MARIANA R. PFAELZER, District Judge.
Plaintiff Enfish, LLC ("Enfish") has sued Defendants Microsoft Corporation ("Microsoft"), Fiserv, Inc., Intuit, Inc., Sage Software, Inc., and Jack Henry & Associates, Inc. (collectively, "Defendants") for infringement of two patents: U.S. Patent Nos. 6,151,604 ("the '604 Patent") and 6,163,775 ("the '775 Patent").
Defendants move for summary judgment on the basis that all asserted claims are unpatentable under 35 U.S.C. § 101.
The abstract of the patents provides a clear explanation of the invention. See '604 Patent, Abstract. The patents are directed to an information management and database system. The patents improve upon prior art by employing a flexible, self-referential table to store data. This table is composed of rows and columns. Each column and each row has an object identification number ("OID"). Rows correspond to records and columns correspond to attributes. The intersection of a row and column comprises a cell, which may contain information for a particular record relating to a particular attribute. A cell also may simply point to another record. Columns are entered as rows in the table. The record corresponding to a column contains information about the column, rendering the table self-referential.
The invention includes an index structure to allow for searching. A key word index contains text from each cell in the table. This index is itself stored in the table. Text cells in the table contain pointers to entries in the index, and the index contains pointers to the cells. This arrangement provides for extended inquiries. See '604 Patent, 2:66-3:6.
The Court shall grant summary judgment if there is no genuine dispute as to any material fact, as supported by facts on the record that would be admissible in evidence, and if the moving party is entitled to judgment as a matter of law. Fed. R.Civ.P. 56; see Celotex Corp. v. Catrett, 477 U.S. 317, 322, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986); Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 250, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986). Ineligibility under § 101 is a question of law.
Section 101 of the Patent Act defines patentable subject matter: "Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title." 35 U.S.C. § 101. Section 101 defines four broad categories of patentable inventions: processes, machines, manufactures, and compositions of matter. "Congress took this permissive approach to patent eligibility to ensure that ingenuity should receive a liberal encouragement." Bilski v. Kappos, 561 U.S. 593, 601, 130 S.Ct. 3218, 177 L.Ed.2d 792 (2010) (internal quotation marks omitted). But § 101 does not encompass all products of human effort and discovery. Laws of nature, physical phenomena, and abstract ideas are not patentable. Diamond v. Chakrabarty, 447 U.S. 303, 309, 100 S.Ct. 2204, 65 L.Ed.2d 144 (1980). These exceptions are well established. See, e.g., Chakrabarty, 447 U.S. at 309, 100 S.Ct. 2204; Diamond v. Diehr, 450 U.S. 175, 185, 101 S.Ct. 1048, 67 L.Ed.2d 155 (1981); Parker v. Flook, 437 U.S. 584, 600, 98 S.Ct. 2522, 57 L.Ed.2d 451 (1978) (Stewart, J., dissenting); Gottschalk v. Benson, 409 U.S. 63, 67, 93 S.Ct. 253, 34 L.Ed.2d 273 (1972); Funk Bros. Seed Co. v. Kalo Inoculant Co., 333 U.S. 127, 130, 68 S.Ct. 440, 92 L.Ed. 588 (1948); Le Roy v. Tatham, 55 U.S. 156, 175, 14 How. 156, 14 L.Ed. 367 (1853).
