Copyright David Webber 1993, Davies Collison Cave , Melbourne, Australia
The new Australian Patent Office practice adopted following International Business Machines Corporation v. Commissioner of Patents[1] and the U.S. Court of Appeals, Federal Circuit decision in Arrhythmia Research Technology Inc. v. Corazonix Corp.[2]appear to have significantly opened the door to patents for computer software. Previously both countries relied solely on the cumbersome U.S. Freeman/Walter/Abele test to determine the patentability of software-related inventions. IBM has forced the Australian Patent Office to apply a completely new test, whereas Arrhythmia provides an avenue for stepping around the Freeman/Walter/Abele test to allow a wider class of software patent claims.
This is a test which the U.S. Court of Custom and Patent Appeals (CCPA) developed in light of three Supreme Court decisions, Gottschalk v. Benson[3], Parker v. Flook[4] and Diamond v. Diehr[5]. The test essentially allows method or process claims for inventions which are implemented by software, and de facto apparatus claims in a "means for" format, provided the claims do not attempt to monopolise or "pre-empt" traditional unpatentable subject matter, such as mathematical algorithms or laws of nature. The test is primarily concerned with mathematical algorithms, as software, by its nature, is used to execute mathematical steps, and mathematical terms and formulae are used to describe software algorithms. Two considerations are involved in applying the test, the first being a determination as to whether the patent claim even recites a mathematical algorithm. Once the first part of the test is affirmed, the second part involves a complicated analysis to determine whether the claim "pre-empts" the recited mathematical algorithm. The CCPA In re Walter[6] said:
"Once a mathematical algorithm has been found, the claim as a whole must be further analysed. If it appears that the mathematical algorithm is implemented in a specific manner to define structural relationships between the physical elements of the claim (in apparatus claims) or to refine or limit claim steps (in process claims), the claim being otherwise statutory, the claim passes muster under section 101[7]. If, however, the mathematical algorithm is presented and solved by the claimed invention, as was in the case in Benson and Flook, and is not applied in any manner to physical elements or process steps, no amount of post-solution activity will render the claim statutory; nor is it saved by a preamble merely reciting the field of use of the mathematical algorithm."
In re Abele and Marshall[8] provides a good example as to how the test is applied and qualifies Walter in view of Diamond v. Diehr. The qualification is summarised in the U.S. Patent Office Manual for Examiners by stating that a claim seeking coverage for a computer program would be non-statutory, when considered as a whole, if it merely recites a mathematical algorithm or mathematical calculation which is not applied in any manner to physical elements or process steps.
The U.S. Patent Office has refused claims to computer programs per se on the basis that the claims relate to a computer language listing of instructions which is a mere idea or abstract intellectual concept that as such is not patentable. Alternatively the claims have been viewed as relating to a collection of printed matter which is also not patentable.
The Australian Patent Office in adopting the U.S. Freeman test simply recited in their guidelines[9] that claims to computer programs per se would continue to be refused, presumably on the basis of previous Patent Office decisions, such as Telephone A/B LM Ericcson's Application[10].
IBM was an appeal heard by the Australian Federal Court on a decision[11] of a Delegate of the Commissioner of Patents in which claims of an IBM patent application were refused on the basis of the Freeman/Walter/Abele test. The invention related to a computer program for producing an improved curve image on a visual display. Two old algorithms, the B spline technique and the interpolating spline technique, were used to compute the coordinate values for the curve, and the invention was said to provide a significant advantage in that integer control points were used instead of floating point control points which gave rise to the improved display. The main claim, claim 1, simply recited a method for producing a visual representation of a curve image, a number of computing steps and a final step of simply displaying the curve without defining the particular environment or any specific physical steps. The Federal Court, however, was willing to conclude that on considering the patent specification as a whole, there was a necessary inference confining the claim to the operation of computers and therefore it was limited to a computer environment[12].
In considering whether the claim then recited patentable subject matter, the Court referred to National Research Development Corporation v. Commissioner of Patents[13], as being the leading authority in Australia[14], where the High Court said
"a process, to fall within the limits of patentability . . . , must be one that offers some advantage which is material, in the sense that the process belongs to a useful art as distinct from a fine art . . . - that its value to the country is in the field of economic endeavour."
The Federal Court in IBM concluded that it is by the production of some useful effect that patent law has been able to distinguish between the discovery of a principle of science and the making of an invention, and that the "production of an improved curve image is a commercially useful effect in computer graphics"[15]. The United States authorities were considered not to preclude this view and the appeal was allowed[16]. IBM was the first time an Australian Court had considered the Patent Office practice for software inventions, and whilst the Court did not disapprove of the adopted practice, it went much further by applying the principles established in NRDC.
In view of IBM, the Australian Patent Office decided in 1992 to adopt a new test[17] for determining the patentability of software inventions. The test is:
"Does the invention claimed involve the production of some commercially useful effect?"
This seems to cast an extremely wide net and it is difficult to envisage what class of software inventions would be excluded, as presumably most would have some commercial application. The test contrasts with the practice of the EPO[18] which involves a consideration of the technical effect which a software related invention has on known equipment. The Australian Patent Office has not gone so far as to condone claims for computer programs per se but nor has it expressly ruled out acceptance of such claims as in the past. Senior officials of the Office have indicated that such claims will be objected to but this would seem to be inconsistent with IBM.
The Office has provided the following illustrations as to how the new test should be applied:
(a) A method-type claim must define a method which, either directly or by clear implication, embodies the commercially useful effect.
(b) A claim to a mathematical algorithm per se is not patentable, because it does not produce a commercially useful effect. A commercially useful effect can only arise when the mathematical algorithm is implemented in some manner to produce a result.
(c) A claim to a mathematical algorithm when used in a computer is patentable so long as a commercially useful effect is produced.
