Software Patents After Fujitsu

New Directions or (another) Missed Opportunity?

Prof. Ian Lloyd
Strathclyde University
i.j.lloyd@strath.ac.uk

Contents

1.	Background
	1.1	Product and Process Patents
	1.2	The Patents Act and the European Patents Convention
2.	Fujitsu's Application
	2.1	Technical Contribution
	2.2	VICOM
	2.3	Merrill Lynch
	2.4	The Decision in Fujitsu
3.	Moving Apart from Europe on Mental Acts
	3.1	A New Kind of Expert System
	3.2	Image Identifying
	3.3	IBM/Text Processing
	3.4	Other EPO Authority
4.	Mental Acts in Fujitsu
5.	Conclusion

Download

---

This is a Case Note published on 30 June 1997.

Citation: Lloyd I, 'Software Patents After Fujitsu . New Directions or (another) Missed Opportunity?', Case Note 1997 (2) The Journal of Information, Law and Technology (JILT). <http://elj.warwick.ac.uk/jilt/cases/97_2fuji/>. New citation as at 1/1/04: <http://www2.warwick.ac.uk/fac/soc/law/elj/jilt/1997_2/lloyd/>

---

1. Background

Patent applications in respect of software related inventions have not fared well before the United Kingdom courts. With the exception of the first instance decision in Gale ([1991] RPC 305), a judgment which was comprehensively overturned in the Court of Appeal, every application brought under the Patents Act 1977 has been rejected as relating to non-patentable subject matter. Applications brought under the terms of the European Patent Convention and adjudicated before the European Patent Office Board of Appeal have been more successful.

At first glance it may appear surprising that the issue of patents for software developments should have reached the courts at all. The Patents Act 1977 provides specifically in section 1(2) that patents are not to be awarded for:

• a discovery, scientific theory or mathematical method;
• a literary, dramatic, musical or artistic work or any other aesthetic creation whatsoever;
• a scheme, rule or method for performing a mental act, playing a game or doing business, or a program for a computer;
• the presentation of information.

Matters, however, are not so straightforward. In his judgment in Fujitsu Ltd's Application ([1996] RPC 511) which was subsequently affirmed by the Court of Appeal (The Times 14 March 1997), Mr Justice Laddie analysed the rationale behind a number of the statutory exceptions. The prohibition against the grant of a patent to a discovery illustrates perfectly the problems inherent in this area. The obvious objection to awarding a patent for a discovery, for example of a new mineral, is that there is no discernable inventive step. However, as was pointed out in the judgment:

most inventions are based on what would be regarded by many people as discoveries. Large numbers of highly successful and important patents in the pharmaceutical field have been and continue to be based upon the discovery of new strains of micro-organisms which exist naturally in the wild.

recognising this fact, the statutory prohibition against the grant of a patent is restricted to the case where the application relates to the discovery 'as such'. In principle, such an approach must be correct. Its practical application has proved more difficult with particular problems surrounding the treatment of what are frequently referred to as 'software related inventions'. In part the problem may lie with the fact that both Act and Convention were enacted in the 1970s. At that time, it was considered that computer programs could be separated from the hardware components and should be excluded from the patent system. Both the report of the Banks Committee in the UK and the initial Guidelines for Examiners produced by the European Patent Office make this point clearly. Over the last 20 odd years the nature of computer programs has changed and expanded and the division between software and hardware has become a matter of chance as much as one of technology. It would be absurd if an invention were to be denied a patent on the basis, for example, that its 'on/off' switch was controlled by software embedded in a microprocessor chip. Equally it is clear beyond doubt that a computer program taken in isolation is not patentable. As with so many areas of the law, the difficult task is to determine where the boundary lies between permitted and prohibited subject matter.

