How Means-Plus-Function Limitation Interacts With Software Industry? Do “Processor,” “Controller,” Or “Program” Invoke 35 U.S.C. §112?

Finished on 2019.05.31  

閱讀本篇文章前可參考：Can AI algorithms be patented under AIA? 及 電腦軟體的智慧財產權保護

I. Introduction of Software Patent Facing 35 U.S.C. §112(f) 

As the USPTO announces the new framework of eligibility test for §101 issues, it again rouses the applicant’s concerns about getting software patents. Does the new test loosen the constraint of passing the Alice/Mayo in prosecution?[1] Well, it may be hard to say so. However, when applicants try to procure software-related patents and try to overcome the “significantly more” requirement, there are other potential risks other than patent eligibility—claims indefiniteness with unclear language. 

While patent drafters may use tangible wordings to get rid of potential patent ineligibility, the tangible-like words such “processor” or “controller” in the claims may not be well-supported by the specification because it may trigger to the examination of Means-Plus-Function claims of 35 U.S.C §112(f). 

On the other hand, even though the claims survived after an unexpected examination to be as means-plus-function claims, the scope of the claim may be narrowed in the specification as the §112(f) regulation. In order to reduce the risk being unintentionally treated as means-plus-function claim, a drafter should have been more careful with the wordings. This article will focus more on the wordings, which the previous cases light upon, and try to speculate some of other words may cause the same issues with this author’s perspective. Also, this article will not discuss on method claims, which involves “steps-plus-function.”

II. What Is Means-Plus-Function Claims?

The scope of a patent determines the value of it. Every inventor would like to broaden his or her patent scope by attempting to define their invention at a higher level of language. To achieve that, for instance, an inventor of a new chemical might claim a set of related chemical substances like a genus instead of specific species. Likewise, an inventor of a mechanical invention may claim or describe an inventive contribution, which any apparatus would implement the conceptual inventive contribution, rather than identify specific patent boundary with actual structural elements. One of the famous cases which is held to be enforceable with this far-ranging claim languages is the Wright Brother’s airplane. In claim 7 of the U.S patent No. 821,393 issued on May 22, 1906, they recited:  

In a flying machine the combination with an [airplane], and [;] 
means for simultaneously moving the lateral portions thereof into different angular relations to the normal plane of the body of the [airplane] and to each other, so as to present to the atmosphere different angles of incidence, of a vertical rudder, and [;] 
means whereby said rudder is caused to present to the wind that side thereof nearest the side of the [airplane] having the smaller angle of incidence and offering, the least resistance to the atmosphere, substantially as described.

In this patent, Wright Brother widely used the means-plus-function language, and there are twenty-three (23) “means” recited in eighteen (18) claims. This broad interpretation of a claim frustrated many late-coming improvements which actually surpassed the technical achievement of Wright Brother’s. Glenn Curtiss, a talented airplane engineer who improved the design of the wings, however, infringed the broad scope means-plus-function claim.[2] Also, Paulhan, a French inventor, infringed the patent. The court held that Wrights’ patent to be pioneering and so entitled to broad scope.[3] The defendant tried to argue the unreasonable claim construction, Curtiss derisively suggested that if someone jumped in the air and waved his arms would infringe the patent.

However, this practice has been ruled out by the Supreme Court in Halliburton Oil Well Cementing v. Walker, 329 U.S. 1 (1946). The Court held that the patent claim was indefinite because it did not specify how the inventive contribution performs the function or limited the invention to the particular contribution(means) which the drafter(patentee) actually invented. The court asserted that it was improper for a claim to describe its “most crucial element … in terms of what it will do, rather than in terms of its own physical characteristics or its arrangement in the new combination apparatus.” A substitute functional language at a point of novelty did not comply with the essence of the patent system. 

