Journal of Law, Information and Science
|
|
Home
| Databases
| WorldLII
| Search
| Feedback
Journal of Law, Information and Science |
|
GAYE MIDDLETON[*]
In the last five years, development and use of the Internet has increased at a phenomenal rate. Operation of the Internet depends on a wide range of technologies. With the shift in Internet use from primarily non-commercial to commercial use, there has been a corresponding shift among creators of Internet technology from making that technology freely available to protecting their intellectual property interests in that technology. Patents have become a powerful means of protecting a creator’s interests in Internet technology. While it is generally recognised that Internet technology is patentable, patenting of two types of Internet inventions, namely software and methods of doing business, remains controversial. This paper examines patenting of these two types of Internet inventions from a historical perspective, culminating in the recent US State Street decision. The discussion focuses on the US, where most recent developments have occurred, with comments on the position in Australia and the European Union. This is followed by a discussion of arguments for and against patenting of Internet inventions, which reveals that most criticisms of Internet patents stem from the patent system itself. Accordingly, the appropriate solution to many of the problems arising from Internet patents is reform of the patent system rather than the abolition of patent protection for Internet inventions.
State Street may have put software patents on the map, but the surrounding region is raw and unchartered.[1]
In the late 1960’s, the United States (“US”) Department of Defense commissioned the development of a network of computers that could communicate remotely with each other, and that had built-in redundancy, so that if certain parts of the network ceased to function, the remainder of the network would remain operational. This was the beginning of the Internet. However, use of the Internet only expanded to the general population in the early to mid 1990’s, when graphical user interfaces were developed and commercial use of the Internet was first authorised. Since that time, the number of Internet users has grown at a phenomenal rate.[2]
Operation of the Internet depends on a vast array of hardware and software. With the shift in Internet use from non-commercial to commercial use, there has been a corresponding shift among creators of Internet technology from making that technology freely available to protecting their intellectual property interests in that technology. While certain aspects of Internet technology can be protected by trade secrets and copyright, patents have become a powerful means of protecting a creator’s interests in Internet technology. The public nature of the Internet makes trade secrets difficult to keep, while copyright only protects the expression of an idea in a particular form. By contrast, patents enable creators of Internet technology to protect an idea embodied in an invention, even against persons who are not aware of the patent.
While there is general global recognition that Internet technology is patentable, patenting of two types of Internet inventions remains controversial; namely, software and methods of doing business. Some sectors of the Internet community are against patenting of Internet technology in the belief that it will stifle development of the Internet, with patent owners controlling future development and holding the rest of the Internet community to ransom by demanding crippling licence fees for the use of patented technology.
This paper examines Internet patents and discusses arguments for and against patenting of Internet inventions.[3] The discussion focuses on the US, where most recent developments have occurred, with comments on the position in Australia and the European Union.
To be patentable, the US Patent Act[4] requires that inventions must be:
(a) statutory subject matter;
(b) new;
(c) non-obvious; and
(d) useful.
Similarly, section 18 of the Australian Patents Act 1990 (Cth) (“Patents Act”) provides that a patentable invention is an invention that:
(a) is a “manner of manufacture”;
(b) is novel;
(c) involves an inventive step;
(d) is useful; and
(e) was not secretly used prior to the priority date of a patent claim.
Potentially patentable Internet inventions include:
(a) communications protocols;
(b) data compression schemes;
(c) encryption and security procedures;
(d) hardware;
(e) computer software including servers, browsers, and search engines;
(f) user interfaces; and
(g) methods of conducting business online.[5]
These categories of inventions are not mutually exclusive.
A survey of the case law and media coverage relating to patents for these categories of inventions reveals that most controversy has surrounded whether software and methods of doing business are patentable subject matter. Accordingly, this paper focuses on the development of patent law principles relevant to these two types of Internet inventions. As the most rapid development in this area of the law has taken place in the US, these principles will be examined primarily from the US perspective.
For many years, the US position was that software was not patentable subject matter within the meaning of section 101 of the Patent Act: namely, it was not a “new or useful process, machine, manufacture or composition of matter, or any new and useful improvement thereof.”[6] Case law had established that laws of nature and abstract ideas, including mathematical algorithms, could only be discovered, not invented, and hence were not patentable subject matter.[7] As software involves the use of mathematical algorithms, US courts for many years held that software was not patentable subject matter.
In Gottschalk v Benson,[8] the US Supreme Court held that a method of converting coded decimal numbers into binary numbers could not be patented, as to patent it would create a monopoly in a mathematical algorithm. In the court’s opinion, such a patent would be equivalent to patenting a law of nature. While the claims of the invention related to a computer program, the court found that the mathematical procedure involved could be carried out without a computer.[9] This decision was criticised because it was unsupported by the authorities on which it relied, and the court did not adequately justify why mathematical algorithms should be judicially excluded from patentable subject matter.[10]
The US Supreme court next considered the patentability of software in Parker v Flook.[11] The relevant invention was a method of updating an alarm limit used during petroleum refining that involved a computer program. The court found that this method was “a process in the ordinary sense of the word”, and therefore was prima facie patentable subject matter. However, the court believed that it was bound by Gottschalk v Benson, and held that the method was unpatentable subject matter.[12]
In Diamond v Diehr,[13] the US Supreme Court held for the first time that software-related inventions were patentable. The invention involved a method for operating a rubber moulding press, where a computer program calculated the rubber curing time. The court reiterated the Benson definition of an algorithm as a “procedure for solving a given type of mathematical problem”. It held that the invention was patentable, and distinguished Benson and Flook as involving attempts to patent mathematical algorithms, whereas in this case, a patent was sought for a rubber curing process which happened to involve a computer program. In this landmark decision, the court confirmed that “anything under the sun that is made by man” is patentable subject matter.[14]
Following this decision, the US courts developed the following two step test known as the “Freeman–Walter–Abele” test[15] to determine whether a software-related invention is statutory subject matter:
(a) Do the claims recite a mathematical algorithm? If so, then
(b) Is the mathematical algorithm applied to or limited by physical elements or process steps?
If the second question is answered affirmatively, then the claims recite statutory subject matter. The Board of Patent Appeals and Interferences elaborated on this test by holding that a claim only recites an unpatentable mathematical algorithm where it “essentially recites … a method of computing one or more numbers from a different set of numbers by performing a series of mathematical calculations.”[16] The two-part test only applied to claims drafted in method format.[17]
Following pressure from anti-software patent lobby groups, the US Patent Office (“USPTO”) adopted a policy of rejecting software patent applications if there was any doubt regarding whether the claims addressed statutory subject matter.[18] However, a new line of cases soon reversed this trend.
