Court: Supreme Court of the United States; March 3, 1981; Federal Supreme Court; Federal Appellate Court
Respondents filed a patent application for a process that involves molding raw synthetic rubber into cured products, addressing industry challenges in accurately measuring temperature within the molding press. The claimed innovation involves continuously measuring the mold's temperature and utilizing a computer to adjust cure times based on a mathematical equation. The patent examiner and the Patent and Trademark Office Board of Appeals initially rejected the claims as nonstatutory under 35 U.S.C. 101, which governs patent eligibility for processes. However, the Court of Customs and Patent Appeals reversed this decision.
The Supreme Court held that the respondents' claims were eligible for patent protection. It clarified that a "process" under 101 encompasses any act or series of acts that transform a subject matter into a different state, making industrial processes like the one described historically patentable. While mathematical formulas themselves are not patentable, the respondents were not seeking to patent the formula but rather the application of that formula within their specific process. The Court emphasized that claims should not be dissected into old and new elements; rather, they should be evaluated as a whole. Furthermore, the patent eligibility under 101 is distinct from the novelty and nonobviousness criteria under 35 U.S.C. 102 and 103.
In conclusion, the Court affirmed that when a mathematical formula is integrated into a process that fulfills the patent laws' intent—such as transforming materials—it meets the requirements of 35 U.S.C. 101. The decision was delivered by Justice Rehnquist.
The patent application filed on August 6, 1975, involves a process for molding uncured synthetic rubber into cured products using a mold that shapes the material under heat and pressure. The respondents assert that their method ensures properly cured molded articles, with successful curing dependent on factors such as article thickness, molding temperature, and curing time. Historically, the industry struggled with accurate curing due to the inability to precisely measure molding press temperatures, leading to inconsistent curing times and results.
The respondents' innovation lies in continuously measuring the mold temperature, which is then fed into a computer that recalculates the optimal cure time using the Arrhenius equation. When the elapsed time matches the recalculated cure time, the computer signals the press to open. This approach is claimed to be novel in the field.
However, the patent examiner rejected the claims, stating they pertained to nonstatutory subject matter under 35 U.S.C. § 101, as the computer-controlled steps were deemed conventional. The Patent and Trademark Office Board of Appeals upheld this rejection, but the Court of Customs and Patent Appeals reversed the decision, arguing that the inclusion of a computer does not render a claim nonstatutory if it addresses a practical problem in rubber molding. The Commission of Patents and Trademarks subsequently sought certiorari, asserting the appellate court's decision conflicted with prior rulings, prompting the Supreme Court to grant the writ due to the significance of the issue.
The Court's examination in Diamond v. Chakrabarty, 447 U.S. 303 (1980), focuses on the interpretation of 35 U.S.C. 101, which allows for patents on new and useful processes, machines, manufactures, or compositions of matter. The interpretation begins with the statute's language, adhering to the ordinary meaning of words unless otherwise defined. The Court warns against imposing limitations not explicitly stated by the legislature. Historical context reveals that the Patent Act of 1793 defined patentable subject matter broadly, and the 1952 recodification replaced "art" with "process," reflecting Congress's intent to encompass anything made by man.
The definition of a patentable process has been consistent, viewed as a method of treating materials to achieve a result. A process is characterized as a series of acts performed on subject matter to transform it into a different state or form, equally patentable as machinery. The tools used in a process are secondary to the process itself, which requires specific actions on particular substances in a defined order.
The eligibility for patent protection for processes remains unchanged since the term's introduction in 1952. Previous cases, including Gottschalk v. Benson, reiterate that the transformation of an article to a different state is key to determining a process's patentability. In this case, the respondents' claims involve a physical and chemical process that transforms raw synthetic rubber into a distinct product, qualifying as potentially patentable subject matter under 35 U.S.C. 101.