After six years, the Congress, however, rejected Halliburton in the 1952 Patent Act by adding the provision that elements of a combination claim may be drafted in functional form. 35 U.S.C. §112(f). Section 112(f) provides: 

III. Why Means-Plus-Function Claims Matter Computer-Implemented inventions?

a. Cases before Alice using means-plus-function overcome §101 issue

To pass the 35 U.S.C. §101, an applicant might recite a physical apparatus fall into the scope of statutory requirements like machines, manufactures and compositions of matter (“process” omitted since this article focus on “means” plus function, instead of method claims). For example, a prosecutor might use terms like “processor” or “rasterizer” as the preamble in a claim. However, how to express a program’s contribution in such “physical” approach? The means-plus-function claims provide the applicant a way to do so. For instance, in Alappat, the court held that a tangible machine with a specific function, implementing software, complies with the statutory scope of 35 U.S.C. §101.[4] It is notable that the disputed claim was drafted with means-plus-function limitations, where actually recited what functions of the program does in each limitation. The claim 15 read: 

A rasterizer … comprising:   
(a) means for determining the vertical distance between the endpoints of each of the vectors in the data list;   
(b) means for determining the elevation of a row of pixels that is spanned by the vector;   
(c) means for normalizing the vertical distance and elevation; and
(d) means for outputting illumination intensity data as a predetermined function of the normalized vertical distance and elevation.

The court found that even though many of the elements recite circuital elements that perform mathematical calculations, the most important point is that:

[T]he claimed invention as a whole is directed to a combination of interrelated elements which combine to form a machine for converting discrete waveform data samples into anti-aliased pixel illumination intensity data to be displayed on a display means. This is not a disembodied mathematical concept which may be characterized as an “abstract idea,” but rather a specific machine to produce a useful, concrete, and tangible result.

Another example is Noah Systems, Inc. v. Intuit, Inc.[5] The patentee survived and passed patent eligibility with the means-plus-function approach, however, failed the §112(b) and (f) requirements of corresponding structure in the specification. Noah alleged that Intuit’s products infringe its claims and all of the asserted claims contained an “access means” limitation: 

A financial accounting system for a first entity such as an individual or a business, said system comprising:
… 
means for providing access to said file of said financial accounting computer for said first entity and/or agents of said first entity so that said first entity and/or said agent can perform one or more activities selected from the group consisting of entering, deleting, reviewing, adjusting and processing said data inputs.

According to §112(b) and (f), means-plus-function claims in a software patent the specification of the patent needed to contain a specific embodiment(algorithm) that support the function associated with the “access means” limitation, otherwise the limitation is indefinite and the claim invalid. However, the application did not disclose an algorithm for performing the function associated with the “access means” limitation. This determination leads to a decision rendering all of the asserted claims indefinite because they lacked the required corresponding structure in the specification. Even though it ultimately invalidated by indefiniteness, it overcame the patent eligibility at the time. 

However, whether the approach of means-plus-function claiming of software could satisfy the current Alicetest become important. It seems logical that a means plus function claim, which will read in the full extent of the innovation disclosed in the specification, could be one that is patent eligible and not considered merely an abstract idea. Although the scope of means-plus-function claims may be narrowed, one can still use it as a safe harbor of eligibility and asserting an infringement action. 

b. After Alice, the means-plus-function claim still feasible for a software-related invention to overcome §101 issue

The Alice didn’t rule out the original requirement of statutory categories— machines, manufactures, and compositions of matter. The first part of the Alice test is to determine whether the claims are directed to an abstract idea, a law of nature or a natural phenomenon (i.e., a judicial exception). If the claims are directed to a judicial exception, the second part of the Alice test is to determine whether the claim recites additional elements that amount to significantly more than the judicial exception:

In Enfish case, the patentee contends that its claimed self-referential table for a computer database was an improvement in existing technology and thus not directed to an abstract idea.[6] The court agreed with the patentee, in light of Noah, based on its interpretation of the claimed “means for configuring” under 35 U.S.C. §112(f) as requiring a four-step algorithm that achieved the improvements, as opposed to merely any form of storing tabular data.  The claims recited: 

A data storage and retrieval system for a computer memory, comprising:
means for configuring said memory according to a logical table, said logical table including:
a plurality of logical rows, each said logical row including an object identification number (OID) to identify each said logical row, each said logical row corresponding to a record of information;
a plurality of logical columns intersecting said plurality of logical rows to define a plurality of logical cells, each said logical column including an OID to identify each said logical column;
and means for indexing data stored in said table.

