Few people outside technical or computer-related fields really know what software is and how it is generated. According to Webster¹s dictionary, it is «something used or associated with and usually contrasted with hardware, the entire set of programs, procedures, and related documentation associated with a system and especially a computer system; specifically: computer programs». The same dictionary defines "program" as «a sequence of coded instructions that can be inserted into a mechanism (as a computer)».
Programmers write their programs in specific languages, known as programming languages; these are standardized ways to instruct a computer. Such languages specify how the computers should behave in various cases and in reply to external events using wording and syntax similar to human languages.
While programmers create their programs using an abstract language, the internal design of the computer is based on binary logic: simple switches that can be on or off. The program, therefore, can't be directly executed by the computer and it must be translated through a process called compilation. Compilation turns the program as written and understandable by people, known as source code into executable code, the set of instructions that can be run directly by the computer. The difference between source code and executable code is a very important one, similar to the difference between a musical score and its live performance: the composer creates and writes the instructions (the score) for a set of components (the orchestra) in order to have a result (the symphony). In the software field, the composer is the programmer, the score is source code; the orchestra acts as a compiler that translates the score into the symphony so that the public can enjoy it. In this context, the executable is the work in its usable form, the live performance generated by the orchestra.
The analogy is valid, within certain limits, when the process is reversed, or rather when from the music played (the executable) one goes back to the score (source): in the case of simple, one-instrument music it is relatively easy to understand the score, much less so in the case of an entire orchestra with dozens of instruments. For software, the process of going backwards from the executable to the source is never a simple matter.
The patent is an instrument created to stimulate development in scientific and practical arts as set out by various national constitutions [1] [2] [3]. Therefore every modification of the law in force must be justified in that sense, verifying a priori if the proposed modification helps development of the sector in question.
In the aim to foster innovation, a core item of patent law in all national systems is disclosure of the «inventive teaching» [4]. In other words, the inventor must describe how to achieve the results about which exclusive exploitation is requested, so that the state of the art can advance.
It's no accident that the European Patent Convention forbids patents on business methods, mathematical theorems, computer programs and other abstract fields [5]; the same rules exist in the Italian laws [6] and we are sure the same applies to most other countries. It's also worth noting how Thomas Jefferson [7] expressed his opinion against the privatization of ideas.
It's important to keep in mind how computer programs fall under copyright regulations, as ratified by all major international agreements: the Bern Convention [8], the TRIPS convention (Trade-Related aspects of Intellectual Property rights) [9], and the WIPO Copyright agreement [10]. On the basis of these legal milestones, every argument on the necessity of protecting software is unfounded. No invention can be covered by both regulations, that is by patent and copyright at the same time.
The concept of protecting inventors lies in the fact that developing an invention requires expensive investments, considering too that not all inventions result in a useable commercial product. The 20-year monopoly on a successful invention also helps to cover research costs that don't result in a product.
This situation has nothing to do with the field of abstract ideas; there are no concrete research costs in developing an idea, so there is no reason to give an exclusive on the use of the presumed invention that would be created in any case.
Anyone who works in a technological field constantly produces new ideas or new programs for computers and often the same procedure is made simultaneously by different groups. Allowing the exclusive use of these creations by one inventor means preventing the independent activity of all the others. In a system where ideas can be patented, those who work in technology find themselves faced with a series of roadblocks. Not even the patent holders are protected, since they will come up against many other patents as soon as they try to produce the product.
It's important to remember that obtaining a patent isn't easy, many small and medium-sized companies simply won¹t be able to take advantage of this possibility and they will have to work in a minefield of patents from their competitors.
Another fundamental characteristic of the patent system is the limited time limit on the guaranteed monopoly of the inventor. The limit is set not to block technical developments but to ensure the inventor a period to profit from the invention in an exclusive fashion and recuperate research investments.
While guaranteeing a 20-year monopoly on a mechanical invention makes sense, that period of time doesn't correlate at all with the productive cycle of a specific software implementation or an abstract idea, which both last two or three years at most. A patent that covers a much longer period of time can't but limit the overall development in the sector and therefore harm those who operate in the field with the exception of those few who can get sufficient patent coverage to not be squashed by those with a bigger budget.