Respondents present a detailed method for curing rubber, starting with loading uncured rubber into a mold and concluding with the press opening post-cure. Such industrial processes have historically qualified for patent protection. The use of a mathematical equation and a programmed digital computer in some steps does not preclude patentability. However, patent laws exclude laws of nature, natural phenomena, and abstract ideas from protection. Prior cases, such as Parker v. Flook and Gottschalk v. Benson, illustrate that algorithms and mathematical formulas alone are not patentable, as they resemble natural laws. In Benson, claims for an algorithm converting binary numbers were deemed unpatentable, as the algorithm lacked practical application beyond programming a computer. Similarly, Parker v. Flook involved a formula for calculating an "alarm limit," which did not encompass necessary operational details for its application.
In contrast, respondents do not seek to patent a mathematical formula but rather a complete process for curing synthetic rubber, which includes the use of a well-known equation without attempting to monopolize it. Their claims incorporate several process steps that, when combined with the mathematical equation and computer use, enhance the curing process's precision. This integration of technology to reduce curing errors does not render the overall process unpatentable.
A claim that pertains to subject matter that is otherwise statutory does not lose its statutory nature merely because it incorporates a mathematical formula, computer program, or digital computer. In previous cases, such as Gottschalk v. Benson and Parker v. Flook, it has been established that processes are not unpatentable solely due to the inclusion of laws of nature or mathematical algorithms. It is now widely accepted that applying such laws or formulas to known structures can be patentable. Justice Stone clarified that while scientific truths themselves are not patentable, the creation of novel and useful structures utilizing such truths can be.
In this context, Arrhenius' equation is not patentable by itself; however, a process for curing rubber that effectively utilizes the equation may be patentable. When assessing the eligibility of a claimed process under patent law, it is crucial to evaluate the claims in their entirety, rather than dissecting them into old and new components. Even if all individual elements of a process are previously known, their novel combination may still qualify for patent protection.
The argument that novelty should be considered under patent eligibility is misplaced, as Section 101 merely outlines the types of subject matter eligible for patent protection, while the specific conditions regarding novelty are addressed in Section 102. The legislative history of the 1952 Patent Act supports this interpretation, indicating that Section 101 establishes eligibility, while Section 102 details the novelty required for patentability.
The excerpt outlines key aspects of patent law, particularly focusing on the distinctions between different sections of the patent statute. Section 101 addresses what constitutes patentable subject matter, while Section 102 defines novelty and other conditions for patentability. It is noted that respondents’ process may ultimately be deemed unpatentable for failing to meet the novelty or non-obviousness criteria under Sections 102 and 103, respectively. However, this potential rejection does not negate the eligibility of the subject matter under Section 101.
The claims in question are viewed as a process for molding rubber products rather than an attempt to patent a mathematical formula, which is not protectable under patent law. The excerpt emphasizes that a mathematical formula can only be patented if it is applied in a manner that satisfies the patent laws, such as transforming an article. The Court affirms the judgment that the claims do satisfy Section 101.
Justice Stevens, dissenting, argues that the majority misinterprets the patent application and conflates the novelty of the claimed subject matter (Section 101) with its actual novelty (Section 102). He highlights the historical context of patent law's evolution concerning computer technology, noting that prior to 1968, principles like the "mental steps" doctrine would have barred patents on computer programs.
The mental-steps doctrine asserts that scientific concepts or mere ideas cannot be patented. It has historically been used to deny patents for inventions based primarily on mathematical formulas or methods of computation, particularly when a mental operation or mathematical computation constitutes the sole novel contribution. The "function of a machine" doctrine similarly deemed unpatentable any process that merely describes a machine's functionality, a principle rooted in 19th-century court decisions. The definition of "process" established in Cochrane v. Deener indicated that patentable processes must result in a physical transformation of materials.
In response to challenges posed by rapidly evolving technology, notably in computer science, the President's Commission on the Patent System recommended in 1965 that computer programs be explicitly excluded from patent law due to the Patent Office's administrative burdens. Subsequently, the Patent Office proposed guidelines in 1966 that deemed computer programs unpatentable but allowed for programmed computers to be part of a patentable process if combined with non-obvious elements resulting in a tangible outcome. These guidelines were officially adopted in 1968 but were soon overturned by the Court of Customs and Patent Appeals.
