25 July 2006

Full Court Finds Asset Protection Method not Patentable Subject Matter

Introduction

In my post of 14 August 2005, I reported that Justice Branson had upheld a decision of the Commissioner of Patents to grant a patent for "An Asset Protection Method". The relevant claim can be seen in that post.

On 18 July 2006, the Full Bench of the Federal Court handed down Grant v Commissioner of Patents [2006] FCAFC, which was an appeal from the case reported in my post. Perhaps to avoid the high costs of litigation, Grant himself appeared in what seemed to be a desperate and vain attempt to challenge settled law in Australia.

In this post, I will have a look at some of the important aspects raised by the Full Bench. However, this matter does not raise any new law, possibly evidenced by the fact that the judgment is only 14 pages long, a refreshing change from some of the turgid documents generated by this court in the past.

Brief Background

In the initial hearing before the Commissioner of Patents, the Deputy Commissioner concluded that the invention did not involve any discovery of a law of nature nor involve the application of technology in some form or other to perform the process claimed. He held that it does not result in an "artificially created state of affairs" and is not a manner of manufacture.

In the Federal Court, Branson J., for some reason, held that the value of the invention was only to those whose assets are ultimately protected - and possibly to their professional advisers. She held that the performance of the invention will not add to the economic wealth of Australia or otherwise benefit Australian society as a whole. For that reason, in her view, the invention was not proper subject matter for a patent. Her honour went on further to state that the claimed invention is a method by which the owner may be insulated from the operation of laws intended to serve the public interest. She held that a court of law must assume that the performance of the invention will not advance the public interest but merely advance private interests. Finally, her honour held that the social cost of conferring on the invention the protection of a patent would not be counterbalanced by any resultant benefit to the public.

This was rather strange reasoning given the wealth of precedent behind her and may in fact have provoked Grant's folly in pursuing an appeal.

Relevant Principles Discussed by the Court

It is not really necessary to get too involved here. Most of these principles are set out in my previous post of 29 May 2006. Some things are worth pointing out, though.

The court set out an interesting review of the historical development of patentable subject matter, which should be read. Their honours point out that business, commercial and financial schemes as such have never been considered patentable in the same way that the discovery of a law or principle of nature is not patentable. For example, in Re Cooper's Application for a Patent (1901) 19 RPC 53, it was held that "you cannot have a Patent for a mere scheme or plan - a plan for becoming rich; a plan for the better government of a State; a plan for the efficient conduct of business."

Business Systems

Mr. Grant contended that the invention is a "business system" as discussed in a report by the Advisory Council on Intellectual Property. The Government was in favour of the patentability of business systems. However, their honours held that the question was not whether a system for use in business was or was not patentable. Rather, they held that an invention must comply with the requirements of the Patents Act and the fact that it may be called a business method does not prevent it being proper subject matter for a patent.

Manner of Manufacture

Mr. Granted argued, for various reasons, that the invention was a "manner of manufacture". As discussed in my post of 29 May 2006, in order for the claimed method or process to be a "manner of manufacture", the method or process must result in an "artificially created state of affairs".

Their honours pointed out that the method of Mr. Grant's invention does not produce an artificial state of affairs, "in the sense of a concrete, tangible, physical or observable effect". They stated further that it was quite different from the invention in Welcome Real-Time SA v Catuity Inc [2001] FCA 445 (17 May 2001), discussed in an article linked in my post of 17 July 2006. The method disputed in that case involved components such as smart cards and point of sale terminals, and produced tangible results. According to their honours, Mr Grant's method resulted in "at best an abstract, intangible situation, namely that a hypothetical, unsecured creditor who recovered judgment against a user of the method could not levy against the user's assets to the extent they were subject to the charge."

Their honours emphasized that a physical effect in the sense of a concrete effect or phenomenon or manifestation or transformation is required. By contrast, Mr Grant's alleged invention resulted in no physical consequence at all.