On occasion, the Federal Circuit has described § 101 as a "coarse eligibility filter," barring only "manifestly abstract" inventions and leaving §§ 102, 103, and 112 as the finer sieves. See Ultramercial, Inc. v. Hulu, LLC, 722 F.3d 1335, 1341, 1354 (Fed.Cir.2013), vacated sub nom. Wild-Tangent, Inc. v. Ultramercial, LLC, ___ U.S. ___, 134 S.Ct. 2870, 189 L.Ed.2d 828 (2014). But in the Supreme Court's last few terms, it has indicated that patentability is a higher bar. See Alice Corp. Pty. Ltd. v. CLS Bank Int'l, ___ U.S. ___, 134 S.Ct. 2347, 2354-55, 189 L.Ed.2d 296 (2014); Ass'n for Molecular Pathology v. Myriad Genetics, Inc., ___ U.S. ___, 133 S.Ct. 2107, 2116, 186 L.Ed.2d 124 (2013); Mayo Collaborative Servs. v. Prometheus Labs., Inc., ___ U.S. ___, 132 S.Ct. 1289, 1293-94, 182 L.Ed.2d 321 (2012); Bilski,
The concern underlying § 101 is preemption: the idea that allowing a patent on the invention will impede innovation more than it incentivizes it. Of course, a court should not overstate this concern. By definition, every patent preempts an area of technology. A patentee with a groundbreaking invention is entitled to monopolize a segment of technology, subject to the limits of the Patent Act.
Concerns over preemption have called into question when, if ever, computer software is patentable. A basic truth is that algorithms comprise computer software and computer codes. See J. Glenn Brookshear, Computer Science: An Overview 2 (6th ed. 2000) ("A machine-compatible representation of an algorithm is called a
But intervening precedents and Congressional action have demonstrated that software is patentable. In Diamond v. Diehr, 450 U.S. 175, 101 S.Ct. 1048, 67 L.Ed.2d 155 (1981), the Supreme Court found patentable a method claim implementing a mathematical formula on a computer. See Diehr, 450 U.S. at 179 n.5, 192-93, 101 S.Ct. 1048 (finding patentable claim on "method of operating a rubber-molding press for precision molded compounds with the aid of a digital computer"). More recently, in Alice Corp. Pty. Ltd. v. CLS Bank International, ___ U.S. ___, 134 S.Ct. 2347, 189 L.Ed.2d 296 (2014), the Supreme Court again suggested that software is patentable. See id. at 2359 (suggesting that software which improves function of a computer may be patentable).
The aftermath of Alice tells a different but misleading story about software patentability. Alice brought about a surge of decisions finding software patents ineligible. See, e.g., buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355 (Fed.Cir.2014) (invalidating claim addressed to a "transaction performance guaranty" performed on a computer); Digitech Image Techs., LLC v. Elec. for Imaging, Inc., 758 F.3d 1344, 1349, 1351 (Fed.Cir.2014) (invalidating method claim for generating and combining data sets for device profile); Eclipse IP LLC v. McKinley Equip. Corp., No. 8:14-cv-742, 2014 WL 4407592 (C.D.Cal. Sept. 4, 2014) (invalidating claims reciting methods for communications). Despite this flurry of § 101 invalidations, in reality, Alice did not significantly increase the scrutiny that courts must apply to software patents. It held only that an ineligible abstract idea does not become patentable simply because the claim recites a generic computer. See Alice, 134 S.Ct. at 2360 ("[T]he claims at issue amount to `nothing significantly more' than an instruction to apply the abstract idea of intermediated settlement using some unspecified, generic computer. Under our precedents, that is not `enough' to transform an abstract idea into a patent-eligible invention." (citations omitted)). Courts must not extend the reach of Alice too far, lest they read in § 101 limitations that do not exist. Cf. Bilski, 561 U.S. at 603,
Other than its narrow holding, Alice reaffirmed that courts must evaluate patent eligibility using the two-part test applied in Mayo Collaborative Services v. Prometheus Laboratories, Inc., ___ U.S. ___, 132 S.Ct. 1289, 182 L.Ed.2d 321 (2012). First, a court must ask if the claim is "directed to one of those patent-ineligible concepts": a law of nature, physical phenomenon, or abstract idea. Alice, 134 S.Ct. at 2355. Second, if the claim is directed to one of these concepts, the court must ask "[w]hat else is there in the claims before us?" Mayo, 132 S.Ct. at 1297. This second step determines whether there is an "inventive concept" that "ensure[s] that the patent in practice amounts to significantly more than a patent upon the [ineligible concept] itself." Alice, 134 S.Ct. at 2355. These steps are broadly stated and, without more, would be difficult to apply. Although the two-part test was created in Mayo, pre-Mayo precedents offer guidance in applying the steps.