In the Arrhythmia case,[19] the Federal Circuit Court considered whether a method for analysing electrocardiograph signals to determine patient vulnerability to ventricular tachycardia that involved a computer program was statutory subject matter. Newman J held that the invention passed the Freeman-Walter-Abele test because the method related to physical process steps that transform a physical electrical signal into another such signal.[20]
Rader J also found that the invention was patentable, holding that courts should only refer to the statute, 35 USC §101, when determining whether claims are directed to statutory subject matter. In Rader J’s opinion, courts should not impose judicial policy considerations on the “literal and predictable” meaning of the statute, leaving exclusions from the statutory subject matter to Congress.[21] Further, Rader J doubted that mathematical algorithms were akin to laws of nature, as they only appeared in human and numerical processes.[22] Newman J reiterated this view in a dissenting judgment in In re Schrader.[23]
The majority of the Federal Circuit Court in In re Alappat advocated the view of Rader J in Arrhythmia and Newman J in Schrader of reliance on the statute rather than the Freeman-Walter-Abele test.[24] In Alappat, the invention was a rasterizer used to smooth out the waveform display in an oscilloscope. In holding that the invention was patentable subject matter, the Federal Circuit Court found that programming a general purpose computer creates a new machine, being a special purpose computer which performs particular functions pursuant to instruction from software. Accordingly, a programmed computer constitutes statutory subject matter.[25] While the mechanism of operation of such a machine can be represented by mathematical formulae,[26] this “is not a disembodied mathematical concept … but a specific machine to produce a useful, concrete, and tangible result.”[27]
Subsequently, the Federal Circuit Court in In re Lowry[28] decided that a data structure stored on a computer disk or other memory device, or created in memory during or as the result of a computer process, is statutory subject matter.[29] In re Beauregard[30] confirmed that computer–executable instructions embodied on a storage medium are statutory subject matter.[31]
In response to these decisions, the USPTO issued revised Guidelines for Examining Computer-Related Inventions (“Guidelines”).[32] The guidelines apply to inventions implemented in hardware and software.[33] While permitting reliance on the Freeman-Walter-Abele test where claims are “directed solely to a process for solving a mathematical algorithm”,[34] examiners must not begin examining a patent application by determining if it recites a mathematical algorithm.[35]
Instead, examiners must first review the entire specification to determine what invention the applicant seeks to patent and how the claims relate to and define the invention. Only then should examiners consider the specification in light of the statutory requirements, including whether the specification discloses patentable subject matter.
The only categories of non-patentable subject matter identified in the guidelines are:
(a) natural phenomena;
(b) “functional descriptive material” (namely, data structures and computer programs) not recorded on a computer-readable medium;
(c) “non-functional descriptive material”, including musical and literary works or data, regardless of whether or not it is recorded on a computer-readable medium; and
(d) a process consisting solely of mathematical operations without a claimed practice application or which simply manipulates abstract ideas.[36]
As “[i]n the Internet environment, software controls everything”,[37] the patentability of software-related inventions enables a vast range of Internet inventions to be patented. These include methods of data distribution, encryption methods, electronic commerce applications, protocols, servers, browsers and search engines.[38] Most Internet patents will either be for a “process” or a “machine”,[39] and claims can be framed in a wide variety of ways, including:
(a) system format for an entire system;
(b) apparatus format for a single device;
(c) method format, defining the invention as one or more steps in a process;
(d) “means plus function” format, where each claim recites a means together with the function performed by that means; and
(e) data structures or computer-executable instructions embodied on a storage medium.[40]
The advantage of the last claim format is that pure software inventions can be patented provided that the claim is for software embodied in a storage medium. Further, this type of patent can be directly infringed by an unauthorised person selling software embodied in a storage medium containing instructions for computer performance of the recited steps.[41]
With the advent of electronic commerce on a significant scale, the patentability of methods of doing business became an important issue as an increasing number of patents were filed for Internet-based financial and business systems. Prior to the State Street[42] decision, uncertainty surrounded the patentability of methods of doing business. While the US courts issued mixed messages in dicta about whether or not methods of doing business were statutory subject matter, none of the cases were decided on the basis of this issue.
If patentable, methods of doing business fall within the “process” category of statutory subject matter.[43] “Process” is defined as including “a new use of a known process, machine, manufacture, composition of matter or material.”[44] The US Supreme Court interpreted “process” to mean a mode of treatment of certain materials to produce a given result. It is an act or series of acts, performed upon the subject matter to be transformed or reduced to a different state or thing.[45]
The US Supreme Court adopted this “transformation and reduction” test in the software cases.[46] On this reasoning, to be patentable, a method of doing business must be more than an abstract idea, and must apply knowledge to produce a “transformation or reduction”.[47] Arguably, a method of doing business which, for example, involves selling, would result in a number of transformations, including the division of goods into new groups or subsets.[48]
While the US courts prior to State Street considered numerous patents involving methods of doing business, none of these cases was decided on the basis that methods of doing business are not statutory subject matter. Despite dicta in some cases to the effect that methods of doing business are not statutory subject matter, the inventions in question were held to be unpatentable on other grounds.[49]
One case frequently cited as establishing that methods of doing business are non-statutory subject matter is Hotel Security Checking Co v Lorraine Co.[50] That case involved a patent for a system of printed forms used “for securing hotel or restaurant proprietors … from losses by the peculations of … employees.” The court held that this system lacked novelty, and therefore was not patentable. In dicta, the court stated that “a system of transacting business disconnected from the means for carrying out the system is not … an art.”[51] However, this does not mean that methods of doing business are not patentable. The court also stated that business methods for the internal use of a business (eg accounting) lack novelty. This statement does not preclude patenting of methods of doing business involved in electronic commerce, as these methods arguably involve distribution and delivery of goods and services to the public (that is, an external process).[52]
In one case, the Court of Customs and Patent Appeals refused to speculate whether methods of doing business are patentable subject matter, but stated that “some methods of doing business might present patentable novelty.”[53] Accordingly, the court did not rule out methods of doing business as being statutory subject matter.[54]
A cash management account system was held to be statutory subject matter in Paine, Webber, Jackson & Curtis v Merrill, Lynch, Pierce, Fenner & Smith.[55] The claims recited “a method of operation on a computer to effectuate a business activity”.[56] This case is not the sole or first example of a patent relating to a business method,[57] demonstrating that the so-called “methods of doing business” exception to statutory subject matter is a myth.
The Board of Patent Appeals and Interferences in In re Schrader held that an application to patent a method for conducting an auction for a number of related items should be rejected for reciting an algorithm and a method of doing business, both of which the Board held were non-statutory subject matter. On appeal, the Federal Circuit Court upheld the rejection on the basis that the claims merely recited an algorithm.[58]
Newman J in dissent considered the patentability of methods of doing business, and held that the USPTO had incorrectly interpreted cases such as Hotel Security Checking Co v Lorraine Co as establishing that methods of doing business are not statutory subject matter. The judge criticised this so-called exception, stating that previous cases had rejected business method patents on other grounds. Newman J also approved of the Merrill Lynch decision.[59]
While methods of doing business were not definitely excluded from the categories of statutory subject matter, the patentability of business methods remained unclear.[60] In recognition of this ambiguity, the 1996 USPTO Guidelines directed examiners not to categorise claims as methods of doing business, but to treat them like any other process claims.[61]
The Federal Circuit Court’s decision in State Street[62] will have an important impact on the prevalence of Internet patents, as for the first time, the court confirmed that:
(a) the Freeman-Walter-Abele test is irrelevant when determining the patentability of software; and
(b) methods of doing business are patentable, provided that the other statutory criteria are met.
Signature Financial Group Inc (“Signature”) owns US Patent No 5, 913, 056 for a data processing system for a “hub and spoke” financial services configuration. This system involves mutual funds (“spokes”) pooling their assets in an investment portfolio (“hub”) organised as a partnership. Using this system, mutual funds administrators can monitor and record financial information and make the necessary calculations to maintain the “hub and spoke” financial services configuration. Administrators can use this system to determine each spoke’s share in the hub, taking into account daily changes in the value of the hub and each spoke’s assets. This facilitates the sale of shares in the spokes to the public at a price that accurately reflects the spoke’s percentage interest in the hub.
State Street Bank & Trust Co (“State Street”) began negotiating a licence from Signature to use the hub and spoke patent. When these negotiations broke down, State Street sought a partial summary judgment that the hub and spoke patent was invalid for failure to claim statutory subject matter under 35 USC §101. The Massachusetts District Court ruled in State Street’s favour, and Signature appealed.
The Federal Circuit Court unanimously reversed the District Court’s decision on the basis that the hub and spoke patent claimed statutory subject matter. In its judgment, the court made the following findings:
The court held that mathematical algorithms are not patentable subject matter to the extent that they are merely abstract ideas. However, if algorithms form part of an invention that produces “a useful, concrete and tangible result”,[63] then to that extent, they are patentable. Signature’s invention used a mathematical algorithm, formula or calculation to transform data, producing “a useful, concrete and tangible result”; namely, a final share price momentarily fixed for recording and reporting purposes.