Although current laws prohibit so-called «software» patents, several thousands of such patents have been granted in Europe, following the US established practice of granting them.
«Software» patents would be forbidden in the USA like they are in Europe, but the courts, which have considerable weight in the US legal system, have turned the rule upside down.
Abstract patents in the USA are usually filed by big companies (like IBM, Apple, Microsoft, ...) that use them to trade with other companies, but they are also filed by juridical entities created just to file patents. These are the so-called «litigation companies», whose main activity is collecting royalties on their own patents, without producing nor inventing anything.
As you may expect, life is not easy for those who are working with technology and haven't yet been acquired by a larger company. It's no surprise that small and medium enterprises are almost missing in the United States.
A number of organisations and authoritative people have denounced these problems, but the legal system has no interest in addressing the issue. Take a look, for example, at the position published by the «League for Programming Freedom» in 1991 [11], or the letter by Donald Knuth, sent in 1995 [12]. References to more recent positions appear below.
The US horror gallery is full of patents that damage market and society [13]. For example, «one-click purchasing» (trivial application of existing web tools) granted to Amazon; a geometrical algorithm to unroll panoramic images (implemented by one of us in two hours before we even knew of the patent); the patent on hypertext link (the one used in all web pages) granted to British Telecom.
In Europe, as suggested earlier, «software» patents are forbidden. However, on October 7th 1997, Mario Monti (European commissioner for the free market) suggested to evaluate introducing legal support for absrtract patents. The proposal is justified by the need «to unify the Europen market to the US one», with the aim of helping the European market.
The European Commission has even funded a study on this topic, but instead of mandating the work to a group of experts in macroeconomic matters, they gave it to the «Intellectual Property Institute» in London. It's clear how such an institute can't show a scientific and unbiased approach to the subject matter. The conclusions of the study [14] claim that «development of US economy had a benefit in software patentatbility», and that «our small and medium entreprises don't show an interest in exploitation of the patent system, but might well change their minds». The only thing that is demonstrated, though, is that currenently such companies side against patentability.
But the trend towards software patentability shown by of the EC directives are a menace to free competition and to small and medium entrerprises. If, on the other hand, software patents aren't introduced, the European Union will benefit from favorable conditions to its economy and to competition among actors, conditions less prone to monopolistic practices hidden behind legal actions that would protect inventions.
The European Commission asked for comments on this matter [15]. The results [16] of the poll show that most of the technical and productive world sides against abstract patents, while replies for patentability come from legal institutions and big corporations that already own patents in the United States. It's especially worth noting how 90% of the small and medium enterprises claimed to be against patents. But the analysis of such replies is covertly but largely biased, as it talks about «economical weight» of the replies, and interpretation of the data is fuzzy, to say the least. [17].
Meanwhile, despite current laws still forbid granting patents on abstract concepts, the European Patent Office already approved more than twenty thousand such patents, getting as far as bending the rules, giving directions to the examiners that go undoubtedly against current laws [18].
The «horror gallery» of European patents [19] already shows a sample of what we are going to face. It goes from a patent that encumber the JPEG graphic format to automtic diagnosis (any diagnosis), from the idea of comparing the student's pronounciation and the teacher's one to automatic resizing of computer window when it is partially obsured by another window. As well as conversion of name encoding from one convention to another.
Problems raised by Patentability of abstract ideas aren't limited to what is and isn't right, there are also serious practical issues. Such problems are recognized by patent supporters as well, although they don't agree with their being inherently unsolvable.
Every abstract concept can be expressed in patentable form by writing it as a «computer program». Therefore, no patent office can evaluate the state of the art of «computer programs», as such art covers everyyhing mankind ever tought.
As a result, most granted patents cover realizations that are obsolete in the very day they are filed. Obviously, once the patents is granted, nobody can act in that segment of knowledge without paying or being brought to a court.
Even Gregory Aharonian, strong supporter of software patents, underlined in 1994 how you can't avoid to grant a huge number of patents on trivial ideas [20].