To complicate matters further as the relevance of the obvious prohibition has declined, so it has also become apparent that software related inventions are vulnerable to challenge under a range of the statutory exceptions. Applications have been rejected on the basis that they relate to a mathematical method, a method of doing business, the presentation of information and a method for performing a mental act, all of which are excluded from the award of a patent. It is difficult to think of any other form of technology whose nature and range of application is so chameleon as to bring it within so many of the statutory prohibitions. Not unnaturally, those seeking patent protection for software related inventions have sought to lay as much emphasis as possible on the task performed by the invention and as little as possible on the contribution made by computer programs. The criterion applied by both the EPO and the UK authorities is to require that the claimed invention produced a 'technical contribution' to the state of the art (also referred to as a 'technical effect' or 'technical application'). The next question, of course, is whether the mere presence of a technical contribution can outweigh the explicit prohibition against patentability?

1.1 Product and Process Patents

Subject always to the requirements of novelty, inventiveness and industrial application, patents may be claimed either for a product or a process (sometimes referred to as apparatus and means). In many instances, applications will combine the two elements. A helpful illustration is provided in Mr Justice Laddie's judgment in Fujitsu:

... it may be useful to consider what the position would be in a case where someone had invented a new way of mowing grass which involved designing a new type of motor with micro sensors and blade adjustment motors on it, the sensors being used to determine both the softness of the grass to be cut and the height of it above the ground and then produced an output which operated the motors so as to adjust the height of the cut, the angle of the blades and the speed at which they rotated. ... considerations of novelty aside, such a device would be patentable and, so it seems to me, would be the mowing method itself.

In the software context, the claim may often be that the equipment operating in accordance with the program's instructions constitutes a novel product whilst the algorithmic steps prescribed by the implementing programs represent a novel process.

1.2 The Patents Act and the European Patents Convention

The Patents Act was enacted in order to enable the UK to ratify the European Patent Convention and provides that judicial notice is to be taken of decisions of the European authorities. As was stated by Nicholls LJ in Gale ([[1991} RPC 305)

It would be absurd if, on the issue of patentability, a patent application should suffer a different fate according to whether it was made in the United Kingdom under the Act or was made in Munich for a European Parliament (United Kingdom) under the Convention.

In spite of this recognition, concerns have been raised that software related applications have been treated more harshly before the UK courts. The concerns are twofold. First as was suggested by counsel for the Comptroller of Patents in the Fujitsu case:

there are recent decisions of the EPO Technical Board of Appeal in which the question of what constitutes a technical advance or contribution seems to have been rather more flexibly interpreted than has hitherto been the case.

Additionally, concern has been expressed that different criteria might be adopted in dealing with challenges based upon the prohibition against the award of a patent for a scheme or method for performing a mental act. Given the increasing use of expert systems and neural networks, this is likely to prove a significant area of development in the coming years. Two decisions of the High Court in the cases of Wang's Application (1991) and Raytheon Co's Application (1993) have been identified as opening up a discrepancy between the UK and European approaches. The recent litigation in respect of a patent application by Fujitsu presented the High Court and the Court of Appeal with the opportunity to consider both these issues.

2. Fujitsu's Application

The technology at issue in Fujitsu will be familiar to anyone with a recollection of chemistry lessons at school and the use of three dimensional lattices to depict molecular structures. Fujitsu's invention sought to bring this concept into the age of virtual reality allowing chemists to depict and manipulate crystal structures on a computer screen. The effect would be to allow the analysis of the properties of new compounds without the need to create these in the real world. The novelty in the claimed invention lay only in the relevant computer programs.

The patent application was rejected in the Patent Office on the basis that it related to a program for a computer as well as to a method for performing a mental act - that of visualising molecular structure. An appeal was taken to the High Court where counsel for the Comptroller of Patents recognised that there had been 'difficulty in applying these statutory exclusions'. '(S)trict application of the guidance laid down in the authorities' it was suggested, 'leads to the exclusion from patentability of a considerable number of inventions which do in reality appear to provide a substantial contribution to the sum of technical knowledge.' The hope was expressed that the decision of the court would provide 'guidance' in determining 'the scope and application of the exclusions'.