Further, the court held that the Freeman-Walter-Abele test “has little, if any, applicability” to determining whether claims disclose patentable subject matter. Citing Alappat,[64] the court held that application of this test could be misleading, because an invention can be statutory subject matter even though it employs non-statutory subject matter such as a law of nature, natural phenomenon or abstract idea. This finding confirmed the trend in previous decisions.[65]
The court held that the District Court erred in construing the claims as process claims. Where, as in this case, “machine” claims are framed in “means for function” format, they may only be construed as “process” claims if the specification does not contain supporting structure corresponding to the “means” elements. This was not the case with the hub and spoke patent.
Additionally, the court held that once it is determined that a claim falls within the categories of statutory subject matter, it is unnecessary to determine which category the claim relates to. Instead, the focus should shift to whether the claim satisfies the remaining criteria for a patentable invention.
For the first time, the court confirmed that methods of doing business are not inherently unpatentable. Describing the business method exception as “ill-conceived” and “no longer applicable”, the court held that methods of doing business should be subject to the same requirements for patentability as any other method or process. The court approved of the USPTO Guidelines, which take this approach. It also pointed out that neither the Federal Circuit Court nor the CCPA had ever found an invention unpatentable on the basis of this exception.
The court noted the District Court’s remarks that the hub and spoke patent was sufficiently broad “to foreclose virtually any computer-implemented accounting method necessary to manage this type of financial structure.”[66] However, the court stated that whether claims are too broad is not relevant to the question of whether those claims recite statutory subject matter.
While the State Street decision does not end the litigation between Signature and State Street, it is likely to have lasting repercussions on Internet patents. State Street has confirmed that two perceived barriers to patenting of Internet technology – namely, the “mathematical algorithm” and “business methods” exceptions to statutory subject matter – are illusory, and that cases decided on the basis of these so-called “judicial exceptions” to statutory subject matter were wrongly decided.
The immediate impact of State Street on Internet patents has been a dramatic increase in the number of business method software patent applications filed at the USPTO.[67] However, the decision has alerted businesses in all sectors, particularly those in banking, insurance and finance, to the importance of obtaining patents as part of their intellectual property portfolio as an offensive and defensive measure to fend off competitors.[68] Steps which businesses should consider taking include:
(a) considering whether patent applications should be filed for software developed by the business;
(b) investigating whether the business has the right to use all current software products without infringing third party intellectual property rights; and
(c) conducting due-diligence on existing software licences to ensure that they contain appropriate warranties and indemnities from the licensor should use of that software infringe third party intellectual property rights.[69]
An indirect but disconcerting aspect of State Street is that the validity and scope of the hub and spoke patent remains undetermined. As judicial developments occur much more slowly than developments in Internet technology, new industries may arise and develop in the shadow of uncertainty as to whether the technology used in those industries is subject to third party patent rights.[70] One commentator speculates that this uncertainty may reduce the amount of venture capital or lengthen the due diligence process for availability of that capital to the Internet and software industries.[71] However, the Internet and software industries were faced by this uncertainty long before State Street.
An important impact of State Street is that it shifts the focus of patent examiners and the courts away from whether business method software is patentable to the appropriate scope of those patents.[72] For example, the District Court in State Street indicated that the hub and spoke patent may be drafted too broadly.[73] This criticism may also be directed at many issued Internet patents,[74] and will be an important consideration for patent attorneys drafting Internet patent applications.
Recently, the Federal Circuit Court of Appeals upheld the State Street decision in AT&T Corp v Excel Communications, Inc.[75] AT&T appealed from a decision that US Patent No. 5,333,184 (“’184 patent”) was invalid for failure to claim statutory subject matter under 35 USC §101 by implicitly reciting a mathematical algorithm. The ’184 patent describes a method of annotating a message record for long-distance telephone calls by the addition of a primary interchange carrier (“PIC”) indicator, which assists long distance telecommunications carriers to provide differential billing treatment for subscribers.
On appeal, the court held that the ’184 patent does claim statutory subject matter. It approved the decisions in Alappat and State Street, and confirmed that a mathematical algorithm may be an integral part of patentable subject matter if the invention as a whole is applied in a useful manner. While Alappat and State Street concerned claims for machines rather than method claims, as in this case, the court confirmed that the scope of 35 USC §101 is the same, regardless of whether the invention is a machine or a process. As the ’184 patent did not claim the Boolean algebra behind the PIC indicator itself, but rather, the application of that algebra “to produce a useful, concrete, tangible result”, the court held that the claimed process was patentable subject matter.
The court rejected the notion that a “physical transformation” is required for a software-related invention to be patentable, holding that the reference to a “physical transformation” in Diamond v Diehr[76] was simply one example of how a mathematical algorithm may be usefully applied. Further, the court held that process claims for software-related inventions do not require physical limitations. The ultimate issue is whether the claims as a whole are drawn to statutory subject matter,[77] rather than whether they are applied to or limited by physical elements, as required by the Freeman-Walter-Abele test. After State Street, the court doubted whether anything is left of the Freeman-Walter-Abele test.
The decision in AT&T v Excel confirms State Street’s findings regarding the patentability of software-related inventions. It also goes beyond State Street by establishing that valid method claims can describe the functional features of the invention without limiting the means of their implementation, clearing the way for broader claims for software-related inventions.
Like the US, the Australian Patent Office (“APO”) for many years rejected patent applications for software-related inventions on the basis that they were not a “manner of manufacture” within the meaning of section 18 of the Patents Act.[78] Software-related inventions were viewed as schemes for operating a known machine, abstract ideas, intellectual processes or algorithms, all of which are unpatentable.[79]
Following the more liberal approach to patenting of software-related inventions in the US and the United Kingdom, the APO issued new guidelines for examination of computer-related inventions in 1986 that adopted the Freeman-Walter-Abele test.[80] While more software patents were issued under these guidelines, the APO experienced similar practical difficulties in applying this test to those experienced by the USPTO.[81]
The IBM case[82] was the first Australian court decision to consider whether software-related inventions are patentable subject matter. The invention concerned was a “method and apparatus for generating curves on computer graphics displays”. One of the objections raised against the patent application was that it claimed a mathematical algorithm, which is not a “manner of manufacture” within the meaning of the Patents Act. An “algorithm” was defined by the court as “a procedure for solving a given type of mathematical problem.”[83]
Rejecting the Freeman-Walter-Abele test, the Federal Court applied National Research Development Corp v Commissioner of Patents[84] (“NRDC”) to hold that the test of whether an invention is a “manner of manufacture” is whether the invention “belongs to a useful art as distinct from a fine art … - that its value to the country is in the field of economic endeavour.”[85] The court held that while the mathematics of the invention were not new, its application to computers to produce an improved curve was commercially useful in the field of computer graphics.[86]
In response to IBM, the APO issued new guidelines for examining computer-related inventions in 1992 which incorporated the IBM test of whether an invention is a “manner of manufacture”.[87] It is now rare for a software patent application to be rejected for not being a “manner of manufacture”.