At the first step of Mayo, the court must identify whether a claim is directed to an abstract idea. To do so, the court must identify the purpose of the claim—in other words, determine what the claimed invention is trying to achieve—and ask whether that purpose is abstract. For example, in Alice, the court concluded that the claims were directed to mitigating settlement risk using a third party, but the claims recited more. The claims outlined an entire process, including creating shadow records, obtaining from an exchange institution a start-of-the-day balance, and so on. See Alice, 134 S.Ct. at 2359. But these steps were meant to achieve the purpose of mitigating settlement risk. The Supreme Court took the same approach in Bilski and Mayo by characterizing the claims in terms of the inventions' purposes: hedging risk and applying a natural law, respectively. See Bilski, 561 U.S. at 611, 130 S.Ct. 3218; Mayo, 132 S.Ct. at 1296-97.
Characterization of a claim is essential to the § 101 inquiry. In Diamond v. Diehr, 450 U.S. 175, 101 S.Ct. 1048, 67 L.Ed.2d 155 (1981), the dispute boiled down to what the majority and dissent were evaluating for abstractness. See id. at 206, 101 S.Ct. 1048 (Stevens, J., dissenting) (faulting majority for characterizing claim by its purpose, which was "constantly measuring the actual temperature inside a rubber molding press"). The Diehr majority took the correct approach by asking what the claim was trying to achieve, instead of examining the point of novelty. Id. at 207, 101 S.Ct. 1048. Courts should recite a claim's purpose at a reasonably high level of generality. Step one is a sort of "quick look" test, the purpose of which is to identify a risk of preemption and ineligibility. If a claim's purpose is abstract, the court looks with more care at specific claim elements at step two.
At step one, prior art plays no role in the analysis. The court does not filter out claim elements found in prior art and evaluate the remaining elements for abstractness. See Caltech, ___ F.Supp.3d at ___ - ___, slip op. at 18-21; but see McRO, 2014 WL 4749601 at *9 (claims must be evaluated in light of prior art because such art is "understood, routine,
Using prior art at step one also impermissibly conflates the two steps of Mayo. Of course, at step two, courts must remember that reciting purely conventional activity will not save a claim, and claim elements found in prior art may occasionally, though not always, constitute conventional activity. Mayo, 132 S.Ct. at 1298. But at step one, the court neither identifies nor disregards conventional activity. That inquiry occurs only at step two.
Once the court has identified a claim's purpose, it must determine whether that purpose is abstract. This task is difficult, especially with regard to computer software. Because software is necessarily intangible, accused infringers can easily mischaracterize and oversimplify software patents. Cf. Oplus Techs. Ltd. v. Sears Holding Corp., No. 12-cv-5707, 2013 WL 1003632, at *12 (C.D.Cal. Mar. 4, 2013) ("All software
If the court finds the claim's purpose abstract at step one, it must then determine whether there is an inventive concept that appropriately limits the claim, such that the claim does not preempt a significant amount of inventive activity. In performing the second step of analysis, the court must be wary of making patentability "a draftsman's art," Flook, 437 U.S. at 593, 98 S.Ct. 2522, but inevitably, drafting plays a key role. Patents that claim
With this concern in mind, the court must disregard "well-understood, routine, conventional activity" at step two. Mayo, 132 S.Ct. at 1299.
The court must also consider the claim elements as a combination. A combination of conventional elements may be unconventional and therefore patentable. See Diehr, 450 U.S. at 188, 101 S.Ct. 1048 ("[A] new combination of steps in a process may be patentable even though all the constituents of the combination were well known and in common use before the combination was made."). Courts should consider all elements as part of the "ordered combination," even those elements which, in isolation, appear abstract. See Diehr, 450 U.S. at 189 n.12, 101 S.Ct. 1048.
Enfish has asserted eight claims from the '604 patent and three claims from the '775 patent. At step one, the Court determines that all of the asserted claims are addressed to abstract ideas. At step two, the Court determines that the claim limitations do not supply an inventive concept that sufficiently limits the scope of the claims. Therefore, all asserted claims are unpatentable.
