Think before You Write: Considerations for Drafting Claims of Software Patents

Jinseok Park

JinseokPark

University of Sheffield,

Department of Law

E-mail :

March 2004

Think before You Write:

Considerations for Drafting Claims of Software Patents

JinseokPark[1]

1. Introduction

Over the last few years, the number of patent applications in the area of software-related inventions has increased rapidly. Software patents, including business methods patents and Internet related patents, have become the focus of controversial issue among the software industry and IP practitioners. Many software patents have been granted by the EPO, USPTO[2], and JPO and there have been many law suits regarding patent infringement.

Software contains abstract manifestation rather than clear description. In many cases, flowcharts, pseudo-code and other graphical methods are not enough to illustrate the unique function of the software. Unlike a hardware invention, only a small part of software inventions “reside” within a single machine and this makes it difficult to prove direct infringement of the invention.

Successful prosecution of software-related inventions is heavily dependent upon how well the specification and claims of the patent are constructed and drafted. In preparing a patent application, it is required for the software patent drafter to thoroughly understand both hardware and software from a technological, business and legal perspective. The drafter also needs toconsider the dynamic and ever-changing software industry so that the software patent can be viable and enforceable throughout the term of the patent.[3] Furthermore, given the emergence of business methods inventions, it is crucial to draft claims that can be tailored for the client's business needs. This is why drafting a software patent is difficult and complicated.

One of the biggest challenges for the software patent counsel face is the barrier to patentability. Rules and regulations governing the patentability, and the claim interpretation of software-related inventions may vary in different jurisdictions. In the EPO, USPTO and JPO, numerous court decisions and resultant guidelines have provided more certainty as to the patentability of computer-implemented inventions.

This paper provides the strategies for drafting and prosecuting software patent applications in the EPO, USPTO and the JPO in order to achieve broad interpretation of the claims both now and in the future.

2. A Basis for Drafting a Patent Specification

Although an invention is eligible for patenting as a matter of substance, the invention may be rejected if the form of the patent application expresses the invention incorrectly. Therefore, drafting the substance of an invention in a correct formof specification and claims will be the crucial factor for successful prosecution of software or any other patent applications.

The embodiment or embodiments of the invention need to be detailed to the degree of providing substantive support to the scope of protection desired, giving insight as to how to put into practice the main idea of the invention. US patent law requires inventors describe the best mode of operating the invention at the date of filing the application.[4] This preventsinventorsfrom withholding some essential or advantageous aspect of the invention while seeking to obtain a valid patenta secret. However, the inventor need not update the best mode although better modes are found after the initial filing.[5] In Europe or Japan, there is no requirement that a best mode be disclosed. Even though at least one way of practising the invention must be included in the application in the EPO[6], it is not necessary that the waymust be the best way, or even a good way. Rather, it is sufficient that the invention described in the specification will enable a person skilled in the art to make the invention or to perform the action as claimed.

In preparation for the specification of computer-implemented inventions, a good description of the structure and operation of the program is important. Source code can be used to view the structure and operation of a computer program, and may be included in the specification to ensure that the specification contains enough information to allow the invention to be performed. Although there is no requirement for the patent specification to include the charts or source code listings[7], it is helpful to include some or all of the source code in some cases.

However, it is not necessary to disclose the software program to meet the written description requirement[8] where the functions of the software program were readily apparent from the specification and one skilled in the art could generate the necessary software program to implement the disclosed functions.[9] The level of ordinary skill in the art may be determined by examining related prior art to see which functions of the program are well known. Therefore, it is desirable to determine, before drafting the specification, which parts of the program would be easily understandable and obvious to a skilled programmer and which would be not. It is also important to know the minimum software programs that are needed to fully describe and support the invention.

For complicated software applications, it might be desirable to include function block diagrams that illustrate the hardware and the structure of the software in terms of the files and functional modules stored in the memory when the software is loaded. In many cases, the block diagrams are used to show essential connection among the processes and links between each of the processes and the relevant data structures.[10] Together with the block diagrams, flowcharts are also helpful for showing a software-implemented method as a sequence of steps or processes. In order to make the features of the software clearly understandable, the flowcharts describe the functions of performed steps when the software is executed. The combination of the function block diagram and flowchart will serve as an important visual aide and facilitate the description of particular software, enhancing the understanding on the main features of the invention. In addition to those, data structures, signal waveforms and graphical user interfaces will be useful to show the operation of computer-implemented inventions.

While preparing for a software patent application, a patent counsel needs to work closely with the inventor and software engineer to analyse the data structure and the functions of the software, emphasising some essential technical contributions or advantages of the software. In particular, for some controversial business methods inventions, it is important to ensure that the technical improvements which the invention provides are identified and clearly presented to overcome the possible rejection of the application.