The IBM decision was followed in CCOM Pty Ltd v Jiejing Pty Ltd.[88] This case involved a cross–claim for revocation of a petty patent relating to the input of Chinese characters into a word processing system. At first instance, the patent was revoked on the basis that it did not claim a manner of manufacture. Referring to a line of United Kingdom cases and Burchett J’s finding in IBM that the invention in that case “involve[d] steps which are foreign to the normal use of computers, and for that reason [was] inventive”,[89] the court found that word processing is not “foreign to the normal use of computers”. Accordingly, the court held that the invention did not overcome any problem in computer operation, and related to the fine arts rather that the useful arts.[90]
On appeal, the Full Federal Court held that the invention was patentable subject matter. It confirmed that “manner of manufacture” is a separate requirement from novelty and inventive step in section 18 of the Patents Act, and that the only test for a “manner of manufacture” is that established in the NRDC decision.[91] Further, the court found that the invention was useful in the economic endeavour of using word processing to assemble text in Chinese language characters.[92]
Accordingly, developments in Australia regarding patenting of software-related inventions have paralleled developments in the US, both regarding the patentability of such inventions and the test for determining patentable subject matter.[93] Patents for software-related Internet inventions are treated no differently to other software-related inventions in Australia, and many of these patents mirror those granted in the US through the medium of Patent Co-operation Treaty applications.[94]
There is no line of authority in Australia relating solely to business method patents. These inventions appear to have been treated no differently to any other invention regarding the test for patentability, and there is no “business method exception” from patentable subject matter in Australia. As Australia’s position regarding the patentability of software-related inventions is similar to the US position, State Street’s findings regarding software-related inventions are likely to be followed in Australia. Accordingly, State Street may have similar repercussions in the Australian financial, insurance and banking industries.
While it is uncertain whether Australian courts will follow State Street in holding that methods of doing business per se are patentable, a software patent relating to a business method or financial system may indirectly protect the business method or financial system itself.[95]
It may be easier to patent software-related Internet inventions in the European Union (“EU”) following two recent decisions of the Board of Appeal (“Board”) of the European Patent Office (“EPO”). Article 52(1) of the European Patent Convention (“EPC”) provides that European patents may only be granted for inventions “susceptible of industrial application”. That Article further states that, among other things, computer programs are not inventions within the meaning of Article 52(1) to the extent to which a patent relates to computer programs “as such”.[96] On the basis of Article 52, the EPO guidelines for examining patents provide that, while computer programs “as such” are not patentable, an invention cannot be denied patentability if it makes a technical contribution to the art simply because it involves a computer program.[97]
In Merrill Lynch’s Application,[98] the British Court of Appeal confirmed that an invention which, as a whole, satisfies the requirements for patentability should not be denied a patent because its novelty and non-obviousness reside in unpatentable subject matter, such as a computer program.[99] Accordingly, the fact that the Merrill Lynch application related to a computerised trading system for stocks and shares did not of itself preclude the invention from being patented.
Despite this decision, the continuing barriers to patenting software in the EU were the exclusion of computer programs as such from the list of patentable subject matter, and the requirement of technical character or effect.[100] However, in two recent decisions,[101] the Board held that computer programs claimed as products can be patented if, when the computer program is run on hardware, it brings about or is capable of bringing about a technical effect which goes beyond the normal physical interactions between software and hardware resulting from the execution of the program’s instructions by the hardware.[102] This may occur where the program solves a technical problem,[103] or where technical considerations are required to implement the invention.[104] Further, the Board held that it is irrelevant when determining patentability whether a computer program itself is claimed, or whether it is claimed as a record stored on a computer-readable medium.[105]
These decisions bring the EU’s position on the patentability of Internet and other software-related inventions closer to the position in the US, Japan and Australia. However, determining what is a “technical effect” is still problematic, and the European Commission intends to address this and other issues regarding patenting of computer software in a directive.[106]
Article 52 of the EPC also provides that methods of doing business “as such” are not patentable inventions. Accordingly, the EU is unlikely to follow the US decision in State Street in this regard. However, as in Australia, obtaining patent protection for computer software embodying a method of doing business can lead to de facto protection for that method.[107]
Patenting of Internet inventions has been widely criticised, which begs the question: should Internet inventions be patentable at all? The Internet community is deeply divided over this issue. However, on closer examination, most criticisms of Internet patents stem from the patent system itself.
A major criticism of Internet patents is that their claims are framed too broadly. Broad claims cover not just the particular invention, but extend outside the scope of the true invention. These types of claim are best demonstrated by example.
One widely criticised patent is CyberGold Inc’s (“CyberGold”) patent[108] for a system which pays Internet users to click on banner advertisements and corporate websites, dubbed “attention brokerage”.[109] CyberGold claims that this patent gives it the “sole right to pay consumers online incentives”.[110] While CyberGold’s competitors have criticised this patent for being too broad,[111] the validity of CyberGold’s patent to date remains unchallenged.
Another broad Internet patent is US Patent No. 5,838,906 granted to Eolas Technology (“Eolas”) for “a system allowing a user of a browser program … to access and execute an embedded program object.” Eolas has commenced proceedings against Microsoft for infringement of this patent by its Microsoft Windows and Internet Explorer applications. An Eolas representative has been quoted as stating that the patent covers “generic technology” that could apply to any product that uses a system to enable Web browsers to access executable programs in Web pages.[112] The breadth of these claims remains untested by the courts.
Another patent criticised by industry experts for being too broad is the patent[113] recently issued to DoubleClick Inc, an online advertising agency, relating to its system of targeted delivery of advertising over networks, including the Internet. DoubleClick’s claims are so broad that they have been described as covering “the basic workings of the entire Internet advertising industry”.[114]
Many Internet patents have broad claims that are not supported by the inventions to which they relate. The only way to counter the criticism of overly-broad Internet patents is for patent examiners to be vigilant when examining patent applications to ensure that the scope of the claims is limited to the invention itself.
Internet patents are often criticised for covering subject matter that is not novel or is obvious. While novelty and non-obviousness are separate criteria for patentable inventions, the same patents are often criticised for failing to meet either criterion.
For example, a time-honoured method of doing business does not become novel simply because it is carried out online. One patent granted for a well-known method of doing business in the online environment that has been widely criticised by online vendors for its lack of novelty is Priceline.com’s reverse auction patent.[115] Internet users use Priceline.com’s system to submit the price that they are willing to pay for items such as airline tickets, and these “bids” are submitted by Priceline to potential vendors. A sale takes place when a vendor accepts a “bid”.[116] The validity of Priceline.com’s reverse auction patent is likely to be tested in patent infringement proceedings recently initiated by Priceline.com against Microsoft.[117] Priceline.com claims that Microsoft’s Expedia.com hotel service, which enables customers to submit their own price for hotel accommodation, infringes Priceline.com’s reverse auction patent.
Internet software and methods of doing business patents are also often criticised for covering obvious “inventions”.[118] Some computer programmers argue that, as it is not unusual for a number of programmers to independently solve a given problem in the same way, using commonly-used techniques, it is ludicrous to grant a patent to one programmer for something that is obvious to many others who are not aware of the patent.[119] However, this argument does not mean that Internet software patents are usually obvious, as the inventive step often lies in identifying the problem itself. Further, in many cases, the solution to a known problem is not easily discovered.[120]
An example of an Internet patent that has been criticised for obviousness is Sightsound.com’s patent relating to “the sale of audio or video recordings in download fashion over the Internet”.[121] This patent has been equated to “patenting … a pencil, a piece of paper and a stamp as a method for writing a letter.”[122]
The only way to ensure that patents are not granted for “inventions” which lack novelty or are obvious is for patent examiners to conduct thorough prior art searches when examining patent applications, and to keep up to date with rapidly emerging new developments in the field. Due to the speed of development of the Internet and the large volume of patent applications filed for Internet inventions, this will continue to be a difficult task.