All asserted claims of the '604 and '775 patents are directed to abstract ideas. Every claim has a similar purpose: storing, organizing, and retrieving memory in a logical table. Memory represents data or information.
The fact that the patents claim a "logical table" demonstrates abstractness. The term "logical table" refers to a logical data structure, as opposed to a physical data structure. See Claim Construction Order at 7, Dkt. No. 86 ("[A logical table has] a data structure that is logical as opposed to physical, and therefore does not need to be stored contiguously in memory."). Under this construction, it does not matter how memory is physically stored on the hardware. In essence, the claims capture the
Given these observations, the Court determines that the claims are addressed to the abstract purpose of storing, organizing, and retrieving memory in a logical table. This abstract purpose does not become tangible because it is necessarily limited to the technological environment of computers. See Alice, 134 S.Ct. at 2358 ("[M]ere recitation of a generic computer cannot transform a patent-ineligible abstract idea into a patent-eligible invention."). Tellingly, in Alice, the Supreme Court defined the purpose of the claims as achieving intermediated settlement; it did not define the purpose of the claims as achieving intermediated settlement using a computer. See Alice, 134 S.Ct. at 2356-57. When a claim recites a computer generically, the Court should ignore this element in defining the claim's purpose.
Because the claims are addressed to an abstract idea, the Court must determine whether the claims contain additional limitations that amount to an inventive concept. The claims do not. Instead, the claims recite conventional elements. These elements, when viewed individually or in a combination, do not sufficiently cabin the claims' scope.
The Court begins by analyzing claim 47 of the '604 patent, which is representative of the patents in general. It recites
These limitations comprise a series of conventional elements. "[C]onfiguring said memory according to a logical table" simply means storing computer information, possibly for later retrieval. The purpose of all tables is to store information for later retrieval. As such, this limitation is conventional because it recites the obvious purpose of all tables. The "logical table" element merely indicates that the table can store memory non-contiguously. Non-contiguous memory allocation is the concept of storing a program in various parts of the memory. The concept of non-contiguous memory is ubiquitous in the art and therefore conventional. See D.M. Dhamdhere, Operating Systems: A Concept-Based Approach 213 (2d ed.2006) ("In the noncontiguous memory allocation model, several non-adjacent memory areas are allocated to a process."); Robert C. Daley & Jack B. Dennis, Virtual Memory, Processes, and Sharing in MULTICS 309 (May 5, 1968), available at http://users.soe.ucsc.edu/sbrandt/221/Papers/History/daley-cacm68.pdf ("Paging allows noncontiguous blocks of main memory to be referenced as a logically contiguous set of generalized addresses."). The recitation of rows and columns is also conventional, because these elements are necessary to create a table. See Mayo, 132 S.Ct. at 1298. The recitation of OIDs is likewise conventional. As the specification makes clear, OIDs assigned to rows and columns are used for "exact retrieval." '604 Patent, 2:1. Certainly, OIDs are helpful for computers to locate information, especially because computers store large amounts of information. But OIDs are essentially labels for each column and each row. See '604 Patent, Fig. 3; 6:47-48 ("[T]he OID's for both rows and columns may be used as pointers."). Using the labels to locate information is a basic concept that humans have long employed. Indeed, labels are often the easiest way to locate information in a table.
Likewise, the limitation of "indexing data stored in said table" is not an inventive concept. The Court has construed "indexing" to mean "organizing data to enable searching." Enfish, 2014 U.S. Dist. LEXIS 46523 at *22. Humans engaged in this sort of indexing long before this patent, and the claim does not put forth an innovative and unconventional method of indexing. A bare recitation of an indexing limitation does not make this abstract idea patentable. Finally, the recitation of computer memory does nothing more than limit the abstract idea to a technological environment. This is not
The same analysis applies to claim 17 of the '604 patent. Claim 17 is drafted in a means-plus-function format and recites:
This claim differs from claim 47 in two ways: it claims a "means for configuring" and a "means for indexing," each of which covers an algorithm described in the specification. The Court first determines whether the "means for configuring" limitation recites an inventive concept. It then determines whether the "means for indexing" limitation recites an inventive concept.