In 1968, the court began to repudiate the "function of a machine" and "mental steps" doctrines, arguing that they misinterpreted legal precedents and undermined the patent system's objectives. This shift in interpretation significantly expanded the categories of patentable subject matter, including computer programs. In subsequent cases, such as In re Tarczy-Hornoch, the court overruled prior doctrines, arguing they led to unfair outcomes. The court later articulated broader principles to accommodate computer technology, retaining only a limited prohibition against patents that would monopolize all uses of scientific principles or mathematical equations.
A computer programmed with a new, unobvious program is considered a distinct machine or an improvement over an unprogrammed version. Consequently, patent protection for new computer programs can be sought if claims are framed in apparatus form. The Court of Customs and Patent Appeals (CCPA) later addressed process claims for computer programs in *In re Musgrave*, rejecting the mental-steps doctrine and establishing that any sequence of operational steps could be patentable under Section 101 if within the "technological arts." This standard was further clarified in *In re Benson*, which affirmed that computers are included in the technological arts, but was ultimately reversed by the U.S. Supreme Court in *Gottschalk v. Benson*, ruling that new mathematical procedures executed on existing computers, similar to mental processes, are not patentable.
In *In re Christensen*, the CCPA revisited *Benson*, ruling that a process claim based on a mathematical formula was not patentable, thus reaffirming the point-of-novelty analysis. Subsequent cases like *In re Johnston* and *In re Noll* distinguished between process and apparatus claims, with the former being more restrictive under *Benson*. The majority in *Johnston* argued that *Benson* applied only to process claims, while dissenting opinions cautioned against limiting its implications. In *In re Chatfield*, the CCPA further clarified that *Benson* does not prevent patenting program inventions as processes unless they would monopolize all uses of an algorithm or formula.
Dissenting judges in a legal case referenced prior rulings, asserting that programs for general-purpose digital computers are not patentable. They pointed out that the Court of Customs and Patent Appeals interpreted the Benson decision to prevent patenting process inventions related to algorithms if the claims would entirely pre-empt the algorithm itself. The case In re Flook, which was later reversed, adopted this interpretation. Prior to the Flook decision, a two-step analysis was developed in In re Freeman to assess program-related inventions, examining whether a mathematical algorithm was claimed and whether the claim would wholly pre-empt that algorithm.
The dissent emphasizes the importance of understanding what the inventor claims to have discovered, stating that the outcome of patent litigation often hinges on the judge's interpretation of the patent application. Specifically, the dissent critiques the patent application by Diehr and Lutton, arguing that it lacks any new findings about the synthetic rubber-curing process itself, the raw materials, equipment, or significant process variables. The dissent contends that the Court misinterprets the claims, suggesting that the application primarily focuses on a method for measuring temperature within a mold, rather than a novel discovery. Three arguments are presented against the Court's conclusion: the application does not specify unique temperature-measuring devices, common temperature-measuring devices existed prior to the application, and it is implausible that a process for measuring temperature was newly discovered in 1975.
The Patent and Trademark Office Board of Appeals determined that the primary distinction between conventional molding press operations and those claimed in the application lies in the computational steps associated with controlling the mold heater and the automatic press opening. This finding was upheld by the Court of Customs and Patent Appeals. Diehr and Lutton's application claims a method for using a digital computer to calculate the duration a rubber molding press should remain closed during the curing process. Their application does not present novel advancements in mold instrumentation or the mechanisms for timing and opening the press; instead, it involves updating curing time estimates through repetitive recalculations based on a known mathematical formula, similar to the method sought to be patented by Dale Flook, which involved continuous monitoring and recalculating of process variables during catalytic conversion. The essence of both inventions centers on algorithms programmed on digital computers, with Flook utilizing multiple variables and Diehr and Lutton focusing on a single variable (temperature). The algorithm they use is based on a well-known formula, highlighting that neither the differences in variables nor the nature of the formulas sufficiently support a distinct patentable claim. The document critiques the court's misunderstanding regarding the distinction between the discovery requirement under section 101 and the novelty requirement under section 102, suggesting that if the claimed invention does not constitute patentable subject matter, the novelty issue does not need to be addressed.