Another unsolvable problem is about how to evaluate the «inventive step» that is mandatory in order for a patent to be granted. Most «software» patents, actually, include no inventive step at all, as shown for example by Stallman [21].
Moreover, as said initially, the «inventive teaching» about the patent must be unveiled. But this requirement is often circumvented when patents on abstract ideas are concerned, as the unveiling is just an approximate description of the problem rather than of the solution to it. This happens despite turning the abstract idea into a practical realization is the most difficult part of the inventive work.
Actually, most examples in the various horror galleries are about patents on the problem and not on the solution to that very problem.
With a situation like the one described above, why is there so much effort to patent software? Isn't software already protected by copyright, and no other area of human production simultaneously falls under the domain of both copyright and patents?
CThose for patenting software generally use two arguments: the «protection of the poor inventor» and the «unification of the international market».
Unfortunately, neither of these arguments is tenable. The "poor" inventor after having invested in legal fees to obtain a patent, must also spend more money on legal fees to defend him or herself from violations of other patents. If each person fences in his or her square meter of land the result is that no one is free to move -- and the only ones to profit are those who sell fences.
The unification of the market is unsustainable as well, as soon as one looks at the situation in the US, where smaller companies often become suffocated or bought out because of presumed violations of patents (examples are part of the aforementioned horror gallery).
It's interesting to note, then, how those in favor of patents on ideas are always those who have a personal interest in the matter: namely managers in patent offices, specialized law practices or companies with a considerable portfolio of patents.
Of course, there are many macro-economic studies on the problem -- all invariably demonstrate how extending the system of patents is harmful to the development of the market and to technology [22].
Fritz Machlup argued back in 1958 that the patent system doesn't bring any advantages to the market it covers [23]. Since then numerous other studies have been done, including the one in 2000 by Bessen and Maskin of the Economics Department of MIT which demonstrated the harmfulness of a patent system in dynamic economic environments [24].
Clearly, studies done by law offices and any kind of «organization to protect intellectual property» should not be taken into consideration for their inherent bias in the matter.
Instituting patents for software is harmful to small and medium-sized companies. The current push in this direction comes from those who have a direct personal interest in the matter (patent offices, law practices, large companies).
Any useful computer program violates dozens of software patents already valid in the US and that would be recognized here as well, so the myth of protecting the single inventor is in fact only a myth.
A number of independent economic studies back up this thesis, while those carried out by opponents are invariably biased by interests.
Until now, small and medium-sized companies have been against patenting abstract ideas, but it's necessary to pay attention to «informative» conferences on the patent problem, keeping in mind the position of the person speaking about it.
The current position favors competitivity of Europe in respect to the US and Japan. There is no reason to change it, to avoid the practice of monopoly and unfair competition in a field as important as new information technologies.
1: «The Congress shall have power to [...] promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries» In: The US Constitution:
2: Article 14 of the German Constitution, according to the Bundestag is not compatible with patenting ideas.
3: The Italian constitution doesn't cover the topic, while patent lasw (CC 2584-2594, law 1127/39, law 244/40, las 360/1994) only describe how patents are implemented.
4: «The European patent application must disclose the invention in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art». European Patent Convention, Art. 83
5: «2. The following in particular shall not be regarded as inventions within the meaning of paragraph 1: [...] (c) schemes, rules and methods for performing mental acts, playing games or doing business, and programs for computers». European Patent Convention, Art. 52.2c
6: Decree 1127/1939 and later changes, in the second paragraph of article 12: «According to the previous paragraph, the following shall not be considered inventions: [...] b) plans, principles and methods for intellectual activities, for games or for business activities, and computer programs» (originally: «Non sono considerate come invenzioni ai sensi del precedente comma in particolare: [...] b) i piani, i principi ed i metodi per attività intellettuali, per gioco o per attività commerciali e i programmi per elaboratori»).
7: «If nature has made any one thing less susceptible than all others of exclusive property, it is the action of the thinking power called an idea» Cited for example in «The Economy of Ideas: A Framework for Patents and Copyrights in the Digital Age (Everything you know about intellectual property is wrong)», di John Perry Barlow (http://www.eff.org/cafe/barlow.html).