Many criticisms of Internet patents stem from the fact that there is no comprehensive database of Internet prior art which can be searched by inventors and patent examiners to determine whether an “invention” is anticipated by the prior art. One reason for this is that many Internet patents relate to software and methods of doing business, both of which were believed for many years to be unpatentable. As a result, there are no complete or accurate records of prior art in the field.[123] This problem is aggravated by the fact that, due to copyright considerations, much of the prior art residing in books cannot be incorporated into the prior art databases.[124]
It is also difficult to find all relevant prior art in existing databases because there are no standard keywords that can be used to conduct a thorough and accurate search. One reason for this is that, when software was thought to be unpatentable, software claims were disguised as hardware claims to avoid rejection.[125] Further, there was no USPTO classification system for software-related inventions until 1991.[126] To compound these difficulties, the language used for software-related inventions is continually evolving,[127] and there is a lack of consistency as to which category Internet patents will be delegated to under the USPTO’s complex classification system.[128]
The lack of a comprehensive prior art database is a major contributing factor to the issue of poor quality Internet patents. It also makes it expensive and time-consuming for Internet companies to conduct a prior art search before using technology, filing a patent or defending an infringement suit, and search results are inconclusive.[129] In turn, this affects patent quality by discouraging companies from conducting thorough prior art searches prior to filing patent applications.
However, these difficulties do not justify the abolition of Internet patents. The solution to this problem relies on developing and improving the prior art databases for Internet inventions, and in standardising the classification system for these inventions.
Criticism of Internet patents has been exacerbated by the attempts of some patent holders to extend the scope of their pre-Internet patents to the Internet.[130] An infamous example is E-Data Corp’s (“E-Data”) attempts to extend its patent for distributing electronic data at retail stores to data distribution over the Internet.[131] After purchasing the patent, E-Data commenced patent infringement proceedings against numerous defendants and issued infringement notices against many others.[132] A New York District Court issued an opinion that E-Data’s actions were “an obvious attempt to expand the scope of its patent beyond that which was intended.”[133] While the case was subsequently dismissed for lack of infringement,[134] E-Data has appealed against the District Court’s interpretation of its patent.
Other recent examples of unsuccessful attempts to extend pre-Internet patents to the Internet include Wang Global’s claims that certain browser technologies infringed its closed system videotex patent,[135] and Home Shopping Network’s claims that its patent for distribution and redemption of store coupons applied to coupon distribution over the Internet.[136]
The essence of many criticisms of Internet patents is that such patents are of poor quality, which reflects the poor quality of the examination leading to the grant of those patents. Factors contributing to the poor quality of the examination of many Internet patent applications include:
(a) failure of many applicants to cite all relevant prior art in their patent applications;[137]
(b) lack of a comprehensive prior art database;[138]
(c) lack of expertise or familiarity of some patent examiners with the technology they are examining;[139] and
(d) the current USPTO system of evaluating patent examiner performance, which rewards speedy rather than thorough examination.[140]
These factors, combined with the large volume of Internet patent applications and the rapidly evolving nature of Internet technology, have made the task of patent examiners evaluating Internet patent applications extremely difficult. However, while these factors affect the quality of Internet patents, they do not address the issue of whether Internet inventions should be patentable. Rather, these issues underscore the need for reform of the patent system.
Internet patents have also been criticised for promoting undesirable practices in the Internet industry which stifle competition. Many members of the Internet community believe that large companies are engaging in patent “warfare”,[141] using their large patent portfolios to stifle competition from existing and start-up smaller companies. Smaller companies often cannot afford the licence fees charged to use the patented technology of large companies. As smaller companies are less likely than large companies to be able to afford the legal fees necessary to defend a patent infringement suit, smaller companies must generally exercise greater caution than large companies when developing new products to ensure that any new products do not infringe existing patents. This can be a difficult task, as knowledge of a patent is not required for infringement.[142] Despite the dubious validity of some Internet patents in the portfolios of large companies,[143] many smaller companies cannot afford the legal fees necessary to challenge those patents.[144]
The situation is arguably no better where smaller companies own the patents. Large companies are better able financially to defend a patent infringement suit and to challenge the validity of patents held by smaller companies.[145] However, due to the enormous expense that both sides would incur in a patent infringement suit,[146] even large companies are choosing to pay licence fees rather than defend a patent infringement suit in court.[147]
Another allegation is that large companies stifle competition from smaller companies by cross-licensing their patent portfolios with other large companies. While large companies argue that cross-licensing is beneficial as it allows them “freedom of operation”,[148] it makes it harder for smaller companies to utilise cross-licensed technologies and entrenches potentially invalid patents in the Internet industry.[149]
Related to these concerns is the growing practice in the Internet industry of filing as many patent applications as possible, despite doubts regarding the patentability of an “invention”. The rationale behind this practice is to ensure that someone else does not attempt to patent something first, and to increase cross-licensing bargaining power.[150]
While the above practices are valid causes for concern, the fundamental problem lies with the patent system, not with Internet patents. To prevent misuse of Internet patents, the solution is not to abolish them, but to “clean up the patent system”.[151]
The widespread criticism of Internet patents brings into question the suitability of patents for Internet inventions. One argument against Internet patents is that they are unnecessary, as the Internet would have developed without patent incentives,[152] as demonstrated by rapid Internet development when its use was non-commercial and when software and methods of doing business were believed to be unpatentable.[153] However, this argument can be applied to any industry, and does not justify abolition of Internet patents. Arguably, patent protection promotes creativity across all industries, and the consequent disclosure of inventions benefits society.[154] Further, as Internet inventions are uniquely vulnerable to copying, patenting can dissuade unscrupulous persons from exploiting inventions created through the intellectual and financial investment of others.[155]
Another argument is that it is better not to grant patents for rapidly evolving Internet technologies, as it is too difficult to determine what is truly novel and non-obvious.[156] However, this difficulty is not confined to Internet patents; it applies equally to any rapidly evolving industry, from the Industrial Revolution through to the modern biotechnology industry. Patents have contributed to the development of all of these industries, despite challenges in keeping abreast of new developments. Accordingly, this argument does not justify abolition of Internet patents.
Internet patenting has also been criticised because the period of patent protection is arguably too long[157] for rapidly evolving technologies with a short commercial life.[158] This criticism can be applied to many high technology industries, not just the Internet. However, the appropriate solution is not to abolish Internet patents, but to decrease the period of patent protection for types of Internet inventions with a historically short commercial life.[159]
The above arguments do not justify abolition of Internet patents. Rather, these arguments advocate for patent system reform.
There are a number of arguments for retaining Internet patents. First, Internet technology results from investment of time, money and effort by the inventor.[160] It is only just that the inventor receives some returns on this investment. Patent protection offers these returns by enabling the inventor to control exploitation of their invention. The fact that a lower investment is required for many Internet inventions than in certain other industries[161] does not make them less worthy of patent protection. Fears that patent protection will stifle development of the Internet are unfounded, as there is no reason why patent protection will affect the Internet more than any other rapidly developing industry.[162]
Patent protection arguably encourages further research and development in Internet technology.[163] While not all inventors are motivated by patent protection,[164] this can be said of inventors generally. Further, patent protection can prompt disclosure of inventions that may otherwise have remained secret, which benefits the community through awareness of and access to those inventions. Internet patents can also deter unscrupulous individuals from copying the inventions of others without investing time, money and effort themselves.[165] This is particularly important in the case of Internet inventions, many of which can be copied easily and inexpensively.
As many Internet inventions involve software, there is arguably no adequate protection for those inventions apart from patenting. Copyright protection is unsuitable for computer programs because:[166]
(a) it only protects one form of expression of a computer program, not its functionality;
(b) it classifies computer programs as literary works, when they are functional rather than literary in nature;
(c) lengthy copyright protection[167] is inappropriate for rapidly evolving Internet technologies; and
(d) copyright protection discourages standardisation, which is crucial to the Internet’s operation, by encouraging development of different forms of the same thing.
By contrast, patents protect functionality and are of shorter duration than copyright.
