The European Patent Office has just released its 1994 Annual Report, always an interesting read of patenting activities in Europe. This year's annual report has a length introductory essay dealing with the problems and controversies surrounding software patents.
The essay is a pretty good assessment of the history and current status of software patenting in Europe and America, though slightly deficient in not discussing the more advanced problems plaguing software patenting around the world (in particular, hardware/software codesign as it impacts statutorial definitions and treatment of software patents).Information and communication technologies, key technologies of our time, are exerting a decisive and growing influence on our lives. The steady shift in emphasis from industrial production to a service economy has caused a surge in the importance of information as a strategically vital instrument for the whole economy; information has long been the fourth factor in industrial production. Mastery of information technology has become a potent ingredient in the performance of a national economy and the competitiveness of commerce and industry. Plans for the future envisage global data superhighways, linking databases and information systems into worldwide networks.
The European Commission recognised the signs of the times and in July 1994, following the Bangemann report "Europe and the global information society", presented an action plan for "Europe's way to the information society". The Commission's plans lays down six requirements for a worldwide information and telecommunications infrastructure. This should be geared to:
- promoting compatibility and interoperability - developing global markets for networks, services and applications - guaranteeing individual privacy and data security - protecting intellectual property rights - promoting co-operation in research and development - monitoring the personal & social effects of the information society
The Commission was right to assign a high priority to intellectual property rights if the information society is to live up to our expectations. The economic impact of information technology is immense, the "information industry" is a growth area par excellence, the rate of innovation extremely high. Information technology pervades every aspect of life from commerce and industry to government; at the same time, it is reaching into people's private lives and homes. The world market for hardware and software will continue to expand, but competition will grow more relentless. The potential market volume for all types of software products in the European Union alone is thought to be around DEM 300 billion.
The information industry needs effective protection for its products and processes to ensure that it can cover the often very high costs of developing new hardware and software. It must be able to defend itself against software piracy and hardware imitation. Hardware such as computers, storage media, microprocessors, peripherals and other electronic devices, needs to be protected, but so does the software without which much of the hardware would be pointless. This includes computer operating systems and other system software, programs for computer-controlled operation and monitoring of industrial machines and processes, software for the microchips in everything from household appliances to vehicles, application programs for a wide variety of tasks and, more recently, software for expert systems, artificial intelligence and neural networks.
The information industry's need for protection is obvious: many of its products - especially software - can easily be copied or imitated. Complaints of rampant software piracy are growing louder. The US Software Publishers' Association reckons that 98% of the personal computer software sold in China in 1994 had been pirated. For Russia the estimate is 95% and for Thailand 92%. Loss of revenue in these countries is estimated at USD 328 million. The Business Software Alliance reckons that illegally copied standard software accounts for 40% of the US market, 76% of the German market, 81% in Japan and 98% in Thailand. The seriousness of the situation is shown by the threat, since averted, of a trade war between the USA and China. The USA accused Chinese authorities of taking inadequate measures against software piracy. Litigation on patent or copyright infringement of hardware or software is on the increase worldwide. At the end of 1994, for example, a New York court ordered the Japanese video games firm Nintendo to pay damages of USD 208 million for patent infringement.
A broad range of protection is available for hardware and software. Information technology products and processes can be patented. Copyright can protect all types of computer program and other software products like video games. The 1991 EU directive on protecting computer programs standardised and strengthened copyright protection in Europe. The WTO/TRIPS agreement will improve copyright protection for computer programs worldwide. Trademarks can be registered and other intellectual property rights acquired for both hardware and software products. Finally, legislation design to combat unfair competition and protect know-how can be an effective weapon in individual efforts to prevent piracy. Most information technology products and processes can be afforded the optimum protection through a skillful choice of methods.
The European Patent Convention clearly states that patents are to be granted for new "technical" inventions which involve an inventive step and are susceptible of industrial application. On the other hand, Article 52 of the Convention lists a number of things and activities which are not viewed as "technical" inventions and are therefore not patentable. These include mathematical; schemes, rules and methods for performing mental acts, or doing business; and programs for computers as such. The European Patent Office (EPO) is thus debarred from granting patents for computer programs as such; they can be protected by copyright. On the other hand, the EPC in no way prohibits the patenting of computer-related inventions, or more precisely, of methods, systems, devices or products in which new computer software plays an important role.