Must the Result be an Application of Science and Technology?

This question arose because in Re Peter Szabo and Associates Pty Ltd (2006) 66 IPR 370, reported in my post of 1 July 2005, the Commissioner had expressed the opinion that "an artificially created state of affairs...required the application of science or technology in some material manner."

Their honours stated that they were "not sure that this is correct". They supported this by stating that to erect a requirement that an alleged invention be within the area of science and technology would be "to risk the very kind of rigidity which the High Court warned against" in the NRDC case, which forms the basis of the "artificially created state of affairs" requirement and is discussed in my post of 29 May 2006.

Justice Branson's Diversion

As mentioned above, Justice Branson had set out some other "interesting" reasons why the patent should not be granted. These included her conclusion that "an invention should only enjoy the protection of a patent if the social cost of the resulting restrictions upon the use of the invention is counterbalanced by resulting social benefits."

According to their honours, however, it was not relevant whether or not the public interest is advanced. Also, they held, the Court is not in a position to determine the balance between social cost and public benefit. Their honours rubbed it in a little by stating that the court's predecessor had already determined the balance in 1623, by rewarding innovation with time-limited monopoly.

Conclusion

Mr Grant's asset protection scheme is not unpatentable because it is a "business method". The proper assessment is carried out by applying the principles that have been developed, irrespective of the area of activity. Their honours repeated the requirement that there be some "useful product", some physical phenomenon or effect resulting from the working of a method for it to be properly the subject of a patent.

17 July 2006

Articles from Presentation

For those of you who may be interested, an article on software and business methods can be accessed here and an article on Australian patent practice can be accessed here. These articles are based on presentations that I delivered at the American Bar Association's IPL Summer Conference and to various firms and corporations in Silicon Valley. Your comments on these articles would be particularly appreciated.

13 July 2006

Some Thoughts on Software Patent Specifications

Introduction

I have recently returned from a trip to the United States, where I delivered a presentation at the American Bar Association's summer conference covering Australian patent practice. I also spoke at a number of firms and corporations in Silicon Valley. During the trip, I had a number of interesting discussions with practitioners and attended a very useful lecture on drafting software patent specifications.

The purpose of this post is to give you some of my thoughts on preparing software patent specifications in the United States and Australia. Recent developments in the United States has made the need for searching and proactive patent preparation clear.

Changes at the United States Patent and Trademark Office (USPTO)

The United States Commissioner of Patents and Trade Marks, John Doll, was a guest speaker at one of the conference lunches. He spoke of an accelerated examination procedure that was about to be implemented by the USPTO. More information concerning the accelerated procedure can be found
here.

In short, the USPTO is prepared to guarantee that a first office action (patent examination report) will issue within 12 months from filing the application provided a number of conditions are met. The most important condition is that the Applicant must carry out a search. The results of the search must be provided in a statement at the time of filing and the Applicant must explain how the invention is distinguished from the results of the search. The specification can also include no more than twenty claims, and only three of those claims can be independent. The conditions are a lot more extensive and amount almost to the Applicant carrying out its own examination (surprisingly, the USPTO will not be reducing its fees if this procedure is selected).

I have mixed feelings about this new procedure, but I understand the problem faced by the USPTO. It is simply caving in under the massive influx of patent applications particularly in the software / business method category. Those practitioners that have been involved in the preparation of specifications covering software products will appreciate the difficulties involved with identifying the invention and then describing it. It would follow that examining such patent applications would also be extremely difficult and time-consuming. As a result, applicants in the software / business method categories have been told recently that it may take up to five years or more for a first office action to issue. Given the fast pace of software development, this is simply an untenable position.

Making it Easier

During the conference, one of the speakers gave some tips on how to speed up the prosecution of a United States software patent application.