It is unclear whether this hope has been realised. Two aspects of the decision will be considered in this article. First, consideration will be given to the operation of the prohibition against computer programs, second to that of the prohibition against patenting a scheme, rule or method for performing a mental act'.

2.1 The Requirement for a Technical Contribution

Although the term 'technical contribution does not appear in either Act or Convention, it has achieved pivotal significance since being adopted by the European Patent Office Board of Appeal in the case of VICOM in 1987 (1987 2 EPOR 74) and endorsed by the English courts in the case of Merrill Lynch (([1989] RPC 561). In the course of his judgment Mr Justice Laddie made extensive reference to European and English precedents and it may be helpful to summarise the key features of some of these cases.

2.2 VICOM

The VICOM application related to the use of a computer for image processing purposes. Data representing the image, in the form of electrical signals, would be processed by the computer so as to enhance the quality of the image as displayed on the computer monitor. This application was rejected by the examiner on the grounds both that it sought protection for a computer program and on the basis that it related to a mathematical method. The electrical signal, it was argued, could be represented in mathematical terms, likewise the processed signal.

The Technical Board of Appeal disagreed. In respect of the program objection it was held that:

Generally, claims which can be considered as being directed to a computer set up to operate in accordance with a specified program (whether by means of hardware or software) for controlling or carrying out a technical process cannot be regarded as relating to a computer program ...

Generally speaking, an invention which would be patentable in accordance with conventional patentability criteria should not be excluded from protection by the mere fact that for its implementation modern technical means in the form of a computer program are used. Decisive is what technical contribution the invention as defined in the claim when considered as a whole makes to the known art.'

It was further recognised that a mathematical method could not be protected directly. Equally it could not be protected indirectly by a claim directed 'to a method for digitally filtering data'. That it was held, would be as abstract and unpatentable as the mathematical method itself. What was required was that the mathematical method should be applied within a specific technical context which, being capable of industrial application would qualify for patent protection. In this event, the mathematical methods could freely be used by third parties for any purpose other than the specified form of image processing. Such an approach overcomes one of the major concerns which has been expressed by opponents of software patents - especially in the United States - that a patent could be infringed by a party working out calculations with pen and paper.

2.3 Merrill Lynch

If VICOM constitutes a landmark decision under the Convention, the Merrill Lynch decision played a similar role in UK patent law. In its report on the patent system, the Banks committee had recommended that an invention making use of a computer program, 'and where the novelty or alleged novelty lies only in the program, should not be patentable'. In Merrill Lynch , the application related 'an improved data processing based system for implementing an automated trading market'. Effectively, the system permitted automated trading in stocks and shares. Before the Patent Office and the High Court the application was rejected on the basis that the only novelty lay in the computer programs involved. By the time the case came before the Court of Appeal, the VICOM decision had been published and endorsed by the Court of Appeal in another patent appeal, that of Genetech's Application ([1989] RPC 147). Although this case was concerned with biotechnology and the application of the prohibition against patenting a discovery, it constituted clear precedent for the proposition that an invention might be patentable if the novelty lay only in prohibited subject matter so long as there was the requisite technical contribution. Ultimately, however, Merrill Lynch's application was rejected on the basis that the end product was no more than a means of doing business, itself prohibited subject matter.

2.4 The Decision in Fujitsu

Relying on the VICOM decision, Counsel for Fujitsu suggested that what was required to bring prohibited subject matter such as computer programs or schemes for performing mental acts within the patent system was that these should be tied to some technical application. In the present case, as in VICOM, the technical component was that the application resulted in the manipulation of images. In VICOM, images of physical objects were processed, in the present case, images of molecular structure.