3. Considerations for Drafting Claims

3.1. The Comparison of the Practices in the EPO, USPTO and JPO

Fundamentally the USPTO, EPO and JPO have the common notion that a patentable invention should be new and non-obvious. Moreover, all three Patent Offices seem to agree that the substance of a claim should take precedence over the form.

In June 2000, the three offices published a report[11] on the practices of business method related inventionsin an attempt to harmonise the patent practices among themselves. They reached a consensus on the general concept of patentability for computer-implemented business methods and the result is as follows.

─A technical aspect is necessary for a computer-implemented business method to be eligible for patenting. This technical aspect should be expressly claimed in the EPO and JPO whereas the technical feature may be implicitly recited in the claim in the USPTO

─To “merely automate a known human transaction process using well known automation techniques is not patentable.”[12] The USPTO and JPO stated that an invention is lack of inventive step if it is to systematise existing human transactions in an applied field by means of a computer, and the transactions are such that they could be realised by a routine application of usual system analysis and system design technologies.[13]

Nonetheless, each Office still takes a different stance on the extent of the requisite “technical nature” to be a patentable subject matter.[14] It appears that the USPTO permits the broadest possible scope for a software patent, focusing on the tangible result rather than the technical contribution of the invention. In fact, there is no requirement in the USPTO that computer related inventions have a “technical character” or solve a technical problem. Therefore even a business method itself is patentable if it merely produces a “useful, concrete and tangible result”. In this regard, the USPTO is more liberal in its approach to business method patents than the EPO or JPO.

Despite the amended examination guidelines and the cases which allowed the patentability of computer programs, it seems that the EPO still holds a more conservative position on the patentability of software and business methods inventions than the USPTO. The EPO has the highest “technical nature” requirement. In other words, the “technical character” with a technical contribution is essential to be a statutory invention in the EPO and the technical contribution should be non-obvious. In the EPO, the mere combination of a process and a technical apparatus (i.e. software loaded onto a computer), in its normal sense, is not enough in itself. For example, a scheme for organising a commercial operation, being of a purely commercial nature, would not be protected by patents in the EPO because it would lack technical character even if the scheme runs on a computer.[15]

In contrast, the scheme is patentable in the USPTO the only requirement being that it produces a “useful, concrete and tangible result”. In this regard, it is desirable for IP practitioners to try to bring out the technical features of the invention when preparing patent applications in Europe. Although it remains to be seen, it is more likely for an EPO patent to pass the US examination standard than the other way round because an invention with technical character has an increased possibility to meetthe requirement of the USPTO by producing a “useful, concrete and tangible result”.

As can be seen from the chart[16] below (Figure.1), the patent applications of the EPO that correspond to the granted US business method patents, were not allowed as frequently as in the USPTO especially in 1997.

Although, the data of the EPO is limited to the year 1997, given the conservative practice for business methods inventions of the EPO, it is thought that this trend would have continued even after the year 1998.

The JPO seems to take a middle road.[17]Japan takes a similar legal stance as the EPO, but on a less extreme ground. Although the JPO requires a technical aspect, this can be achieved by drafting a patent claim to specify a computer or by combining a process with a technical apparatus. Accordingly, the actual implementation of patentability appears to be more similar to the United States than to Europe. Nevertheless, there is a great difference in the practices on inventive step between the US and Japan. Unlike the practice of the US, the utility of an invention cannot be used for justifying the inventiveness of the invention in Japan. Instead, technical effect or technical contribution is required as a ground of inventive step and this is quite similar to the practice of the EPO.

To summarise, while business method claims that recite the typical steps of a method may meet the utility requirement of the USPTO, they may not be allowed in the EPO or JPOif there is no expressed recitation regarding computer hardware, technological features, and technical effects. Software patent drafters must understand the essential requirements of patentability of software in each jurisdiction and consider modifications to their patent specifications in accordance with the guidelines and the case law of the country where a patent is being sought.

3.2. Claims with Broad Scope of Interpretation

[1] Considerations for Broad Literal Interpretation

With broad claim interpretation in mind, claim drafters need to enhance the scope of literal infringement. Literal infringement examines whether the alleged device falls exactly within the boundaries of the claims of the patent by comparing the textual meanings of the claims with the features of the challenger’s device. In an attempt to provide a basis for broader claim interpretation, patent counsels should consider the followings.[18]

More than any thing else, it is desirable to avoid using words that can lead to narrow interpretation. For example, statements including words like “critical”, “must”, “required”, “necessary”, “only”, “always” and “never” might be interpreted narrowly. For the same reason, using phrases and arguments like “In order to satisfy the objects of the invention…”[19], “All of the embodiments include…” and “The invention includes…” are likely to lead to narrow interpretation. It is also better to avoid expressions such as “characterised by…” as it limits claims by specific environment or use unless this is desired. Unusually long preambles may not be appropriate for the same reason[20].

