all programmers. This is similar to the way in which scientific laws are available to all scientists. Scientific laws and mathematical theorems are fundamental truths. So are algorithms. Laws and theorems often depend on other laws or theorems. An algorithm often depends on other algorithms. It is doubtful that science or mathematics could have progressed to the levels they have today if anyone had been able to patent these laws or theorems. Fortunately, it is difficult to imagine anyone ever having held a patent on, for example, Archimedes' principle or Pythagoras's theorem.
A composer wishing to write music with harmonies uses chord progressions. Chord progressions are the equivalent of a computer program's algorithms. Suppose that a composer of nursery rhymes wishes to end a nursery rhyme with an air of finality. The composer may choose a chord progression called the "perfect cadence". (The perfect cadence chord progression is "V, I" meaning "the chord of 5 followed by the chord of 1".) For example, the last two chords in "Three blind mice" implement a perfect cadence.
If the person who wrote the very first nursery rhyme had been able to patent the perfect cadence, perhaps "Three blind mice" may never have been written or performed. Possibly, only a few simple nursery rhymes may ever have been composed. Fortunately, the development of music has not been held back by patents.
Software patents could have a detrimental effect on the quantity and quality of new software, and its speed of development.
The "cadential six four" (Ic, V, I) is a development of the perfect cadence (V, I). The cadential six four is easily recognisable in the last three chords of “God save the queen”. The composer who developed the cadential six four would still require permission from the patent holder of the perfect cadence. Just imagine that every composer who writes a new arrangement of the British national anthem were obliged to obtain permission from some (possibly foreign) commercial enterprise. It would be a national scandal. No doubt, adding insult to injury, money would be demanded too, even though the patent holder had taken no part in the development of the new arrangement.
Software patents could easily inhibit the development of software, resulting in similar unfair practices.
Of course, it may be argued that simple algorithms would not be patented. Patents might only be granted for complex, specialised algorithms. Even so, software development could be held back or prevented for just the same reasons.
Haydn wrote a piece of music called the "St Anthony chorale". Suppose that Haydn's rich and powerful employer (Prince Esterházy) had been granted a patent for the chord progressions (algorithms) in this work. Suppose also that Haydn subsequently moved to a new employer. Haydn could have been prevented from correcting, modifying or developing his own work. Other composers could have been prevented from using Haydn's piece as a source of inspiration. In fact, although this was years later, Brahms wrote a famous work called "Variations on a theme of Haydn" which was based on the "St Anthony chorale".
Software patents could reduce the production of better, more reliable software. It could prevent timely elimination of bugs, correction of mistakes, improvements and replacements for software.
The proposed patenting of algorithms might not only stifle software development. It might also place a lot of money and power in the hands of those who are already rich and powerful. Large companies and corporations that have done little or no work on the original algorithm or software might acquire huge sums. They might also be able to reduce competition, and even eliminate it.
Western church music uses the plagal cadence. The plagal cadence is the progression of two chords (IV, I). It harmonises the “Amen” traditionally sung in church services and at the end of hymns. That might seem to be uncontroversial since composers of church music would be unlikely to be sued by their own Church. (Let us avoid the complication of disagreements between rivalreligious sects.) However, the patent might still inhibit the development of secular (non-church) music.
Imagine that the Church had been granted a patent for the plagal cadence. The chord progression (algorithm) for a basic blues is I, I, I, I, IV, IV, I, I, V, IV, I, I. You can see the plagal cadence (IV, I) appearing twice here. The Church might decide not to allow the development of the plagal cadence into the twelve-bar blues. Think of the thousands of pieces of blues, jazz, rock and pop that we might never have heard. The theme music of the old television series "Batman" may never have reached our screens! Of course, the Church might have allowed selectively some “morally correct” blues music in order to generate an income. Even so, it is hard to imagine that we would have had even a fraction of the rich heritage we enjoy today.
Patents are most likely to be granted to (and enforced by) rich and powerful entities. This threatens to reduce or eliminate the competition and creativity that drives development forwards. The threat would be greatest for economies with significant number of small businesses (SMEs). Allowing a patent on a fundamental truth such as an algorithm is reminiscent of the “Big Brother” suppression in George Orwell's 1984.
Musical instruments were invented, not discovered. Arguably, chord progressions were discovered rather than invented. European musicians discovered them by playing and listening to chords, They liked the effects of certain progressions and used them in their compositions. Chord progressions are the fundamental truths that underlie musical harmony. Fortunately, they were never patented in Europe, or anywhere else.
Similarly, Archimedes' principle was discovered, not invented. Of course, many inventions based upon Archimedes' principle must have been patented, but not the principle itself. Anything that floats relies on the underlying fundamental truth discovered by Archimedes. Fortunately, it was not possible to patent Archimedes' principle either.
Can we now afford to allow algorithms to be patented in Europe? Is a thriving, vibrant European software industry under threat?
1. Explanation of copyright: http://www.intellectual-property.gov.uk/std/faq/copyright/what.htm
2. European Commission proposals: http://europa.eu.int/comm/internal_market/en/indprop/comp/index.htm
3. Patentable inventions: http://www.european-patent-office.org/legal/epc/e/ar52.html#A52_2
4. Explanation of patent: http://www.intellectual-property.gov.uk/std/faq/patents/what.htm