After surveying relevant UK and EPO authority, Mr Justice Laddie concluded that four propositions could be drawn relating to the patentability of software related inventions:

1. The types of subject-matter referred to in section 1(2) are excluded from patentability as a matter of policy. This is so whether the matter is technical or not.
2. The exclusion from patentability is a matter of substance not form. Therefore the exclusion under section 1(2) extends to any form of passive carrier or recording of excluded subject matter. Thus, merely because a piece of paper is in principle patentable (save to the extent that it lacks novelty), it is not permissible, for example, to record a literary work (section 1(2)(b)) or a computer program (section 1(2)(c)) on a piece of paper and then seek patent monopoly for the paper bearing the recorded work. Similarly it is not permissible, without more, to seek protection for a computer program when it is stored on a magnetic medium or when merely loaded into a computer.
3. Prima facie a computer running under the control of one program is a different piece of apparatus from the same computer when running under the control of another program. It follows that a claim to a computer when controlled by a program or to a method of controlling a computer by a program or to a method of carrying out a process by use of a computer so controlled can be the subject of patent protection. However, because the court is concerned with substance not form, it is not enough for the designer of a new program to seek protection for his creation merely by framing it in one of these terms. The court or patent office must direct its attention not to the fact that the program is controlling the computer but to what the computer, so controlled, is doing.
4. Therefore a data processing system operating to produce a novel result would not be deprived of protection on the ground that it was a program as such. On the other hand, even if the effect of the program is to make the computer perform in a novel way, it is still necessary to look at precisely what the computer is doing, i.e. at the nature of the process being carried out. If all that is being done, as a matter of substance, is the performance of one of the activities defined under section 1(2) as unprotectable, then it is still unprotectable.

On the basis of these criteria, he concluded:

just as it would be possible to obtain a patent, considerations of novelty aside, for a faster chip or a more effective storage medium or a computer containing such a chip, there is no reason in principle or logic why modification of the computer to achieve the same speed or storage increase by means of software should be excluded from protection.

Attention must therefore be focused on what the claimed invention achieves rather than the manner in which it accomplishes this.

At first instance, it is unclear whether the Fujitsu application was regarded as constituting nothing more than a computer program. Holding that 'prima facie, they avoided the program exclusion', Mr Justice Laddie continued:

The real issues it seems to me is whether the application also avoids the other exclusions ... If it does not, the application will fail. Whether in those circumstances, the grounds of failure are stated to be that the invention is only for a program or that it is, for example, a method for performing a mental act is a matter of semantics.

The issue was considered in more detail in the Court of Appeal. Here, Aldous LJ held that the application should be rejected on this basis. Comparing the claimed invention with that at issue in VICOM he concluded that:

In VICOM, the technical contribution was provided by the generation of the enhanced display. In the present case, the combined structure is the result of the directions given by the operator and use of the program. The computer is conventional as is the display unit. The displays of crystal structures are provided by the operator. The operator then provides the appropriate way of superposition and the program does the rest. the resulting display is the combined structure shown pictorially in a form that would in the past have been produced as a model. The only advance is the computer program which enables the combined structure to be portrayed quicker.

It might be suggested that a quicker method of producing a display of molecular structure should be regarded as patentable in the same way as would a better mousetrap or a method for producing better quality images. Certainly the comments made regarding the conventional nature of the computer and the fact that 'virtual reality' displays are substituting for traditional three dimensional models might be ground for challenge on the issues of novelty and inventiveness, but the distinction drawn with the situation in VICOM exacerbates rather than clarifies the issue what is required to produce a technical effect.

3. Moving Apart from Europe on Mental Acts

The Fujitsu case illustrates well what is becoming a feature of our information society. Computers and computer programs are being used in situations and for activities which were previously the province of humans. The machine may well substitute in whole or in part for human thought. In this situation, it is very possible that the end product of the process might be classed as a scheme or method for performing a mental act and any claim for patentability rejected on this basis.