The "means for configuring" limitation does not recite an inventive concept. "Means for configuring" covers a four-step algorithm:
See Claim Construction Order at 8-9. This algorithm does not constitute an inventive concept. The first step recites the creation of a logical table that need not be stored contiguously in computer memory. As discussed above, non-contiguous memory allocation is a basic idea in computing and is therefore conventional. As the Court also discussed above, creating a logical table on a computer is an abstract concept that is not patentable without something more. The second step recites two conventional elements: assigning OIDs and allowing for OIDs that may vary in length across databases ("variable-length OIDs"). As discussed with regard to claim 47, assigning OIDs is a conventional step. Variable-length OIDs are likewise conventional. The element merely allows OIDs across databases to comprise a different number of bits, "depending on the precision required." See '604
The third step recites the conventional element of creating columns from information stored in rows, in order to render the table self-referential. A table is self-referential if each column is defined by information stored in one or more of the table's rows. Enfish's Opp'n to Non-Infringement at 11-12, Dkt. No. 280. Figure 3 of the '604 patent shows one example of this: column 126 has the definition "Employed By," which corresponds to row 136. See '604 Patent, Fig. 3; '604 Patent, 7:16-19 ("The column definition is stored as a record in the table
All four steps of the algorithm are conventional, and as an ordered combination, they remain conventional. The algorithm creates a table with labeled rows and columns, where one row defines the type of information in each column. This is the description of a purely conventional table, and nothing more. Nothing in this algorithm sufficiently limits the scope of the claim.
As with the "means for configuring" limitation, the "means for indexing" limitation recites no inventive concept. "Means for indexing" covers a three-step algorithm:
See Claim Construction Order at 9. The first two steps are the essence of indexing. Extracting key phrases and storing these phrases in an index are steps that are necessary for indexing. As a result, they are purely conventional. The fact that the index is stored in the logical table itself does not sufficiently limit the scope of the claim. The logical table is an obvious place to store an index for the same logical table, much like the back of a textbook is an obvious place to store an index for that textbook. The third step is likewise conventional. An index's purpose is to point to the location of information, and it is unremarkable for the cells to contain information pointing back to the index, in order to aid in further data retrieval. Again, an inventor could not patent a hand-drawn table with pointers from a text cell to an index, and this concept remains unpatentable when applied to a computer. This concept is not sufficiently inventive enough to cabin the claims, because it would preclude inventors from performing a basic step to maximize the potential of indices. Viewing these steps as an ordered combination adds nothing, because the algorithm as a whole represents a conventional process of indexing.
Viewing claim 17 as an ordered combination does not change the result. Claim 17 describes how to store information in a table and use an index to find information in that table. These ideas in combination are purely obvious, conventional activity. Therefore, claim 17 is unpatentable.
The other asserted claims add little to the substance of claims 17 and 47. Claims 16 and 46 of the '604 patent requires OIDs to be of variable length, but this limitation is insignificant and conventional. As discussed above, varying the length of identifying labels, whether on a computer or otherwise, is a basic concept. This limitation simply acknowledges the reality that computers may assign a different numbers of bits to different labels. Viewing the claims' elements as a combination adds nothing: the result is a conventional table for storing and retrieving information, implemented on a computer. Claims 16 and 46 are therefore unpatentable.