Also, the meanings of “means for configuring” supported by corresponding algorithms in the specification read as:

1. Create, in a computer memory, a logical table that need not be stored contiguously in the computer memory, the logical table being comprised of rows and columns, the rows corresponding to records, the columns corresponding to fields or attributes, the logical table being capable of storing different kinds of records.
2. Assign each row and column an object identification number (OID) that, when stored as data, ….
3. For each column, store information about that column in one or more rows, rendering the table self-referential, the appending, to the logical table, ….
4. In one or more cells defined by the intersection of the rows and columns, store and access data, which can include structured data, unstructured data, or a pointer to another row.

Enfish successfully provided enough disclosure in the specification and asserted the claim was eligible with improving technology. I think it is because of the requirements of means-plus-function, requesting a patent reader has to interpret the claims heavily on the specification’s disclosure, the court ultimately accepted the disclosure and said: 

The specification [] teaches that the self-referential table functions differently than conventional database structures.  According to the specification, traditional databases, such as “those that follow the relational model and those that follow the object oriented model,” are inferior to the claimed invention. 

The court then affirmed the claim read on the specification successfully increased the capability of the prior art, and cited that a specification’s disparagement of the prior art is relevant to determine the scope of the invention.[7]

Another case, Amdocs, recited means-plus-function claim, successfully conquered the step two of Alice test and ruled that it was significantly more than an abstract idea.[8] Amdocs focused on a representative claim 1 recited:

A computer program product embodied on a computer readable storage medium for processing network accounting information comprising:
computer code for receiving from a first source a first network accounting record;
computer code for correlating the first network accounting record with accounting information available from a second source;
and computer code for using the accounting information with which the first network accounting record is correlated to enhance the first network accounting record.

Without much doubt, this claim quickly was regarded as a means-plus-function claim and an abstract idea. The Court noted that the language in claim 1 could avoid abstract ideas if it was described as “focusing on correlating two network accounting records to enhance the first record;” or “focusing on a computer program that includes computer code for receiving initial information, for correlating that initial information with additional information, and for using that additional information to enhance the initial information.” 

However, the patentee overcame the second step of the Alice test. With the narrow interpretation from the specification, the claim was tied to a specific structure of various components, which is narrowly drawn to not preempt a broad abstract idea and does not merely combine the components in a generic manner, but instead achieve a technological solution with a specific embodiment.

Although the reason for the claim being eligible may not be directly caused by the means-plus-function approach, a diverse claim sets in an application gave a patentee more possibilities to assert his rights. 

c. The limited expressions of software functions but numerous way to accomplish functions  

Why applicants would like to use this facially narrower way to recite their invention? Besides the possible reason that provides the foregoing, another reason might ground on the characteristics of “equivalence” to the claim construction. The statute in the last phrase states—“and equivalent thereof”—permits some broadening of the scope, however, such equivalents must have been known as of the time of the invention filed(or invented). This unique claim interpretation is critical for a software-related invention because it retains some flexibility of extending a terminology to its equivalents. For example, a specific function may be run by thousands of different algorithms, which each of the substitutes are created by different coders. In addition, the engineers always come out with new programs, frameworks, models, algorithms or terminologies for specific functions or goals. In the field of neural networks, there are a lot of different structures/frameworks for processing inputs data and converging outputs toward a given goal. However, in the claim language, it has a limited way to recited the function of “convergence,” and it is difficult to describe with structural wordings. Therefore, a means-plus-function claim like “means for converging” could be a possible solution. 

Nonetheless, a drafter may avoid using intangible-liked words such as “means” to reduce the possibility of failing the patent eligibility, but instead use physical words like “apparatus,” “processor,” or “module.”  However, a term lacking “means” will nonetheless be construed under §112(f) if the “challenger demonstrates that the claim fails to recite sufficiently definite structure or else recites function without reciting sufficient structure for performing that function.”[9] The crucial question is “when” and “how” to decide a claim should be read as a means-plus-function claim, because sometimes it matters a live-or-dead situation to litigation, and that may be a patentee’s last resort. 

IV. When Does 35 U.S.C §112(F) Apply?

Before digging into the analysis of means-plus-function implementation, a prosecutor should note that the claim construction standard in prosecution is “broadest reasonable interpretation” (BRI). Conversely, in litigation, the claim construction standard is “Phillips standard.”[10] The Court of Appeals for the Federal Circuit, in its en banc decision InreDonaldson held that a “means-or-step-plus-function” limitation should be interpreted as follows:[11]

Per our holding, the “broadest reasonable interpretation” that an examiner may give means-plus-function language is that statutorily mandated in paragraph six. Accordingly, the PTO may not disregard the structure disclosed in the specification corresponding to such language when rendering a patentability determination.