In respect of the interpretation of this ground of exclusion concern has centred both on the relationship between the subject matter and the requirement for a technical effect, a situation where the factors discussed above in the computer program context are equally apposite, and also as to the definition of the ground per se. Concerns have been expressed that the UK courts may be adopting a more restrictive approach than is the case in the EPO. Prior to Fujitsu , this ground of objection had been at issue in the cases of Wang ([[1991] RPC 463) and Raytheon ([1993] RPC 427).

3.1 A New Kind of Expert System

In Wang , the claimed invention related to a novel form of expert system. This was held to be unpatentable on the ground that it related to nothing more than a computer program. Reference was also made in the case to the statutory prohibition against the grant of a patent in respect of a scheme or method for performing a mental act. Counsel for Wang argued that the phrase "scheme, rule or method for performing a mental act" only applied to methods which were capable of being performed in the human mind. The operation of the expert system, although seeking to produce results similar to those arrived at by a human expert, utilised steps and procedures which would not be replicated by such a person. This interpretation was rejected by Aldous J (as he then was) who held that:

Just as a claim to a disk containing a program can be in fact a claim to an invention for a computer program, so can a claim to steps leading to an answer be a claim to an invention for a method for performing a mental act. The method remains a method for performing a mental act, whether a computer is used or not. Thus a method of solving a problem, such as advising a person whether he has acted tortuously, can be set out on paper, or incorporated into a computer program. The purpose is the same, to enable advice to be given, which appears to me to be a mental act. Further, the result will be the advice which comes from performance of a mental act. The method may well be different when a computer is used, but to my mind it still remains a method for performing a mental act, whether or not the computer program adopts steps that would not ordinarily be used by the human mind.

3.2 Image Identifying

The decision in Wang was approved in the subsequent case of Raytheon's Application . At issue in this case was a method of automatically identifying objects such as ships. The image of the object's silhouette would be captured by some form of imaging device such as a camera and transformed into digital format. The digitised image would then be processed by a standard computer. This process involved making a comparison with a library of images stored on the computer in order to select the most appropriate match. A patent was sought for the process but was rejected within the Patent Office on the grounds that the application related to no more than a method for performing a mental act using a computer. This interpretation was upheld in the Patents Court where Mr Julian Jeffs QC (sitting as a Deputy Judge) held that the phrase "a mental act" had to be construed in its normal sense. This was equated with the possibility of explaining in words how a mental act had been performed. Thus:

I have long known that seven eights are fifty-six. By an analytical process I could demonstrate how that result is derived. It is certainly a mental process. But if I meet a friend in the street and I recognise him, it is hard to explain in words precisely why. No doubt the act of recognition could to some extent be analysed and expressed in words, but no combination of words would enable a total stranger to pick my friend out of a crowd. To move on to a more abstract conception, if I smell an orange, I know that it is an orange. I recognise it by such a process of comparison calling up the smell of oranges that I have smelt in the past, but no words could be formulated that would enable someone to recognise the smell of an orange who has never smelt one. Yet in recognising the orange I have no doubt that I am performing a mental act. It would no doubt be possible to devise an apparatus which would achieve the same result by a process of chemical analysis. Such an alternative process of recognition would not be a mental act and the apparatus might well embody one or more inventions. There would be a technical advance, but that is not what is happening in the present invention.

The approach developed in Wang and Raytheon would deny protection to any software related innovation performing a function which, in principle (author's emphasis) could be carried out within the human brain even though this would not reflect the normal process of human reasoning. This approach sits easily with at least some of the jurisprudence of the EPO

3.3 IBM/Text Processing

This application referred to a novel method for correcting homophone errors in a document, e.g. the use of the word 'where' when the context of the document required 'were'. Such a facility is an important feature of speech recognition systems but is also a process which is carried out (often imperfectly) within the brain of an author. The application, it was held, related only to known and standard apparatus and was described in functional terms corresponding to the mental steps which would be carried out by a human performing the same text processing operations. Holding it unpatentable the Board of Appeal ruled that:

Since the only conceivable use for a computer program is the running of it on a computer, the exclusion from patentability of programs for computers would be effectively undermined if it could be circumvented by including in the claim a reference to conventional hardware features, such as processor, memory, keyboard and display, which in practice are indispensable if the program is to be used at all. In the opinion of the Board, in such cases, patentability must depend on whether the operations performed involve an inventive step in a field not excluded from patentability.