The application must be rejected under 101 without considering 102, as the novelty of unpatentable subject matter, such as a formula for updating alarm limits, is irrelevant. The analysis begins with understanding the inventor's claimed discovery, which in this case is a method for programming a digital computer to calculate curing times in a familiar process. This method is assumed to be novel, unobvious, and useful, but the key question remains whether it qualifies as patentable subject matter. If the method is categorized as an "algorithm," as defined in Gottschalk v. Benson and Parker v. Flook, and lacks any additional inventive concept, it is not patentable. Both Benson and Flook established that algorithms are considered "laws of nature" and thus not patentable processes under 101.
Flook also dismissed the idea that an algorithm could gain patentability through specific postsolution activities, deeming them "insignificant" or "token." However, the postsolution activity in Flook was as crucial to the process as the mold-opening activity in the current case, indicating that both activities are significant to their respective industrial processes. Yet, neither activity contributes to the inventive concept claimed by the applicants.
In Gottschalk v. Benson, it was established that a digital computer program solving a mathematical problem is not a patentable process. Parker v. Flook reiterated that adding non-novel postsolution activities cannot make an algorithm patentable. Thus, Claims 1 and 2 of the Diehr and Lutton application, as well as Claim 11, which involves familiar presolution activities, must be rejected. The Court does not argue that the computer program is a patentable discovery, and treating it as part of the prior art confirms that the application lacks a claim for patentable invention, warranting rejection under 101 by the Patent Office and the Board of Appeals.
The decision regarding whether computer programs should receive patent protection is a complex issue that the Court is not authorized to resolve. Past cases, including Gottschalk v. Benson and Parker v. Flook, highlight the difficulty and importance of this matter, with various stakeholders, including patent attorneys and industry representatives, presenting conflicting views influenced by economic interests. Despite arguments for the necessity of patent protection for software industry growth, it has been noted that the industry thrives without it. Concerns persist about the patent system's capacity to handle an influx of applications if computer programs were deemed patentable.
The Federal Government has shown consistent opposition to patenting program-related inventions, a stance reflected by past Commissioners of Patents and Tribunals. Criticism has emerged regarding the clarity of patentability rules and the ambiguous classification of "algorithms" as unpatentable subject matter. This ambiguity raises fears about the potential for broad exclusions of any process from patentability. The author suggests that a clear ruling should be made that no program-related invention is patentable unless it contributes to the art without relying solely on a computer. Additionally, the term "algorithm" should be defined as synonymous with "computer program." The author believes the current case does not adequately address these issues and calls for a reversal of the prior decision due to the invention's reliance on computer usage.
The excerpt outlines a method for operating a rubber-molding press using a digital computer to improve the precision of curing compounds. Key elements include:
- **Variables and Constants**: The method incorporates the natural logarithm of the total required cure time (v), an activation constant (C) unique to each batch, the mold temperature (Z), and a constant (x) based on mold geometry.
- **Rheometer**: A rheometer measures the flow of viscous substances, which helps determine the activation constant for the specific compound being molded.
- **Temperature Management**: The press cools while open, leading to variable reheat times. The respondents propose a solution by using a thermocouple to continuously monitor the mold's temperature during closure.
- **Claims**: The respondents' application includes 11 claims. Notable examples:
1. **Claim 1**: Details a method for operating the press, involving the use of a database for parameters, initiating a timer upon closure, monitoring mold temperature, calculating cure time using the Arrhenius equation, and automatically opening the press when the calculated cure time matches the elapsed time.
2. **Claim 2**: Specifies measuring the activation energy constant with a rheometer and updating the database when there are changes in the molded compound.
3. **Claim 11**: Describes a broader method for manufacturing molded articles, including heating the mold, installing rubber, closing the press, monitoring elapsed time, and maintaining the mold temperature during the curing process.
Overall, the document emphasizes the systematic approach to improving curing accuracy in rubber molding through continuous temperature monitoring and computational calculations.