In Arrhythmia, the U.S. CAFC considered an appeal from a decision [19]of the U.S. District Court for the Northern District of Texas which found that U.S. Patent 4,422,459 (the Simpson patent) did not claim statutory subject matter under 35 USC 101. The invention related to a process for analysing electrocardiograph signals in order to determine whether a patient is vulnerable to an acute type of heart arrhythmia known as ventricular tachycardia. The main claim recited a method for analysing electrocardiograph signals by converting a characteristic part of the signals into a time sequence, applying the segments in reverse time order to a high pass filter and then comparing the output of the filter with a threshold level. All of the processing steps could be executed using a digital computer with appropriate software and connected to an electrocardiograph machine. The basis of the District Court's objection was that the claims of the Simpson patent were directed to a mathematical algorithm.
Newman, J. of the Federal Circuit Court found that the claims of the Simpson patent passed the Freeman/Walter/Abele test[20]. He said the claimed method related to physical process steps that transform one physical, electrical signal into another and that the view that there is nothing necessary physical about signals is incorrect. The method was held to comprise an otherwise statutory process whose mathematical procedures are applied to physical process steps. Of particular interest, however, is the basis on which Rader, J. of the CAFC also found for the appellant.
After a detailed consideration of the mathematical algorithm cases of the Supreme Court and the CCPA, Rader, J. concluded that the Supreme Court in Diamond v. Diehr had stressed that the Courts should simply consider the statute, 35 USC 101, when determining whether a claim recites patentable subject matter[21]. In defining patentable subject matter, section 101 refers to, inter alia, a "useful process" and Rader, J. said:
"Courts should give "process" its literal and predictable meaning, without conjecturing about the policy implications of that literal reading . . . If Congress wishes to remove some processes from patent protection, it can enact such an exclusion. Again in the absence of legalised limits on the meaning of the Act, courts should not presume to construct limits. The Supreme Court directed this Court to follow the Act."[22]
He further found that "Diehr limited Benson and its progeny to three classes of unpatentable subject matter - laws of nature, natural phenomena, and abstract ideas"[23] and that the claims of the Simpson patent in defining a process and corresponding apparatus, claimed patentable subject matter within the language of section 101[24]. As far as Rader, J. was concerned "the Supreme Court's most recent message is clear: when all else fails (and the algorithm rule clearly has), consult the statute"[25]. In stating that the Simpson patent did not, in any event, present a mathematical algorithm[26], he questioned the barring of mathematical algorithms from patentability and in the footnotes of his judgement commented that laws of nature are excluded from being patentable as they are not new within the meeting of section 101, and although the Supreme Court in Diehr concluded that a mathematical algorithm is like a law of nature and cannot be the subject of a patent, he felt it was difficult to determine how or why mathematical algorithms are "like" laws of nature. He felt a mathematical algorithm does not appear in nature at all, but only in human and numerical processes.
The new Australian practice only requires a software invention to produce a commercially useful effect in order to constitute patentable subject matter. In the U.S., Arrhythmia provides support for the contention that a claim to a software invention need only recite a new process or apparatus. Both criteria should be relatively easy to meet as it is the function of most computer programs to execute a specific process which is commercially useful.
Software patents have been obtained and continue to be actively sought by hardware manufacturers, such as IBM, but it is also now the software houses, such as Microsoft Corporation, and telecommunications companies which are obtaining and seeking software patents. There is a distinct trend in the U.S. towards relying more on patent protection and narrowing copyright protection for software to its traditional limits, as indicated by decisions such as Computer Associates International, Inc. v. Altai, Inc.[27]. This seems primarily due to the often debated problems in determining the scope of copyright protection for software, and the utilitarian nature of software, which has evolved, in a number of areas, to succeed hardware in importance, both commercially and functionally.
Patents having traditionally been granted for hardware and there no longer seems to be any logical basis for excluding patents for software. A software development will still need to be sufficiently new and inventive to qualify for a patent, which is a significantly more onerous requirement to meet than the low threshold of originality required to gain copyright protection. Also the published patent specification provides third parties with at least some guidance as to the scope of protection afforded to the patentee. Difficulties associated with patents for software, of course, need to be attended to, such as disclosure requirements and examination standards and the extent of prior art material available to patent examiners. Work in this regard has already commenced in the U.S. with the establishment by the Software Patent Institute of a non-patent prior art database for U.S. Patent Examiners.
Unless there is a dramatic turn around in U.S. and Australian practice, the day when claims to computer programs per se can be obtained does appear to be too distant.
1. (1992) AIPC 90-853, 22 IPR 417
2. 22 USPQ 2d 1033
3. 175 USPQ 673, 1972
4. 198 USPQ 193 (1978)
5. 209 USPQ 1, 1981
6. 205 USPQ 397, at 407
7. Section 101 is the U.S. patentability provision, 35 USC 101, which recites "whoever invents or discovers any new or useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this Title".
8. 214 USPQ 682
9. "Guidelines for Considering the Patentability of Computer Program Related Inventions", Australian Patent Office, 1986
10. 1975 FSR 49
11. (1991) AIPC 90-781
12. (1992) AIPC 90-853, at 38162
13. (1959) 102 CLR 252
14. Ibid, at 38163
15. Ibid, at 38165
16. Ibid
17. Patent Examiners Manual 35.65, August 1992
18. European Patent Office Guidelines, C-IV 2.3
19. Arrhythmia Research Technology, Inc. v. Corazonix Corp., No. CA 3-88-7745-AJ (N.D. Tex. October, 1990)
20. 22 USPQ 2d, at 1038
21. Ibid, at 1044
22. Ibid, at 1042
23. Ibid, at 1043
24. Ibid, at 1044
25. Ibid
26. Ibid, at 1043
27. 23 IPR 385