As was commented by Mr Justice Laddie, the Board of Examiners appear to be attempting to distinguish the 'substance' of an application from the 'form' in which it is expressed.

3.4 Other EPO Authority

Although the decision in IBM/Text Processing resulted in rejection of the application, other EPO decisions not cited in Fujitsu demonstrate the complexity of the issues involved. The case of Koch and Sterzel's Application (1988) concerned claims relating to an invention whereby a data processing device operated to ensure that patients received the optimum dosage of radiation from an X Ray machine. The novelty lay only in the processing apparatus and again the task involved was one which had hitherto been carried out by a human operator. The Board of Appeals in holding that the invention did not relate to prohibited subject matter, considered and rejected as contrary to the Convention's requirements case law of the German Federal Court of Justice to the effect that:

a teaching is not technical if in its essence it states a rule that can be carried out without employing controllable natural forces other than human brainpower, even if the use of technical means appears expedient or indeed the only sensible and hence the necessary procedure, and even if reference is made to these technical means in the claims or descriptions.

The decision in Koch does not appear to decide anything beyond the fact that the German criteria were unduly restrictive. A further decision relating to an application from IBM ( IBM Corp/Reading Age, 1990 ) is more explicit. This application concerned a system for checking automatically the text of a document in order to highlight words having a reading age higher than that specified for its readers. The system would go on to present a list of alternative formulations which would meet the appropriate age requirements. Again, the equipment could be seen as replicating functions traditionally carried out by human editors. Although the particular application was rejected, the Board of Appeal held that such a development might be patentable if the technical manner in which the process was conducted involved an advance on the state of the art even though (author's emphasis) the steps taken might correspond to those performed in the mind of a human.

4. Mental Acts in Fujitsu

Holding that the application related to no more than a scheme or method for performing a mental act, Mr Justice Laddie stated that:

In VICOM, the Board explained that a mathematical method could be distinguished from a patentable process based on it in that the former involved an abstract concept in which numbers were worked on to produce new numbers whereas in a patentable process a physical entity was worked on and a new physical entity was produced. Very similar concepts apply to the distinction between methods for performing mental acts and processes methods or apparatus based upon such acts. Excluded mental acts must include those mental activities which involve a significant level of abstraction and intellectual generality. Rules as to the planting of potatoes in which the operator is instructed to measure and evaluate matters such as the type of soil, location, weather and availability of irrigation is a method for performing a mental act. Directions to plant one seed potato every metre is not. It is a precise process. In this case, Fujitsu's application leaves it to the operator to select what data to work on, how to work on it, how to assess the results and which, if any, results to use. The process is abstract and the result of use of it is undefined. What is produced is not an inevitable result of taking a number of defined steps but is determined by the personal skill and assessment of the operator. As such it consists in substance of a scheme or method for performing a mental act and is unpatentable.

The decision in Fujitsu provides no specific answer to the definitional questions raised above. It does appear to support a restrictive view of the scope of patentability. In some respects, the result seems somewhat paradoxical. A computer controlled potato planting machine which plants a potato every metre might be patentable. Were the identical machine to have more sophisticated software allowing account to be taken of factors such as soil conditions, weather and the presence of other crops, no patent could be awarded even though this second machine might appear more technologically advanced and deserving of protection. Again, reference to the role of the Fujitsu system in presenting information for the operator to act upon might be compared with many other pieces of machinery. The example might be presented of a novel form of navigational aid incorporated on an aircraft flight deck. This might present the pilots with information as to height, speed and direction of flight. Assuming that the device made use of mechanical components, there would be no bar to its patentability.