Repetitive calculations of the Arrhenius equation are conducted during the curing process of rubber in a press to determine the total required cure time (v), defined by the formula ln v = czx, where c is the activation energy constant for the rubber, z is the mold temperature at each calculation, and x is a constant related to the mold's predetermined geometry. Each calculated cure time is compared with the elapsed time from an interval timer, and the press is opened when both times are equal, allowing for the removal of the cured rubber article.
The term "process," as outlined in 35 U.S.C. § 100(b), refers to methods that can include new uses of known processes, machines, or compositions. Historical context from Corning v. Burden clarifies that a process is not patentable by itself but falls under the broader category of "useful art." Processes are distinct from machines, as they involve chemical actions or natural elements, while inventions typically relate to mechanical devices. Patent eligibility for processes, including those for curing rubber, was acknowledged as early as 1854, emphasizing that patents are granted for the method or means of achieving a result, not the result itself.
In Tilghman v. Proctor, the U.S. Supreme Court affirmed that patents can be granted for processes, not limited to machines or compositions, emphasizing that a manufacturing process qualifies as an art under patent law. Charles Goodyear's patent for the vulcanization of rubber was highlighted, where the method of curing rubber with heat, sulfur, and mineral salt was deemed adequate for patenting, despite the apparatus not being patented or relevant.
The definition of "algorithm" is explored, with the petitioner offering a broad interpretation as a step-by-step procedure for achieving a result. This contrasts with the narrower definitions used in previous cases like Benson and Flook, where the Court did not assess the patentability of algorithms under the broader definition presented by the petitioner.
In Flook, the necessity for specific operational parameters in applying a formula was noted, indicating that the patent application lacked sufficient detail on selecting critical variables. The court reiterated principles from Funk Bros. Seed Co. v. Kalo Inoculant Co., stating that discovering a natural phenomenon does not grant monopoly rights unless applied in a novel and useful way. The discussion also touches on the procedural aspect of dissecting claims into old and new elements, suggesting that if a claim's components are all old, it may not constitute statutory subject matter, referencing the precedent set in Flook regarding mathematical algorithms.
The argument presented is flawed because it misinterprets the decision in Flook, which did not rule out the consideration of mathematical algorithms in the Section 101 patent eligibility analysis. Accepting the petitioner's view would render all inventions unpatentable, as they can all be traced back to fundamental principles of nature, which, once understood, could appear obvious to implement. This interpretation would contradict previous rulings, such as Gottschalk v. Benson and Cochrane v. Deener, regarding process patent eligibility.
Section 102 outlines conditions for patentability, specifying that an invention is not patentable if it was previously known, used, published, abandoned, or if the inventor did not independently create it. Additionally, it states that the determination of priority of invention involves considering the dates of conception and reduction to practice, along with the diligence shown by the first inventor.
The claims in Flook extended beyond merely presenting a mathematical formula; they involved a specific calculation producing a new "alarm limit" and applying it to various processes in the petrochemical industry. However, these claims did not encompass every possible application of the formula. The court clarified that limiting the patent's scope to a particular technological application does not make a mathematical formula patentable. The reasoning in Flook aligns with the current analysis, reinforcing that a mathematical formula cannot be patented simply by confining its use.
A mathematical formula in itself is considered nonstatutory subject matter for patent protection regardless of its intended applications. Merely adding post-solution activities to a formula claim does not render it patentable, as established in Flook, which highlighted the lack of practical applications or explanations in the patent application concerning variable selection or processes involved. The dissent argues that the claims merely represent an improved method for calculating curing times, but the claims encompass a comprehensive process for curing rubber that culminates in a novel synthetic rubber product. The patent's eligibility does not hinge on the novelty of individual steps but on the overall innovative combination described. The patent is justified by the novel integration of existing elements rather than their standalone novelty. The dissent's reasoning misinterprets the patent application by excluding non-novel steps, contrary to the principle that a claimed invention can be patentable even if some components are not novel. Additionally, historical context regarding early computers illustrates the evolution of programming methodologies, with significant advances occurring in the 1940s and 1950s.