Claims 1, 2, 31, and 32 of the '604 patent and claims 31 and 32 of the '775 patent likewise recite a conventional element: requiring a single row in the database that defines all the columns. See Claim Construction Order at 6-7 ("[T]he implication is clear: at least one fully-populated row is required, i.e., at least one row with values defined for each column."); Enfish's § 101 Opp'n at 20, Dkt. No. 299. As discussed above, a common concept in tables is a fully populated row defining the information in each column. This concept was accomplished in tables created long before computers. Implementing this idea on a computer is so obvious as to be conventional. See Mayo, 132 S.Ct. at 1298 ("Purely `conventional or obvious' `[pre]solution activity' is normally not sufficient to transform an unpatentable law of nature into a patent eligible application of such a law."). As with the variable length OIDs, the Court's analysis of the claim as a combination does not change the Court's conclusion. The combination of elements in the claims results in an unremarkable table,
Claim 47 of the '775 patent is substantively identical to claim 47 of the '604 patent. The claim uses synonyms to describe the invention and recites conventional elements, such as a computer system and cells having address segments. Enfish does not argue that claim 47 of the '775 patent adds any concept which makes it uniquely patentable. Thus, claim 47 is also unpatentable.
Enfish's asserted claims are unpatentable because they apply longstanding concepts about storing information in tables to the technological environment of computers. But this does not mean that every software invention that uses longstanding concepts is unpatentable. To satisfy § 101, software inventions just need something sufficiently more than a recitation of a longstanding concept.
To understand this idea, imagine three computer programs directed at a longstanding concept: determining the best series of moves in a chess game. The first program calculates moves through a "brute force" method—that is, it determines the best moves by testing various moves. A claim for this program is unpatentable. Chess players have long used some form of brute force calculation to determine their subsequent moves. The fact that a computer can perform brute force calculations faster than humans is irrelevant. Rapid processing of data is a generic function of computers.
The second chess program determines future moves by considering various factors, such as the safety of the king, the development of pieces, and the number of pieces each side has. Again, a claim for this program would be unpatentable without something more. Chess players have long used these factors to evaluate positions in chess games. A claim for this program takes fundamental ideas about chess and has a computer apply them. Again, it is irrelevant that a computer may apply these ideas more effectively than a human.
The third chess program determines future moves by evaluating various factors, just like the second program. But this third program does more. It allocates different amounts of computer memory to different factors, and it reallocates memory at different stages of the game, as some factors become more important and others less important. A claim for this program is patentable. The claim is not merely addressed to an abstract idea. It is addressed to an inventive computing concept: dynamic memory allocation. The claim's computer elements are not generic. Rather, the claim recites a modern, computerspecific concept to solve the modern, computer-specific problem of scarce memory. Although the claim implements longstanding ideas about chess, computing concepts form a significant part of the claim. The combination, of these chess ideas and computing concepts constitutes patentable subject matter.
In contrast to this last example, tables are an age-old solution to an age-old problem. Tables are a basic building block of research and development and are not patentable. This remains true when tables are implemented on computers. Undoubtedly, tables improve data storage on a computer, but only inasmuch as tables improve data storage when used in
The Court today decides that Enfish's asserted claims are unpatentable. This determination does not mean that the invention is valueless. Many useful inventions are unpatentable, despite the tremendous effort that went into their creation. But usefulness is not the current standard for patentability, and the Court must follow the principles established by the Supreme Court. See McRO, 2014 WL 4749601 at *12 ("[T]he revolutionary nature of an abstract idea does not weigh in favor of patentability.").
The Court is mindful of the fact that inventors are the casualty of courts' evolving § 101 jurisprudence. The filing date for Enfish's patents is March 5, 1998. Only a few months after this date, the Federal Circuit created a generous standard for patentability, holding a process was patentable if it created a "useful, concrete, and tangible result." State St. Bank & Trust Co. v. Signature Fin. Group, 149 F.3d 1368, 1375 (Fed.Cir.1998). For years, patentees relied on this low bar when writing their applications to the Patent and Trademark Office. But the rules changed in Bilski, Mayo, and Alice. Many inventors drafted their patents for an age of patent law that no longer exists, and inventors have suffered the consequences of shifting jurisprudence.
Predictability and stability are crucial for a successful patent system, and courts bear the responsibility for developing a consistent § 101 standard by earnestly following the guidance of higher courts. Based on the Supreme Court's precedents, this Court concludes that all asserted claims are unpatentable. Therefore, the Court grants the Defendants' motion for summary judgment.
IT IS SO ORDERED.
Name Age Height Abby 27 5'11" Ben 42 5'8" Carla 39 5'5"