5. Conclusion

It is now 20 years since patent law was reformed by the Act of 1977. Not surprisingly, the legislation was based upon a view of computer technology as it was practised at the time. In the intervening years, not only has the technology permeated into every aspect of life, the development of microprocessors has rendered almost redundant distinctions between hardware and software - to the extent that the term computer program is seldom used today. From a situation of existing as a rather small adjunct to the industrial society, information technology has become pivotal to the information society. Software development has changed from a craft to an industry. The turnover and profits of software companies such as Microsoft dwarf those of the vast majority of industrial enterprises. The development of satisfactory forms of protection is a matter of great importance.

One of the legislative trends of the 1980s was to provide that computer programs are protected under the law of copyright. Certainly copyright provides an acceptable and appropriate form of protection for most computer programs which do not possess significant elements of novelty or originality. Copyright, however, particularly given precedents in the United States and the UK placing limits on the scope of protection against non-literal copying, is less suitable as a vehicle for protecting innovative works. Competitors can readily discern the underlying - and unprotected - ideas and replicate these without the necessity to engage in literal copying of any of the code used in the original. In such situations, the attractions of the patent system are apparent. In return for disclosing details of the techniques employed, the patent holder secures monopoly protection against reproduction of the novel ideas.

When the topic of the patentability of computer programs was discussed by the Banks Committee in the 1970s. the issue was agreed to be finely balanced. Ultimately, the Committee recommended against eligibility on grounds both of principle and practice. In terms of principle it was argued that no significant distinction existed between programs and methods of mathematical calculation which had always been excluded from protection. Practical difficulties were also identified, the committee commenting:

... were programs to be patentable, very real and substantial difficulties would be experienced by the Patent Office in searching applications for program patents even were the search material available in suitably classified form. The issues of novelty and obviousness would be so difficult of determination that patents of doubtful validity would be likely to issue.

These arguments cannot be discounted. It may have been preferable had the relatively hard line against patentability advocated by Banks been enforced by the courts. Once the dam had been broken by the decisions in VICOM and Genentech, the line has proved impossible to hold. In Fujitsu, Mr Justice Laddie commented that the distinction between the prohibition against programs and that relating to methods for performing a mental act was 'a matter of semantics'. In respect of may of the decisions and distinctions drawn, it may be suggested that the issue of patentability has been submerged in a semantic sea. Whilst accepting that there may be reasons of principle why no software patents should be issued, it is more difficult to accept at this level that an image processing system should qualify whilst a virtual reality system would not.

This is not to criticise the patent authorities who have to implement what may be considered an outmoded system. In respect of the approaches in the UK and Europe it may be suggested that if discrepancies do exist it is because the EPO decisions have not always followed a consistent line. In such a situation it is impossible for national courts to adhere to two (or more) conflicting lines of authority. The decision in Fujitsu clarifies some important issues and cannot validly be criticised for failing to reconcile the irreconcilable. The next step is one for the legislature. Assuming that the patent system is to continue in the information society, the obvious step would be to repeal all the statutory prohibitions and place reliance upon the core requirements that an invention be:

• novel;
• involve an inventive step and:
• be capable of industrial (commercial?) exploitation

Such an approach would allow the authorities to concentrate upon the merits of an application rather than play word games. There would remain the practical difficulty identified by Banks of determining issues of novelty and inventiveness in an area of rapidly changing technology. Certainly, much of the controversy in the United States where software patents appear to be much more readily available has centred on the argument that patents have been awarded to developments that were not truly novel or inventive. Against this it must be remembered that the award of a patent is by no means conclusive as to questions of validity. Any patent may be challenged by any person at any time. Indeed the notion of examining patents for novelty and inventiveness prior to award is a comparatively recent development in our law.