In 1964, the Copyright Office initiated the registration of computer programs, marking a significant development in intellectual property law. That same year, the Patent Office Board of Appeals issued its first published opinion on the patentability of computer-related inventions in Ex parte King. The landmark case In re Prater, decided in 1969, is recognized as the first major judicial ruling addressing the subject-matter patentability of computer program-related inventions. Prior to this, the Court of Customs and Patent Appeals had handled a similar challenge in In re Naquin, rejecting it based on inadequate disclosure. Subsequent cases, including Gottschalk v. Benson, further shaped the legal landscape regarding the patentability of algorithms and software.
The "function of a machine" doctrine, established in Corning v. Burden, emphasizes that patents cannot be granted for abstract functions or effects, but rather for the specific machines that achieve those functions. This doctrine has been reaffirmed in various rulings over the years, establishing a framework for understanding the limits of patent protection in the context of technological innovation and software development.
The Commission's report highlights significant uncertainty regarding the patentability of computer programs, indicating that direct patent applications for programs have been rejected based on nonstatutory subject matter. Efforts to circumvent these rejections by framing claims as processes or machines have further complicated the issue and should be disallowed. The Patent Office's guidelines were influenced by the mental-steps doctrine and the definition of "process" from Cochrane v. Deener. A dissent by Judge Kirkpatrick and Chief Judge Worley criticized the majority's decision to abandon a well-established rule in patent law, asserting that the doctrine was both equitable and effective, having been implicitly adopted in the 1952 Patent Act.
In the Prater case, a patent application for a method of processing spectrographic data faced rejection based on the mental-steps theory, with the only novel element being an unpatentable mathematical principle. The apparatus claim was similarly rejected due to the assumption that the mathematical principle was prior art, revealing no patentable invention. The Court of Customs and Patent Appeals later rejected the Patent Office's analytical procedure for this claim, a methodology later adopted in Parker v. Flook. Under the "point of novelty" approach, claims that hinge solely on steps involving mental operations are rejected as nonstatutory. Judge Baldwin, authoring the second Prater opinion, disagreed with the "technological arts" standard for process claims and warned that it could allow purely mental processes to be patentable.
Judge Baldwin's interpretation of Musgrave was supported by the Court of Customs and Patent Appeals in In re Foster, highlighting ongoing discussions about the patentability of computer-related inventions. The court's decisions in cases like In re McIlroy and In re Waldbaum drew heavily on Musgrave and Benson. In re Ghiron reasserted a previous rejection of the "function of a machine" doctrine. Although the mental-steps doctrine was not explicitly addressed in Benson, some commentators believe the case implicitly relied on it, while others argue its analysis aligned with the doctrine. The Court of Customs and Patent Appeals later reversed a decision in Dann v. Johnston on different grounds, emphasizing that Christensen did not indicate a return to earlier claim analysis methods. This position was consistently reaffirmed in subsequent cases, including In re de Castelet and In re Freeman. The analysis of claims discussed in the context of Flook was derived from O'Reilly v. Morse and shared similarities with the mental-steps doctrine. The court criticized the Solicitor General's arguments in Flook for obscuring statutory requirements, describing them as "subversive nonsense." Historical context was provided through a reference to Tilghman v. Proctor, which recognized manufacturing processes as art under the law.
Modern rubber curing methods are fundamentally based on Charles Goodyear's discovery of vulcanization, which was first demonstrated 120 years ago by heating a mixture of rubber and sulfur. This process has established a standard for transforming crude rubber into a commercially viable product, enhancing its resistance to heat and cold and increasing mechanical strength. The term "cure," coined by Goodyear for vulcanization, is now widely recognized in the industry.
Respondents assert that their innovative contribution lies in the precise measurement of the mold's internal temperature. Claim 1 of their patent application describes a method utilizing a digital computer to operate a rubber-molding press for precision molded compounds. This method improves upon traditional practices by determining the optimal moment to open the mold, rather than relying on approximations. The patent application includes an "Abstract of the Disclosure," detailing a system where an interval timer records the mold closure time and measures the temperature inside the mold cavity every ten seconds, sending that data to the computer.