Transitional phrases identify the invention and provide the technical context of the invention, linking the preamble and the body of the patent claim. Typical transitional phrases in claims differ by countries. For example, one of the transitional phrases normally used in the US is “…wherein”. Claims in the EPO prefer the phrase “…characterised in that…”.

In the EPO, the Guidelines for Examination states that[21]:

While in everyday language the word “comprise” may have both the meaning “include”, “contained” or “comprehend” and “consist of”, in drafting patent claims legal certainty normally requires it to be interpreted by the broader meaning “include”, “contained” or “comprehend.

There was a case in the Board of Appeals of the EPO that touched on the meaning of “comprising”. The Board held that the term “comprising” may be used legitimately in a claim where the intention is not to define a total composition, but instead to identify one essential constituent thereof.[22] In this context, use of open transitional phrases such as “comprising” are better than closed transitional phrases such as “consisting of” or “which consist of”. This is because the former phrase can be interpreted to encompass all other devices that include all of the elements and limitations of the claim whereas the latter is interpreted as only something that has exactly same elements and limitations of the claim.[23]

[2]Considerations in view of the Doctrine of Equivalents

In the US, inthe case of a patent infringement inquiry, two stages of analysis are normally performed by courts to investigate the infringement. The first stage is literal infringement and the second one is the doctrine of equivalents infringement. If no literal infringement is found, the second step of analysis is applied: the doctrine of equivalents infringement. At this stage, claims are interpreted beyond their strict literal meaning and the scope of the claims may be extended to the features that are equivalentto those literally claimed. This suggests that the exclusive rights provided by a patent are based on, but not exclusively limited to, the language of its claims.

The doctrine of equivalents arose from judicial efforts to stop competitors who would make insignificant modifications and substitutions to the claimed invention, although adding nothing, in order to avoid literal infringement.[24] However, courts have struggled, while using the doctrine of equivalents, to balance the competing public policies of avoiding a fraud on the patent and the need for reasonable certainty for the public regarding the scope of the patentee’s exclusive rights.

Since there is no international standard as to whether to adopt the doctrine of equivalents or not, this issue is left purely to national legislation. While the doctrine of equivalence has been extensively applied to patent cases in the United States, Europe and Japan have been less active in adopting the doctrine for interpreting claims. Given the significance of claim interpretation in patent infringement, the following subsection highlights the claim drafting considerations in view of the doctrine of equivalents. Since the doctrine of equivalents has been predominantly asserted in the US, the following will be especially suitable for US patent applications.

On the whole, claim terms are construed according to their ordinary meaning, unless the patentee has defined those terms “specially” within the specification or file history.[25] Numerous court decisions state that patent claims must be given their ordinary, accustomed, or dictionary meanings unless examination of the specification, prosecution history, and other claims indicate that the inventor intended otherwise.[26] In this context, if possible, using a technical and generic term widely known and used in the art will prevent unintended interpretation of the invention.

When drafting a patent with the doctrine of equivalents in mind, it is necessary to include specific examples of the generic term and to list equivalent structures. It is also advisable to describe components of the invention generally followed by a list of possible alternatives, showing them in drawings. Moreover, it should be emphasised that the listed devices fall within the generic art-recognised term and are considered to be readily interchangeable with the devices of the embodiments.[27] The patent counsel needs to consider whether he can provide a broad definition in the specification for each claim element. For example, a chair can be defined as “a seat for one person, which has a back, usually four legs, and sometimes two arms.”[28]

According to the doctrine of "prosecution history estoppel", an amendment of a claim limitation may restrict that limitation within its literal scope and prevent the scope of the claim from being extended to devices or methods using an equivalent to the limitation. In most cases, such an amendment would have been made to overcome a rejection by a patent examiner based onprior art while prosecuting the patent. This indicates that the subjectmatter that was surrendered during the prosecution of the patent may not be resurrected through the concept of equivalence. Therefore, in view of the doctrine of prosecution history estoppel,it is essential to minimise narrow claim interpretation by avoiding disclaimers or disavowals of literal claim scope.

The patent drafter needs to be careful with the meaning of even small words such as indefinite articles “a” or “an”. The Federal Circuit usually interprets “a” as meaning “one or more”.[29] But, in the KCJ Corp.case[30] it held that sometimes “a” is limited to “one”. It may be necessary to explain the meaning of “or”, whether it means to encompass one or another or both.[31] Furthermore, describing the component of the invention with words like “conventional” or “existing” is risky since it may limit the feature to existing examples.[32]