First of all, a search is strongly recommended. The search results should be described in a background section so that the Examiner can understand the field of the invention. A good background also helps to provide a point of reference when defining the invention. It is usual practice for the background section to describe the disadvantages associated with the known technology. If the background section contains the search results, such a description can be made all the more meaningful. Practitioners need to be careful here, of course. An in-depth description of prior art is useful but can also provide material for a revocation action.

Secondly, the specification should point out why the invention is patentable in light of the search results. Clearly, this should help the Examiner to understand the relevance of the elements or integers of the claims and make his or her job easier. Some practitioners may disagree, pointing out that this may provide a road map for a revocation action against the patent. That may be so, but I believe that the advantage of having an decent and relatively speedy examination outweighs any prejudice against the applicant resulting from this approach.

Thirdly, the specification should employ simple and clear language. Words that are unique to the field of the invention should be defined in a "Definitions" section of the patent specification. The purpose of this is to avoid the need for the Examiner to waste time looking up meanings of words. Practitioners need to make sure that all words used are free of any ambiguity. If there is a possibility of ambiguity, either a definition should be provided or a different word used.

Fourthly, and controversially, the speaker mentioned that practitioners should limit the number of claims. The scope of the search and examination carried out by the Examiner is determined by the claims, particularly the number of independent claims. By limiting these, the scope of the search can be determined more quickly by the Examiner.

Fifthly and finally, as many drawings as possible should be filed together with the patent application. In fact, if possible, it should be possible to comprehend a working example of the invention just by looking at the drawings. In the case of software patent applications, the drawings should include system diagrams, flowcharts and block diagrams that illustrate preferred embodiment/s of the invention to the reader. This will make life a lot easier for the Examiner.

Tricks up our Sleeves?

In Australia, we have the option of requesting, at no extra cost, expedited examination. This will speed up examination. If we follow the guidelines for making it easier described above, examination should proceed quite rapidly.

Our innovation patent system provides a neat mechanism for obtaining early enforcement of patent rights. I have discussed our innovation patent system in my post of
4 May 2005. While that post points out the dangers of the system, an innovation patent can be very useful when used in concert with a standard patent application.

It is possible to file a divisional patent application for an innovation patent from a standard patent application. This is useful if the Applicant wishes to enforce patent rights before the standard patent is granted, due to the relatively quick prosecution of the innovation patent application. Provided the alleged infringement can be described in the specification of the parent application, the claims of the innovation patent can be crafted to capture the infringement. This is an extremely effective tool and we have used it on a number of occasions to enforce clients’ patent rights during prosecution.

Concluding

The filing of software patent applications has become a key strategy of most R&D firms in Silicon Valley. At the moment there is a quest among many of these firms to develop a way in which patents can be granted quicker while still being subject to a decent examination and thus a stronger presumption of validity. Instead of simply complaining about the problems of backlog and poor examination, practitioners can and should find ways to engage the problem in a constructive manner.

03 July 2006

Shedding Some Light on Business Methods

Introduction

In my post of 29 May 2006, I discussed how Australia has become comfortable with software patents. This post builds on that by setting out how we deal with "Business Methods" in Australia and has particular implication for patents covering e-commerce products.

In that post, I discussed how our law developed to deal with the term "manner of manufacture". In this post, I have a look at one of our most significant cases in connection with e-commerce. I then carry out our tests on a patent claim considered in the well-known United States case of Ex Parte Lundgren, that was heard by the United States Board of Patent Appeals and Interferences on 20 April 2004. I also consider a granted claim and explain why it was granted.

A Significant Case

In Welcome Real-Time SA v Catuity Inc [2001] FCA 445, Heerey, J. had the opportunity to consider a process for generating a reward system or buyer's loyalty scheme. This case is worth reading, especially the section dealing with "Manner of Manufacture" which has a discussion of the NRDC case mentioned in my previous post.