Many critics of Internet patents do not advocate their abolition. Rather, they urge for patent system reform to address the underlying problems manifested through Internet patents.[168]
Internet inventors can improve the quality of Internet patents by:
(a) considering whether their “invention” is worthy of a patent, having regard to the criteria for patentability, the invention’s likely commercial life and the cost and time involved in obtaining and defending a patent;[169]
(b) conducting a thorough prior art search and citing relevant prior art to the Patent Office;[170] and
(c) adequately defining an invention in a patent application without going beyond the scope of the invention.[171]
Risks of Internet patent infringement can be reduced by:
(a) conducting a thorough prior art search before implementing new technology;
(b) taking out adequate patent infringement insurance; and
(c) obtaining an indemnity regarding patent infringement from the supplier when relying on technology supplied by others.[172]
To allay some of the concerns regarding the potential consequences of Internet patents, proposals have been made for a mandatory licensing scheme for Internet patents. One proposal requires Internet patent holders to license those patents for a pre-determined fixed fee, while prohibiting cross-licensing.[173] Another proposal suggests that national governments compulsorily acquire key Internet technologies for a mandatory licensing pool.[174] For a small fee, anyone could use those patents, and patent holders would be compensated from those fees.
Criticisms of patent licensing are closely related to criticisms of Internet patents, and include that:
(a) persons who wish to use a particular technology may not be aware that it is patented;
(b) it is difficult to discover whether a particular technology has been patented due to the lack of a comprehensive prior art database; and
(c) as numerous licences may be required to lawfully implement a single technology, any benefit in implementing that technology may be lost through high licence fees.[175]
Abuses of the patent licensing system are not limited to Internet patents.[176] Accordingly, the solution lies in reform rather than rejection of patent licensing. Aided by anti-trust and trade practices laws,[177] a mandatory licensing scheme with pre-determined fee limits would arguably overcome a number of these abuses.
Another safeguard against abuses of Internet patents is the increasingly powerful role of standardisation. The very nature of the Internet requires that each of its components is interoperable. This requires participants in the Internet industry to adopt universal Internet standards.[178] As many different industry participants have patented various key components of Internet standards, each of these participants must integrate their technologies by freely licensing their patents for the Internet to continue to function seamlessly.[179]
Bodies comprised of web developers such as the Web Standards Project (dubbed “WaSP”)[180] and the World Wide Web Consortium (“W3C”) are working to develop and gain industry support for universal Internet standards. W3C requires its members to disclose all intellectual property interests in Internet technologies prior to those technologies being adopted as part of a standard, and for members to license relevant technologies on a “reasonable and non-discriminatory basis”.[181]
Unfortunately, standardisation can be compromised because:
(a) W3C did not formally require disclosure of intellectual property rights by its members until late 1998, long after a number of standards had been adopted;
(b) a large number of patents and patent applications must be considered when developing a standard; and
(c) licensing rules for technologies adopted in standards are still being developed.[182]
These difficulties are illustrated by W3C’s recent conflicts with Microsoft and Intermind. W3C adopted cascading style sheets as a standard just prior to Microsoft being granted a patent for that technology. Under pressure from W3C, Microsoft has agreed to freely license that patent.[183] Recently, W3C requested Internet users to assist it in finding prior art with which to challenge a patent issued to Intermind for Platform for Privacy Preferences (“P3P”), a privacy standard worked on for several years by W3C.[184] Intermind claims to have informed W3C of its patent application at an early stage. While Intermind has stated that it will endeavour to provide reasonable access to P3P, it intends to defend its patent.[185]
Industry experts have criticised large corporations, claiming that they further undermine standardisation efforts by supporting standards when they do not have a strong presence in a particular field, then altering those standards once they have become “open” to meet their own business needs.[186]
Anti-trust and trade practices laws are likely to play an important part in fostering Internet standards. For example, the US Federal Trade Commission filed an anti-trust suit against Dell Computer Corp when Dell worked on setting a standard for VL-bus technology, and promptly sought to enforce its patents in that technology on adoption of a standard. Dell relinquished its patents after the anti-trust suit was filed.[187]
These recent events demonstrate that anti-trust laws and the economic and practical need for Internet standards may together play an important role in safeguarding the Internet industry against misuse of patent protection.
The recent explosion in Internet use has given rise to a corresponding explosion in the number of Internet patents and patent applications. This increase in Internet patents is also fuelled by the relatively recent recognition that software-related inventions and (at least in the US) methods of doing business are patentable.
Internet patents have been widely criticised. While many of these criticisms are valid, they relate to problems within the patent system rather than to Internet patents. The solution to these criticisms is not to abolish Internet patents, but rather, to reform the patent system and educate its users. Together with the development of and support for universal Internet standards, these actions can lead to patents only being granted for Internet technologies which deserve patent protection, and will promote the traditional role of patents in protecting and encouraging development of the useful arts.
© Gaye Middleton 1999
[*] Gaye Middleton is a solicitor in the intellectual property and information technology group at Mallesons Stephen Jaques, Brisbane. She holds a Bachelor of Laws with first class honours and a Bachelor of Science specialising in pharmacology and biochemistry from the University of Queensland. Gaye also holds a Master of Laws from the University of Queensland, focusing on intellectual property law. Her legal experience includes preparation of a wide range of technology agreements and contracting under the government information technology conditions.
[1] Katz, A ““State Street” May Place Start-ups in Peril – Looking Beyond “State Street” – The Future of Software Patents” New York Law Journal 19/1/99. Available at http://www.nylj.com/tech/011999t2.html
[2] In the US, the number of Internet users rose from an estimated 12.4 million in 1995 to 43.4 million in 1997. (Source: Fram, R, Radin, M and Brown, T “Altered States: Electronic Commerce and Owning the Means of Value Exchange” (December 1997). Available at http://www.hewm.com/search/art.shtml?id=216.) In Australia, the number of Internet users trebled between 1995 and 1998, there being an estimated 1.6 million users in early 1998. (Source: Cook, P Introduction to Fitzgerald, A et al Going Digital St. Leonards: Prospect Media Pty Ltd, 1998 at vi, citing Department of Industry, Science and Tourism Report Electronic Commerce in Australia Canberra, 1998.)
[3] All references in this paper are to utility patents, not design patents.
[4] Title 35 United States Code.
[5] See generally Basile, A “Role of Patents Online”. In Smedinghoff, T Online Law – The SPA’s Legal Guide to Doing Business on the Internet Reading: Addison–Wesley Developer’s Press, 1996 at 246-248; Lee, L and Scott Davidson, J Intellectual Property for the Internet New York: John Wiley & Sons Inc, 1997.
[6] 35 USC §101.
[7] Swinson, J “Software Patents in the United States” (1993) 4 JLIS 116 at 117.
[8] 175 USPQ (BNA) 673 (1972).
[9] Lahore, J Patents, Trademarks and Related Rights North Ryde: Butterworths, 1996 at 639.
[10] Chisum, D “The Patentability of Algorithms” (1986) 47 U. Pitt. L. Rev. 959 at 971. Cited in Lee and Scott Davidson n 5 at 122.
[11] 198 USPQ (BNA) 193 (1978).
[12] Rose, C “Protecting the Internet User Interface”. In Lee and Scott Davidson n 5 at 122-123.
[13] 209 USPQ (BNA) 1 (1981).
[14] Id. at 6.
[15] Based on In re Freeman 573 F.2d 1237 (CCPA 1978), In re Walter 618 F.2d 758 (CCPA 1980) and In re Abele 684 F.2d 902 (CCPA 1982).
[16] Ex parte Logan 20 USPQ2d 1465, 1468 (BPAI, 1991).
[17] Swinson n 7 at 122.
[18] Rose n 12 at 124-125.
[19] Arrhythmia Research Technology Inc v Corazonix Corp [1992] USCAFED 298; 22 USPQ2d 1033 (Fed. Cir. 1992).
[20] Webber, D “Software Patents: A New Era in Australia and the United States?” (1993) 5 EIPR 181 at 183.
[21] Arrhythmia Research Technology Inc v Corazonix Corp [1992] USCAFED 298; 22 USPQ2d 1033, 1042.