8:Convention de Berne pour la protection des oeuvres litteraires et artistiques http://www.law.cornell.edu/treaties/berne/overview.html
9: Art. 10 - Computer Programs and Compilations of Data: «1. Computer programs, whether in source or object code, shall be protected as literary works under the Berne Convention (1971)». The original text can be retrieved from http://www.wto.org/english/tratop_e/trips_e/t_agm3_e.htm
10:Article 4 - Computer Programs: «Computer programs are protected as literary works within the meaning of Article 2 of the Berne Convention. Such protection applies to computer programs, whatever may be the mode or form of their expression». http://www.wipo.org/eng/diplconf/distrib/94dc.htm
11: «Against Software Patents», published by the League for Programming Freedom (http://lpf.ai.mit.edu/Patents/against-software-patents.html) and «Why Patents Are Bad for Software», by Garfinkel, Stallman e Kapor (http://lpf.ai.mit.edu/Links/prep.ai.mit.edu/issues.article).
12: Donald Knuth is one of the most important computer scientists. The letter he sent to the US Patent Office in February 1994 shows how software patents are a hard to most people working with computers, to the sole advantage of the legal system and a small set of inventors. His letter is available from http://lpf.ai.mit.edu/Patents/knuth-to-pto.txt.
13: The US patent horror gallery is at http://lpf.ai.mit.edu/Patents/.
14: The study is available as a PDF document (http://europa.eu.int/comm/internal_market/en/indprop/study.pdf) and a summary is available from http://europa.eu.int/comm/internal_market/en/indprop/studyintro.htm).
15: The consultation has been opened on Oct 19 2000, it is available as a PDF document (http://europa.eu.int/comm/internal_market/en/indprop/soften.pdf) and a summary is available from http://europa.eu.int/comm/internal_market/en/indprop/softpaten.htm.
16: The replies, published in July 2001, are available from http://europa.eu.int/comm/internal_market/en/indprop/softreplies.htm, while their analysis is in a PDF file: http://europa.eu.int/comm/internal_market/en/indprop/softanalyse.pdf
17: For example, on page 14 of the analysis it underlines how the percentage of small and medium enterprises for and against software patents are equivalent (15%). The real information, however, is that 90% of SME's side against software patents; their representing 15% of collected replies is irrelevant.
18: See the press release by EuroLinux: http://petition.eurolinux.org/pr/pr14.html.
19: The European horror gallery collected by FFII includes both examples of badly designed patents (http://swpat.ffii.org/vreji/pikta/mupli/index.en.html) and example of delelopment projects blocked by patents (http://swpat.ffii.org/vreji/pikta/index.en.html).
20: «In short, we have strong exponential growth in the number of software patents, and everything that derives from these patents, including lawsuits. Now it can probably be shown that the number of unobvious, novel, unpublished news ideas in the software world grows linearly at best.». In http://www.eff.org/Intellectual_property/crisis_softpatent.article.
21: «The Anatomy of a Trivial Patent», published by the LPF: http://lpf.ai.mit.edu/Patents/anatomy-trivial-patent.txt.
22: The FFII has made a number of these studies available at: http://swpat.ffii.org/vreji/minra/siskuen.html.
23: Fritz Machlup, «The Economic Foundations of Patent Law», 1958, http://www.ipmall.fplc.edu/hosted_resources/jepson/unit1/aneconom.htm: «If we did not have a patent system, it would be irresponsible... to recommend instituting one».
24: James Bessen and Eric Maskin, «Sequential Innovation, Patents, and Imitation», 2000, http://www.researchoninnovation.org/patent.pdf. The FFII page contains an interesting criticism of this model, by an anonymous patent lawyer: http://swpat.ffii.org/vreji/papri/bessenmaskin00/indexen.html.
Copyright (C) 2001,2002 Alessandro Rubini <rubini@linux.it>.
Copyright (C) 2002 Stefano Maffulli <stef@zoomata.com>.
La copia letterale e la distribuzione di questa pagina nella sua integrità sono permesse con qualsiasi mezzo, a condizione che questa nota sia riprodotta.
Ultimo aggiornamento: $Date: 2002/12/13 10:13:48 $