The application emphasizes that constant recalculation of mold time based on actual temperatures leads to accurate curing, significantly reducing the likelihood of defective products. A continuous monitoring system is employed to assess mold temperature, utilizing devices like thermocouples located within the mold cavity to capture the temperature at the material's surface. Historical references indicate that the critical nature of the vulcanization process has long necessitated precise instrumentation and skilled management within modern rubber manufacturing environments.
Instruments are available that can indicate, record, or control vulcanization processes by monitoring time, temperature, and pressure, and can automate operations like mold opening and closing. Critics of the Flook decision highlight similarities between the inventions in Flook and Diehr, noting that both involve an initial calculation, continual remeasurement and recalculation, and control based on calculated values. The main difference lies in the claim drafting; the Diehr claims resemble those in Flook, but their analysis suggests that if Diehr's claims were redrafted in Flook's format, they would likely be deemed unpatentable under 35 U.S.C. § 101. Additionally, the Court's suggestion that Diehr's application could face challenges under 35 U.S.C. §§ 102 and 103 is misleading, as the applicants have already addressed all objections except for those related to 35 U.S.C. § 101. Thus, the Court is effectively deciding that the patent will be issued.
The Court of Customs and Patent Appeals previously distinguished between the patentable subject matter requirement under § 101 and the novelty requirement under § 102, as seen in earlier cases such as In re Shao Wen Yuan and Halliburton Oil Well Cementing Co. v. Walker. The lower court erred in its interpretation of Gottschalk v. Benson and Parker v. Flook, reflecting an expansive reading of § 101. The opinion in Flook emphasized that determining the type of discovery sought for patenting must precede assessing its novelty or obviousness. For a claim to be statutory, it must possess substance beyond merely reciting an equation or formula. The Court criticized the approach that focuses on the industrial context of the applicants' discovery rather than the specific claims made. This perspective assumes that as long as claims describe a specific application of a discovery, they qualify as patentable subject matter, which contradicts Flook and is deemed untenable. The principle that laws of nature cannot be patented stems from the understanding that they do not align with the type of discoveries intended for protection under the statute. Additionally, the patent application describes a method involving a digital computer with a data storage bank to improve a process involving time-temperature cure data, illustrating the application of the invention in a simplified context.
The computer continuously queries data storage to determine the appropriate time for curing a set of temperatures, recalculating every second until the elapsed time matches the computed duration. Once aligned, the computer activates the mold-opening device automatically. The referenced Figure 1 in the application is included in the document. The term "algorithm" is explained, highlighting that the sought patent relates to a method for programming a general-purpose digital computer to convert binary-coded decimal signals into pure binary form. The procedures described are generalized algorithms for solving mathematical problems related to numerical conversions, leading to specific program applications.
In the case of Flook, the Court emphasized that post-solution activities do not make an unpatentable principle patentable, regardless of their conventional nature. It was noted that any mathematical formula could have a post-solution activity attached without conferring patentability, as the concept of patentable subject matter is not malleable. The Court of Customs and Patent Appeals erred by conflating the novelty of the claimed process with the novelty of the algorithm itself, indicating that the process must be new and useful, independent of the algorithm's status in prior art. The amicus curiae argument in Flook asserted that the incentive for patents in software development is as crucial as for hardware, reflecting the increased complexity of software provision.
The economic value of patents varies significantly between large financial institutions and smaller software companies, with patents potentially being critical for the survival and financing of the latter. For banks, the existence or possibility of obtaining a patent can influence loan decisions, while for investors, it may be a key factor in investment choices. Clarity regarding patent availability for software inventions could enhance innovation and competition, countering concerns raised by hardware manufacturers. Despite the rapid growth of the software industry, which has outpaced that of the hardware industry, there are objections to patent protections due to challenges in examining software applications, as noted by the President's Commission on the Patent System. Historical precedents indicate that prior to 1968, patentability for most computer programs was largely dismissed, but subsequent changes allowed for reconsideration. Various authorities have suggested that terms related to computer programs and algorithms may be synonymous, indicating a convergence in understanding within the field.