[22] Webber n 20 at 183.
[23] [1994] USCAFED 373; 30 USPQ2d 1455, 1460 (Fed. Cir. 1994).
[24] Swinson, J “Recent Software Patent Developments in the United States” (1994) 5 JLIS 281 at 289.
[25] Id. at 287-8.
[26] In re Schrader [1994] USCAFED 373; 30 USPQ2d 1455, 1579 per Newman J.
[27] Id. at 1557.
[28] 32 USPQ2d (BNA) 1908 (Fed. Cir. 1994).
[29] Lee and Scott Davidson n 5 at 69-70.
[30] 35 USPQ2d (BNA) 1383 (Fed. Cir. 1995).
[31] Lee and Scott Davidson n 5 at 69-70.
[32] United States Patent and Trademark Office Examination Guidelines for Computer–Related Inventions, 1996.
[33] Id. at 1.
[34] Ibid.
[35] Id. at 2.
[36] Id. at 8-11, 15-19.
[37] Lee and Scott Davidson n 5 at 93.
[38] Webber, D “Intellectual Property in Internet Software”. In Fitzgerald, A et al Going Digital St Leonards: Prospect Media Pty Ltd, 1998 at 66.
[39] However, provided that the invention falls within one of the categories of statutory subject matter, it is not necessary to decide exactly which category it falls within. (In re Alappat 31 USPQ2d 1545, 1589 per Rader J; affirmed State Street Bank & Trust Co v Signature Financial Group, Inc (23/7/98 Fed. Cir., No. 96-1327) per Rich J).
[40] Lee and Scott Davidson n 5 at 66-69.
[41] Id. at 71.
[42] State Street Bank & Trust Co v Signature Financial Group, Inc (23/7/98, Fed. Cir., No 96-1327).
Available at http://www.law.emory.edu/fedcircuit/july98/96-1327.wpd.html
[43] 35 USC §101.
[44] 35 USC §100.
[45] Cochrane v Deener [1876] USSC 41; 94 U.S. 780, 788 (1877).
[46] Gottschalk v Benson 175 USPQ (BNA) 673 (1972); Diamond v Diehr 209 USPQ (BNA) 1 (1981).
[47] Fuelling, M “Manufacturing, Selling, and Accounting: Patenting Business Methods” (1994) JPTSO 471 at 477, 479.
[48] Id. at 479.
[49] Id. at 480.
[50] 160 F. 467 (2d Cir. 1908).
[51] Id. at 467.
[52] Fuelling n 47 at 482.
[53] In re Wait 24 USPQ 88 (CCPA 1934).
[54] Contra Ex parte Murray 9 USPQ2d 1879 (BPAI 1988), which cited In re Wait as authority for methods of doing business being non-statutory subject matter.
[55] 218 USPQ 212 (D.Del. 1983) (hereafter “Merrill Lynch”).
[56] Id. at 220.
[57] For early examples of “methods of doing business” patents, see Swinson, J “Patents in Cyberspace -Electronic Commerce Patents”. In Fitzgerald, A et al Going Digital St Leonards: Prospect Media Pty Ltd, 1998 at 53. For further examples, see Fuelling n 47 at Appendix B.
[58] [1994] USCAFED 373; 30 USPQ2d 1455.
[59] See discussion in Swinson n 24 at 284-285.
[60] Fuelling n 47 at 492.
[61] United States Patent and Trademark Office Examination Guidelines for Computer–Related Inventions, 1996 at 1.
[62] State Street Bank & Trust Co v Signature Financial Group, Inc (23/7/98, Fed. Cir., No 96-1327).
[63] Citing In re Alappat 31 USPQ 2d at 1557.
[64] Ibid.
[65] Swinson n 24 at 289.
[66] 38 USPQ2d 1530, 1542.
[67] The number of business method software patents filed with the USPTO during 1998 increased by more than 40% over such filings in the previous year. See Sandburg, B “Speed over Substance? The PTO’s expansion is producing record patent issuances – and mounting criticism” IP Magazine March 1999. Available at http://www.ipmag.com/monthly/99-mar/sandburg.html at 3.
[68] Id. at 4; Ellis, W and Chatterjee, A “Shakeout on State Street” IP Magazine November 1998. Available at http://www.ipmag.com/98-nov/ellis.html at 2.
[69] Ellis and Chatterjee n 68 at 8-9.
[70] Katz n 1 at 2.
[71] Id. at 3.
[72] Id. at 4.
[73] 38 USPQ 2d 1530, 1542.
[74] See section 7.1.1 below.
[75] 21/4/99, Fed. Cir., No 98-1338 per Plager, Clevenger and Rader JJ. The United States Supreme Court has refused to review this decision.
[76] 209 USPQ (BNA) 1 (1981)
[77] Citing In re Alappat 31 USPQ2d 1545, 1557 (Fed. Cir. 1994).
[78] For example, see British Petroleum Co Ltd’s Application (1968) 38 AOJP 1020.
[79] Webber n 38 at 61.
[80] Fitzgerald, A and Phillips, S “Patentability of Software in Australia: CCOM v Jiejing” (1994) 5 JLIS 296 at 297.
[81] Webber n 38 at 62.
[82] International Business Machines Corporation v Commissioner of Patents [1991] FCA 625; (1991) 22 IPR 417.
[83] Id. at 419.
[84] [1959] HCA 67; (1959) 102 CLR 252.
[85] Id. at 275, cited by the Federal Court in IBM at 423.
[86] International Business Machines Corporation v Commissioner of Patents [1991] FCA 625; (1991) 22 IPR 417 at 424.
[87] Webber n 38 at 62.
[89] International Business Machines Corporation v Commissioner of Patents [1991] FCA 625; (1991) 22 IPR 417 at 424.
[90] [1993] FCA 653; (1993) 27 IPR 577 at 594.
[91] CCOM v Jiejing (1994) 27 IPR 481 at 509-510.
[92] Id. at 514.
[93] Fitzgerald and Phillips n 80 at 304.
[94] Over 500 Internet related PCT applications were published in Australia in 1997. See Swinson n 57 at 48.
[95] Swinson, J “Patenting Electronic Commerce”. Unpublished paper 7/4/99 at 6.
[96] Article 52(2), (3) European Patent Convention.
[97] Guidelines for Examination in the European Patent Office, Chapter IV.2.3.; Liesegang, E “Software Patents in Europe” [1999] CTLR 48 at 49.
[99] For further information regarding the Merrill Lynch decision and subsequent decisions, see Lloyd, I Information Technology Law London: Butterworths, 1993 at 218ff.
[100] Liesegang n 97 at 51.
[101] Decisions T0935/97-3.5.1 and T1173/97-3.5.1, each dealing with an IBM patent application.
[102] Basinski, E “Computer Program Product Claims Allowed by the European Patent Office - Impact on Software Patent Claiming”. In Morrison and Foerster’s firm newsletter, March 1999, reproduced in Internet Patent News Service 29/3/99.
[103] For example, see Decision T208/84, VICOM System Inc’s Application [1987] OJ 14.
[104] For example, see Decision T769/92, SOHEI/Computer Management System [1995] OJ 525.
[105] Basinski n 102.
[106] Liesegang n 97 at 51.
[107] For example, see Decision T769/92 SOHEI/ Computer Management System, where a computer system for coupling administration and management tasks was held to be patentable subject matter.
[108] US Patent No. 5,794,210.
[109] Richtel, M “Are Patents Good or Bad for Business Online?” New York Times Cyberlaw Journal 28/8/98. Available at http://www.nytimes.com/library/tech/98/08/cyber/cyberlaw/28law.html
[110] Sandburg, B “Madness in PTO’s E-Commerce Method?” IP Magazine 27/8/98. Available at http://www.ipmag.com/dailies/980827.html
[111] Ibid.