With this in mind, a drafter may need to note that a narrower scope may be interpreted than a limitation that is not crafted in “means plus function” format. However, since a court applies Phillips standard during litigation, the scope is the same or narrower than BRI standard. This difference, however, may not be significant with respect to means-plus-function claims.

To determine whether a claim limitation invokes 35 U.S.C. 112(f), according to MPEP §2181, examiners will apply §112(f) to a claim if it meets the 3-prong analysis: 

a. Prong 1: The Claim Limitation Uses the Term “Means” or “Step” or a Generic Placeholder (A Term That Is Simply A Substitute for “Means”)

With respect to the first prong of this analysis, a claim element that does not include the term “means” or “step” triggers a rebuttable presumption that 35 U.S.C. 112(f), does not apply. When the claim limitation does not use the term “means,” examiners should determine whether the presumption that 35 U.S.C. 112(f) does not apply is overcome. The presumption may be overcome if the claim limitation uses a generic placeholder (a term that is simply a substitute for the term “means”).

b. Prong 2: The Term “Means” or “Step” or the Generic Placeholder Must Be Modified By Functional Language

With respect to the second prong of this analysis, it must be clear that the element in the claims is set forth, at least in part, by the function it performs as opposed to the specific structure, material, or acts that perform the function. In other words, if the phrase “means” does not link to a specific function in the claim language, it is not under §112. The mere use of the term “means” with no associated functional context rebuts the presumption of §112(f) invoked. A function must be recited within the claim limitation, but it is not necessary that a particular format is used.

c. Prong 3: The Term “Means” or “Step” or the Generic Placeholder Must Not Be Modified By Sufficient Structure, Material, or Acts for Achieving the Specified Function

With respect to the third prong of this analysis, the term “means” or “step” or the generic placeholder recited in the claim must not be modified by sufficiently definite structure, material, or acts for achieving the specified function. Examiners will apply 112(f) to a claim limitation that uses the term “means” or “step” or generic placeholder associated with functional language, unless that term is (1) preceded by a structural modifier, defined in the specification as a particular structure or known by one skilled in the art, that denotes the type of structural device (e.g., “filters”), or (2) modified by sufficient structure or material for achieving the claimed function.

d. The Specific Terms For Software

As for the software-related inventions, a drafter, of course, can recite “means” with the strong presumption purposely and let the claims to be construed under §112(f). Also, an applicant can recite other terms that are more specific than “means” to match the software industry. The phrases like “module for,” “system for,” “unit for,” or “apparatus for,” which may be treated as non-structural functional claim as long as they fulfill the foregoing two-prong: (b) the phrases must be modified by functional language and (c) the phrases must NOT be modified by sufficient structure, material, or acts for achieving the specified function. For example, “a module for determining the inputs data …” may be treated as §112(f) because the “module” is modified by functional language “determining,” and there is no other modifier correlate to the “module.” However, if the claim is like “a detent module for…” may not fall in §112(f) scope because the word “detent” denotes a type of structural device with a generally understood meaning in the mechanical arts.[12]

However, in Zeroclick, a CAFC case overruled the district court that the phrases “program” and “user interface code,” these seemly broad and high-level terms, have not been treated as means-plus-function terms.[13] The Claim 2 of the No. 7,818,691 patent and claim 19 of the No. 8,549,443 patent are recited:

2. A graphical user interface (GUI), which may comprise an update of an existing program, that may fully operate a GUI by a two-step method of movement of a pointer (0) to operate one or more functions within the GUI, wherein, said existing program is any existing program that can operate the movement of the pointer (0) over a screen (300) and has one or more functions operated by one or more other methods apart from said two-step method, and/or one or more functions operated by said one or more other methods in said existing program can be updated to operate by said two-step method, … second by the completion of a subsequent movement of said pointer (0) according to a specified movement generates a ‘click’ event, thereby triggering one or more functions within the GUI.  