As applications for this type of patent increased steadily in number during the mid-80s, the EPO held in-depth consultations on the issue with representatives of industry, patent attorneys and national patent offices. Then, in 1985, it published its revised guidelines for examination with a view to enabling appropriate protection to be made available for software technology. The revised guidelines specify that an invention possesses the technical character required for patentability if as a whole it makes a "technical contribution" to the known state of the art. Using the tried and trusted "problem and solution approach", the examiner assesses whether the invention uses technical means to solve a technical problem. The yardstick is whether the problem and its solution involve technical considerations and know-how or whether this know-how is of a non-technical nature involving, for example, mathematics, linguistics or pure programming. The concept of technology is given a broad interpretation covering the processing, storage, representation and retreival of information by technical means. If a software-related invention meets these criteria for technical character, then the fact that a computer program is involved in its realisation does not detract from its patentability. Programmable controllers and manufacturing or monitoring systems operating under computer control can therefore generally be patented. This is also the case when the internal operation of a known computer is controlled by a program so that a technical effect results. Crucial to this approach is that the inventive achievement may be based on a new program or algorithm.
The approach initiated by the revised guidelines in 1985 has proved basically satisfactory and offered wide scope for patenting software technology. Since the watershed VICOM decision in 1986, appeal board case law has accumulated over 40 decisions on the patentability of software related inventions. In the main, these have confirmed the Office's grant practice. Since its founding, the EPO has granted over 11,000 patents for software-related inventions in the core areas of information technology alone, in digital data processing, data recognition, representation and storage (IPC classes G06 and G11). Fewer than 100 applications in this area were refused on the grounds that the subject-matter lacked technical character. In addition, several thousand patents were granted in other technical areas such as automotive and mechanical engineering where the rapid pace of innovation derives from the use of computer and software technology, e.g. program-controlled processors. The survey carried out by Dr. Hans Goldrian on behalf of the EPO into grant practice and client satisfaction in the electricity/physics sector found broad agreement with the EPO's approach to software-related inventions.
In a number of contracting states - Germany, the Netherlands, Sweden and the United Kingdom - the national patent offices and courts have aligned their approach with that of the EPO and use the same or similar criteria to decide on the patentability of software-related inventions. Nonetheless some decisions in UK courts and German patent courts have applied more restrictive criteria than the EPO in evaluating the technical character of such inventions and have refused patent protection when the sole inventive contribution has been a computer program which was the expression of a purely mental act.
Trilateral co-operation and the goal of harmonising US, European and Japanese patent grant practice led the EPO in 1989 to commission a comparative study of how the three offices handled computer-related inventions. The study found that although they used radically different methods to decide whether an invention was patentable, they generally reached similar conclusions. In 1993 the Japanese Patent Office revised its examination guidelines, bringing them more closely into line with EPO practice. In the USA, a number of decisions of the Court of Appeals for the Federal Circuit in Washington have made it much easier to patent any type of computer software, thereby further lowering the barriers against software patents. The EPO will be analysing these decisions carefully to see whether they have any bearing on European patent practice.
It is often said that a patent is only as good as the preceding search. This could be especially true of software patents. Greg Aharonian, director of a private patent search service in the USA, concluded from his own study of the problem that patent offices are not in a position to carry out a reliable search for software-related inventions. He argues that a high percentage - something like 75% - of the 5000 or so software patents granted in the USA in 1993 should actually have been refused on the basis of prior art which was known but had not been found.
A sensational, if not exactly representative, example of this problem was Compton's 1993 US patent 5241761 for multimedia software, which caused such an outcry that USPTO Commissioner Lehman ordered a re-examination. As a result of the this second examination the patent was revoked in its entirety on the basis of 18 citations which had not been considered during the first examination.
Aharonian's assertion may well be exaggerated, but cannot be rejected out of hand. Patent office are aware that software-related inventions pose special problems for searches. The major difficulty is that much of the relevant state of the art is to be found not in patent documents but in non-patent literature, eg articles in the specialist press, conference papers and company documents. Accessing these sources can be laborious and the material is often not ideally presented for search purposes. Moreover the patent classification system has not kept pace with the development of software technology.
[.....a paragraph of nonsense .....]
The European, Japanese and US Patent Offices will also do their utmost within the framework of their trilateral co-operation to bring about lasting improvements to searches on software-related inventions.
It has recently been suggested that the solution to the problem of protecting computer software lies in discarding thesection of Article 52 EPC which prohibits the patenting of computer programs as such. Those who advocate this approach cite the WTO/TRIPS agreement which calls for unrestricted patent protection for computer programs. Even if the EPO does not share this view, the suggestions will have to be more carefully evaluated when the time comes to revise the EPC. On the other hand, it is important to realize that the Convention already offers appropriate patent protection for new software technology. The Office is confident that it is fulfilling its mandate to grant European patents for new and inventive technical achievements in all fields, including information technology, and that is meeting the needs of the information industry.