The relevant claim reads as follows:

1. Method of processing coded information during a purchase or payment operation by a customer, holder of a card with a chip, at a trader's, in which the contents of the memory of the chip card are read and a coupon is or is not printed on the basis of the information arising from the contents of said memory, characterized in that, with the memory of the chip card including a first identification file, termed the Member file, identifying the card-holding customer, a second accounting file, termed the Points file, and a third file, termed the Behaviour file, relating to the behaviour of the card holder towards the user trader or traders,

a specified algorithmic processing is performed dependent, on the one hand, on the date of the operation and, on the other hand, on the information contained in said files, including the Behaviour file, the algorithmic processing including a step of incrementing or decrementing the Point [sic] file by a predetermined number of points depending on the frequency and/or the nature of first, second or xth visit by the card holder over a time period of specified duration, the coupon is printed only if the number of points contained in the Points file is greater than a specified value then data is written to the Points file, new information is written to the Behaviour file, and said coupon is or is not printed on the basis of the result of said algorithmic processing.

Justice Heerey found the United States matter, State Street Bank & Trust Co v Signature Financial Group 149 F 3d 1368 (1998) persuasive. Those familiar with this matter will recall that it concerned a patent for a data processing system for implementing an investment structure. Mutual funds ("Spokes") pooled their assets in an investment portfolio ("Hub") organised as a partnership. This system allowed for consolidation of costs of administering the funds combined with the tax advantages of a partnership.

Previous decisions of the United States Supreme Court had held that mathematical algorithms are not patentable subject matter because they are merely abstract ideas. However, in State Street, it was held that the transformation of data representing discrete dollar amounts by a machine through a series of mathematical calculations into a final share price constituted a practical application of a mathematical algorithm because it produced "a useful, concrete and tangible result" in the form of a final share price momentarily fixed for recording and reporting purposes.

His honour held that the reward system was not a “business method” in the strict sense. Rather, he held that it produced an artificial state of affairs in that cards could be issued for different loyalty programs of different traders as well as different programs offered by the same trader. All this could be done instantaneously at each retail outlet. The result was beneficial in a field of economic endeavour - namely retail trading - because it enabled many traders (including small traders) to use loyalty programs and thereby compete more effectively for business. Such competition was beneficial to consumers, both in the general sense and in the sense that they could obtain benefits in the form of discounts and free goods and services.

His honour gave an interesting example of what would not be patentable. That crackpot genius, Henry Ford, had the bright idea of stipulating that suppliers of components for the Model T Ford supply the components in crates fabricated with planks of suitable dimensions for use as floorboard planks for the Model T. According to his honour that business practice would not constitute patentable subject matter as it did not result in an artificially created state of affairs.

A Granted Claim

Set out below is a claim from Australian patent no. 732272 for "An Advertising System" that was granted on 12 April 2001:

A method of attracting traffic to a network site, including:
publishing a network site for a plurality of radio advertisers, the site having an index to the advertisers; and
advertising the address of the site on radio cooperatively with respective advertisements of the advertisers.

Clearly, by carrying out the steps of that method, we end up with "an artificially created state of affairs" in the network site with an index.

The Ex Parte Lundgren Claim

In the United States, the Board of Patent Appeals and Interferences dealt with a claim that had the following steps (I have removed large portions for the sake of clarity):

1. A method of compensating a manager.

2. Choosing an absolute performance standard from a set of absolute performance standards.

3. Measuring an absolute performance of each firm.

4. Determining a performance comparison base.

5. Comparing said measurement.

6. Determining a relative performance measure.

7. Determining the managerial compensation amount.

8. Transferring compensation to said manager.

If we apply the NRDC test to the above steps, all we end up with is a well-compensated manager. I don't think that could be regarded as an "artificially created state of affairs".

A Last Word

In order to assess a claim for patentable subject matter, one has to consider the result of the claim. I am tempted to refer to some form of “transformation” of hardware. However, it is important to stick to the terminology of the courts. There has to be a change in the state of affairs and that change has to be artificially created. It follows that if the specification makes no reference to hardware of any sort, there cannot be an “artificially” created state of affairs and the relevant claim would not be suitable for Australia.