19. A device capable of executing software comprising: … a processor connected to the touch-sensitive screen and configured to receive from the screen information regarding locations touched by the user’s finger; executable user interface code stored in a memory connected to the processor; the user interface code executable by the processor: the user interface code being configured to detect one or more locations touched by a movement of the user’s finger on the screen without requiring the exertion of pressure and determine therefrom a selected operation; and the user interface code is further configured to cause one or more selected operations, which includes one or more functions available to the user interface code of the device, to deactivate while the user’s finger is touching one or more locations on the screen.

The court held that even though these two phrases “program” and “user interface code” facially like means-plus-function terms, its might not be automatically the substitutes of “means.” However, if the context strongly supported the plain and ordinary meaning of the terms it may not within the statute. The court further cited Williamson that the essential inquiry is “whether the words of the claim are understood by persons of ordinary skill in the art to have a sufficiently definite meaning as the name for structure.” The court also pointed out that there are three:

1. The fact that claims incorporate functional language does not automatically convert the words into means for performing such functions. Many devices take their names from functions they perform (filter, brake, clamp, screwdriver, lock, etc.)
2. The context strongly supported the plain and ordinary meaning of the terms. For example, claim 2 of the ‘691 patent recites “[a] graphical user interface” which may comprise “an update of an existing program” (emphasis added by the court) Claim 19 of the ‘443 patent recites that the “user interface code” to the code “stored in a memory connected to the processor.” A person skilled in the art would understand these terms to refer to conventional graphical user interface programs or code, existing in the prior art at the time of the inventions.
3. The district court findings do not compel the conclusion that the terms “program” or “user interface code” are used as a substitute for “means.” Further, the patent owner makes no such acknowledgment. Accordingly, the district court erred by effectively treating these terms as nonce words.

According to the USPTO “informative opinions” on functional claiming, the BPAI determined that the term “processor” is a nonce word that merely substitutes for means-plus-function language, is not the name of structure, and invokes § 112(f).  The four decisions identified as informative are: Ex Parte Lakkala, Ex Parte Erol, Ex Parte Smith, and Ex Parte Cadarso.[14]

“Controller” is another term where it may be unclear if the term connotes structure or not.  Sound View Innovations analyzes the term “controller,” and based on the specific facts found that the term “controller” is structural and does not invoke §112(f).[15] The claim 1 of patent No. 6,125,371 read as:

1. A processing system for use with a database of data records, said database stored in a memory, comprising:
a time stamping controller that assigns a time stamp to transactions to be performed on said database;
a versioning controller that createsmultiple versions of ones of said data records affected by said transactions that are update transactions;
and an aging controller that monitors a measurable characteristic of said memory and deletes ones of said multiple versions of said ones of said data records in response to said time stamp and said measurable characteristic thereby to increase a capacity of said memory.

The court pointed out in the specification the patentee put on the definition of “controller.” The court quoted, “Those skilled in the art should be familiar with the use of controllers in processing environments generally and, more specifically, with main memory databases,” and “Controllers may be implemented in software, firmware, hardware, or some suitable combination of at least two of the three.” The court construed the language to mean that “‘controller’ connotes, not one particular type of controller, but controllers implemented in any of those mediums.”  The court further said, “Put otherwise, this statement does not define the term ‘controller’ broader than the class of structures ‘controller’ would ordinary connote,” and that “[i]n fact, the patent stresses in that part that ‘controller’ is used in its ordinary sense.” The statement above may correspond to the prong 1, and out of the scope of substitutes of “means.”

The court also regarded the modifiers “time stamping,” “versioning,” and “aging” to the “controller” as terms of functional monikers given to distinguish between which controller is being referenced, which may be corresponded to the prong 3, giving sufficient structure, material, or acts for achieving the specified function to the “controller.”

V. Considerations of Practicing 35 U.S.C 112(f) In Software Industry

It is hard to say a few decades in the future, there will come out other new phrases to be effectively regarded as means-plus-function terms. However, the liberty of the patent system in the United States gives a drafter more room for being creative. An applicant may still figure out other terms to be treated as “means-plus-function” limitation. In addition, a prosecutor may even use languages with uncertainty to push the boundaries in litigation, or even to be its litigation strategy.

Without a doubt, if a software inventor tries to use means-plus-function claims to be his claim set in a clear mindset. It is better to use unambiguous languages like “means for” or other well-established phrases like “module for,” “component for” for his claims languages. However, the inventor would be asked for putting enough effort on disclosing the corresponding structure in a way that one skilled in the art will understand what structure will perform the recited function in the claim. Otherwise, the patent would be lack of disclosure, failing to comply with §112(b).