[112] Niccolai, J “Microsoft sued over broad Web patent” The Industry Standard 2/2/99. Available at
http://www.thestandard.net/articles/article_print/0,1454,3374,00.html
[113] US Patent No 5,948,061, issued 7 September 1999, entitled “Method of Delivery, Targeting, and Measuring Advertising over Networks”.
[114] Oakes, C “Patent’s Net Result: Nothing?” Wired News 13/9/99. Available at http://www.wired.com/news/print/0,1294,21726,00.html
[115] US Patent No. 5,794,207.
[116] Farmer, J “A battle is brewing for Internet patents” (1998). Available at http://www.insidebiz.com/hamptonroads/law/law100798.htm
[117] These proceedings were filed by Priceline.com on 13 October 1999. See Wolverton, T “Priceline.com files suit against Microsoft” CNET News.com 13/10/99. Available at http://news.cnet.com/news/0-1007-200-851952.html?tag=st.ne.1002
[118] McKeough, J and Stewart, A Intellectual Property in Australia Sydney: Butterworths, 1997 at 246 (in relation to software patents).
[119] Garfinkel, S “Patently Absurd” Wired Magazine July 1994. Available at http://www.wired.com/wired/archive/2.07/patents_pr.html at 2.
[120] Swinson n 7 at 125-126.
[121] Caruso, D “Patent Absurdity” New York Times 1/2/99. Available at http://www.nytimes.com/library/tech/99/02/biztech/articles/01digi.html
[122] Ibid.
[123] Garfinkel n 120 at 6-7; Sandburg n 67 at 5-6.
[124] Garfinkel n 120 at 7.
[125] Id. at 5-6.
[126] Swinson n 7 at 135.
[127] Wayner, P “How can they patent that?” Salon Magazine 9/3/99. Available at
http://www.salonmagazine.com/21st/feature/1999/03/09feature.html
[128] Garfinkel n 120 at 7.
[129] Id. at 2.
[130] Sandburg n 110 at 4.
[131] US Patent No. 4,528,643.
[132] At 7 April 1996, 43 lawsuits had been filed, and infringement notices had been sent to 139 other companies. See Tasty Bits from the Technology Front (online newsletter) 7/4/96. Available at http://www.tbtf.com/archive/04-07-96.html
[133] Jones J in Interactive Gift Express Inc v Compuserve Inc et al 95 Civ. 6871 (BSJ) (SDNY 15 May, 1998).
[134] The case was dismissed on 12 March 1999. See Sandburg, B “E-Data backs off patent claims” The Recorder 2/4/99. Available at http://www.lawnewsnet.com/stories/A438-1999Apr1.html
[135] Luening, E “Wang patent claims dismissed” CNET News 8/5/98. Available at http://www.news.com/News/Item/0,4,21887,00.html
[136] Sandburg n 110 at 4.
[137] Aharonian, G “Does the Patent Office Respect the Software Community?” In Internet Patent News Service 15/5/99.
[138] See discussion at section 7.1.3. Patent examiners have indicated that the new automated patent search system recently installed by the USPTO exacerbates searching difficulties through its faults and lack of “user-friendly” qualities. See Internet Patent News Service 15/11/99.
[139] Sachs, R “Software Patents and Internet Business Methods in the Wake of State Street Bank”. In Fenwick & West’s firm newsletter, Fall 1998. Reproduced in Internet Patent News Service 2/2/99.
[140] Sandburg n 67 at 6.
[141] Caruso n 122 at 3.
[142] Garfinkel n 120 at 11.
[143] Caruso n 122 at 3.
[144] Katz n 1 at 5.
[145] Ibid. An alternative view is put forward in McKeough n 119 at 246, where the authors state that “small companies have begun to terrorise larger concerns through the wielding of patent monopolies.”
[146] These costs are estimated to be an average of $2.5 million for each side. See Lipton, B “Start-up wins e-commerce patent” CNET News 10/8/98. Available at http://www.news.com/News/Item/0,4,25111,00.html
[147] For example, rather than defend the patent infringement suit brought against it by E-Data, IBM chose to licence the technology. See Tasty Bits from the Technology Front n 130; McKeough n 119 at 246-247.
[148] Wayner n 128.
[149] Caruso n 122 at 3.
[150] Lipton n 147, quoting Craig Opperman, patent attorney with Cooley Godward in the US. See also Internet Patent News Service 30/4/99.
[151] See quote by Greg Aharonian, US patent agent, in Wayner n 128.
[152] Garfinkel n 120 at 9.
[153] McKeough n 119 at 247; Swinson n 7 at 132.
[154] Swinson n 7 at 134.
[155] Id. at 133.
[156] Lessig, L “The Problem with Patents” The Industry Standard 23/4/99. Available at
http://www.thestandard.com/articles/article_print/0,1454,4296,00.html; Richtel n 109 at 2.
[157] Patent protection lasts for 20 years in all countries that have implemented Article 33 of the TRIPS agreement.
[158] Lessig n 157; Clark, T “Will patents help e-commerce?” CNET News 26/8/98. Available at
http://www.news.com/News/Item/0,4,25703,00.html
[159] For example, the Australian petty patent system provides patent protection for a maximum of six years (s68 Patents Act 1990 (Cth)). Petty patents are also granted more quickly than standard patents, which may prevent the Internet community from being ambushed by “submarine patents” relating to technologies in common use.
[160] Swinson n 7 at 117.
[161] Garfinkel n 120 at 9.
[162] See section 7.1.7 above.
[163] Swinson n 7 at 134.
[164] Garfinkel n 120 at 9 cites the example of Professor Ronald Rivest, who developed and disclosed RSA public key cryptography three months before applying to patent that system.
[165] Gordon, S “The Very Idea!: Why copyright law is an inappropriate way to protect computer programs” [1998] EIPR 10 at 11.
[166] Id. at 10-12.
[167] For example, Australian copyright protection lasts for the life of the author plus 50 years (s33(2) Copyright Act 1968 (Cth)).
[168] For example, see Garfinkel n 120 at 10; Wayner n 128 citing Greg Aharonian; Internet Patent News Service 4/5/99 at 2.
[169] Webber n 38 at 67; McDonald, D et al “Intellectual Property and Privacy Issues on the Internet” (1996) 55 South Dakota Business Review 1.
[170] Sachs n 140.
[171] See section 7.1.1 above.
[172] Swinson n 57 at 52.
[173] Garfinkel n 120 at 10, citing a proposal by John Preston of the Massachusetts Institute of Technology.
[174] Malamud, C “Building a Park on RSA” Wired Magazine October 1996. Available at http://www.wired.com/wired/archive/4.10/rsa.html
[175] Swinson n 7 at 135–136.
[176] Id. at 138.
[177] Ibid.
[178] Sandburg, B “The Standards Bearers” The Recorder 10/3/99. Available at http://www.law.newsnetwork.com/stories/mar/e031099d.html
[179] Tim Berners-Lee, director of the World Wide Web Consortium, has criticised patents as “getting in the way of a common, universal Web.” See Krochmal, M “Berners-Lee Says Patents Obstruct an Open Web” TechWeb 12/5/99.
Available at
http://www.techweb.com/wire/story/TWB19990512S0013
[180] Further information about WaSP can be obtained at http://www.webstands.org
[181] Sandburg n 179.
[182] Ibid.
[183] Ibid.
[184] Sullivan, J “Volunteer Army to Fight Patent” Wired News 3/5/99. Available at
http://www.wired.com/news/print_version/politics/story/19452.html?wnpg=all
[185] Ibid.
[186] For a discussion of these issues in relation to the current instant messaging protocol dispute, see Festa, P “Why open standards are a myth” CNET News 6/8/99.
Available at
http://news.com/News/Item/Textonly/0,25,39915,00.html
[187] Sandburg n 179.
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/JlLawInfoSci/1999/8.html