The substitutes of “means” may seem not that important as if an applicant made it clear that supporting structure in the specification is a pre-requisite for means-plus-function claims to be found to include statutory subject matter for software-related inventions. In other words, each application is unique to the world, and it needs a patent drafter’s deliberate carving to make it ready to put into a business.

VI. Conclusions

In conclusion, there are many different ways to recite an inventive contribution with respect to software inventions. However, never forget to take the “means-plus-function” as one of the options when a prosecutor is drafting an application. Sometimes intentionally recited one or more than one claim sets based §112 may be a better approach. To have various types of claims in an application, may give the patentee more chance to allege and survive in litigation. Nonetheless, the prosecutor should note that there is NO magic language that must appear in a claim in order for it to fall within the scope of §112, it must be clear that the element in the claim is set forth, at least in part, by the function it performs as opposed to the specific structure, material, or acts that perform the function.

On the other hands, there are often advantages to be found in exploiting areas of uncertainty in the law. As discussed above, it is now possible to draft extremely broad claims which appear to read directly on prior art inventions. Through the use of definitional language included in the specification or affidavits provided during prosecution, the practitioner can obtain patents which are potentially very broad.

Additionally, as discussed above, such claims would also be very useful in litigation. Since the exact scope of such a functional claim is very poorly defined, its definition could presumably be adapted to cover various types of accused devices. Similarly, experts would presumably be able to argue that the scope of equivalents includes only as much corresponding structure as would be necessary to cover the accused device and avoid invalidity.

Last but not least, the intellectual property protection for software gradually increases worldwide. Even though patent protection may give the strongest exclusive right against third parties compared with other types of IP protection, it may not always be the best approach in business dealings.

「姓名標示─非商業性─禁止改作」授權條款臺灣3.0版
本著作採「創用CC」之授權模式，僅限於非營利、禁止改作且標示著作人姓名之條件下，得利用本著作。

---

Author

楊杰凱 專利師
Jack Yang Patent Attorney
2019.05.13

總承國際專利商標事務所｜總承會計師事務所｜主持專利師｜商標代理人｜國立清華大學工業工程管理學系｜清大創新創業學分學程｜華盛頓大學法學院法學碩士｜接受新事物，突破舒適圈。主要提供公司成立及智財服務，從工商登記、商標設計、商標申請到專利申請。“Stay Positive. Work Hard. Make it happen.”

---

Reference

• [1] CLS Bank Int’l v. Alice Corp. Pty, 717 F.3d 1269 (2013). [Back]
• [2] Wright Co. v. Herring-Curtiss Co., 211 F. 645 (2nd. Cir. 1914). [Back]
• [3] Wright Co. v. Paulhan,177 F. 261, 260 (2nd Cir. 1910). [Back]
• [4] In re Alappat, 33 F.3d 1526 (1994). [Back]
• [5] Noah Sys. Inc. v. Intuit Inc., 675 F.3d 1302 (2012). [Back]
• [6] Enfish, LLC v. Microsoft Corp.,822 F.3d 1327(Fed. Cir. 2016). [Back]
• [7] Openwave Sys., Inc. v. Apple Inc., 808 F.3d 509, 513–14 (Fed. Cir. 2015). [Back]
• [8] Amdocs (Israel) Ltd. v. Openet Telecom, Inc., 56 F. Supp. 3d 813. [Back]
• [9] Williamson v. Citrix Online LLC, 792 F.3d 1339 (Fed. Cir 2015) (en banc). [Back]
• [10] See Phillips v. AWH Corp., 415 F.3d 1303 (Fed. Cir. 2005) (en banc). [Back]
• [11] InreDonaldson Co., 16 F.3d 1189, 1194 (Fed. Cir. 1994). [Back]
• [12] Greenberg, 91 F.3d at 1583, 39 USPQ2d at 1786. [Back]
• [13] Zeroclick, LLC v. Apple Inc., 891 F.3d 1003. [Back]
• [14] https://www.uspto.gov/patents-application-process/appealing-patent-decisions/decisions-and-opinions/key-decisions. [Back]
• [15] Sound View Innovations, LLC v. Facebook, Inc., 2017 U.S. Dist. LEXIS 76412. [Back]