Guest Post -- Not Yet Evolved? Where Is 35 U.S.C. § 101 Going?

By Brittany Knight* and Anthony Sabatelli** --

Software is a generic term to refer to a collection of data or executing code used by a computer to implement arithmetic operations or logical operations.  Back in 1936, Allen Turing, in his landmark paper On Computable Numbers proposed the capable use of instructions (i.e., software) to execute functions through his Turing Machine, an early precursor of the general purpose computer[1].  Today the modern computer is instrumental for complex computations, data analytics, high-powered graphing, simulations, and so on.  Although, the development and use of software for these computer functions has become mainstream, determining the patent eligibility of software inventions under 35 U.S.C. § 101 has been less intuitive.

In the wake of the now famous Mayo and Alice decisions, patent subject eligibility is determined through a 2-step inquiry.  The first step asks whether, the patent claim is directed to a natural phenomenon, biological process, or abstract idea.  Software is generally categorized as being "abstract" because it does not fulfill the more concrete criteria for being a "thing":  i.e., a process, machine, manufacture, or composition of matter.  Because software claims often fail the first step of the Mayo/Alice inquiry, the inquiry is next directed to the second part, which determines whether the patent claims add "significantly more" to bring them back within the fold of eligibility.  However, meeting this second step has proven challenging in the court of law when it comes to determining patent eligibility of modern computational tools.

The recent Federal Circuit case of BSG Tech LLC v. BuySeasons, Inc. reiterates these challenges, but also what is needed to overcome them.  BSG Tech sued BuySeasons for infringement on claims for the use of "self-evolving generic indexing of information" (U.S. Patent Nos. 6,035,294, 6,243,699, and 6,195,652).  Indexing software is used to organize and search information based on various parameters and values.  The problem, BSG Tech points out in their '294 patent is that current indexing methods are incapable of sorting through the enormous amount of information available on the internet in an efficient and accurate way.  For instance, if one wants to search for a specific website one most likely would need to know part of the URL unless the subject for the webpage is otherwise included in the web address.  Otherwise, one may choose to search by keywords.  However, keyword searching can restrict the search to subjects indexed by that keyword (e.g., searching for "red" car would provide only cars that were identified as "red" and not magenta or crimson cars).  Additionally, a search for red Mercedes™ might also retrieve irrelevant information such as "a story about a woman named Mercedes wearing a red dress."  To combat this issue of specificity, online stores often use hierarchal indexing to narrow searches based subject themed drop-down menu selection (e.g., All, Books, History, Military, strategy, The Art of War by Sun Tzu).  And yet, applying these current "generic" indexing to large databases or searching the internet would require a substantial amount of effort and cost not to mention, a massive list of search parameters since search criteria can vary greatly by subject matter (i.e., make and model are more appropriate for vehicle searches rather than houses or history books).

Therefore, BSG Tech's '294 patent proposes a "self-evolving generic indexing of information" that would permit users to access search parameters based on what other users may have previously conducted.  BSG argues this method allows users of varying degrees of knowledge to improve search parameters for the benefit of everyone across the network.  As an example, users may access historical usage which would provide the frequency of parameters and values used previously by other users to index different subjects.  BSG asserts that their new software provides a significant improvement over current prior art databases.  However, the Court concluded that these improvements lacked novelty and did not actually add "significant more."  Although, the definition of "significantly more" is technically a "question of law," the idea is that the claims provide more than conventional or routine standard.

Although BSG described the use of "summary comparison" or "relative historical" as guiding parameters in their indexing software, the Court viewed "increasingly narrowing classifications" as inadequate grounds for inventive concept or as a significant improvement upon the current standard.  In the past, other cases have had more success arguing the validity of their "abstract idea."  In Enfish LLC vs. Microsoft Corp., the Court overturned the District Court's decision that Enfish LLC's claims for a "self-referential table" was patent ineligible (U.S. Patent Nos. 6,151,604 and 6,163,775).  For background, Enfish established a logical model that allowed users to create data tables based on a "self-referential" system where by all individual data pieces could be referenced to each other within the same table.  In the conventional relational database each entry was referenced individually by type and not as a collective whole.  The difference with this case was that Enfish made an effective argument that their claims made significant improvements over the generic method which required individual tables for each comparison of data.  Instead, their method allowed users to (1) create one table to input and reference data which (2) increased the efficacy for data storage and (3) data retrieval.  It was later determined, with the overturn of the District Court's decision, that the claims instead described an inventive strategy for indexing information that improved upon the current functionality of computers and is therefore, not directed to an abstract idea.

Determining the eligibility of patents is not as strong and fast a rule when it comes to abstract lines crossing into concrete means.  With the case of software programs, improving upon a computational machine to facilitate a function may revolutionize an industry, manufacturing of a substance, or provide the ability to accurately measure simultaneous variables at once to execute a process.  With BSG, if their product has significant utility like presented in Enfish, their invention of "self-evolving generic indexing of information" may have had a shot at being perceived as more than just an abstract entity and instead, a marketable way for multiple users, across a network, to augment a reference database.  However, the claims were seen as an insignificant improvement on conventional indexing standards and denied patent eligibility.  Perhaps the speed to which software has been incorporated into our daily lives surpasses our current ability to conceptualize the validity of abstract inventions unless they are presented with concrete strategies.  The Mayo/Alice test is clearly not fully evolved.

[1] Turing, A. M. (1937), On Computable Numbers, with an Application to the Entscheidungsproblem. Proceedings of the London Mathematical Society, s2-42: 230-265. doi:10.1112/plms/s2-42.1.230

* Brittany Knight is a Ph.D. Candidate in the Biomedical Sciences Ph.D. Program in the Neuroscience Department at the University of Connecticut.  Prior to attending the University of Connecticut, Brittany obtained her B.S. in Psychology and a minor in Biology with Global Honors with Distinction from Lock Haven University of Pennsylvania.
** Dr. Sabatelli is a Partner with Dilworth IP

Comments

[EG]

"The Mayo/Alice test is clearly not fully evolved."

Hey Brittany and Anthony,

I disagree: the nonsensical and broken Mayo/Alice framework is unworkable. As I've said repeatedly, I could do better using a Ouija board than to apply this unworkable framework. This Frankenstein monster was created by SCOTUS. They should never have created it in the first place. But their failure to at least fix it by defining what they mean by "abstract ideas" so us "mere mortals" can even understand what this framework is represents appalling conceit and arrogance on their part.

[Paul F. Morgan]

Alice-Mayo complaints would get more sympathy if not so often presented in cases with obvious obviousness issues as broadly functionally claimed. Isn't a "self-evolving generic indexing of information" that would permit users to access search parameters based on what other users may have previously conducted been part of what Google and others have been providing for internet search result listings from their beginnings?

[EG]

"Alice-Mayo complaints would get more sympathy if not so often presented in cases with obvious obviousness issues as broadly functionally claimed."

Paul,

There might be fewer such complaints if SCOTUS would do its job, and tell us "mere mortals" what they mean by "abstract idea." Instead, as SCOTUS brazenly did in Alice, as they did before in Bilski, they simply refuse to do so. The fact that they interjected "inventive concept' into that framework simply compounds their error by conflating 101 with 103. So if obviousness is the "real issue," stick with resolving it under 103, not under 101.

[Skeptical]

I would posit the opposite of:

"Alice-Mayo complaints would get more sympathy if not so often presented in cases with obvious obviousness issues as broadly functionally claimed."

These types of cases for which other sections of the law would EASILY work only all the more show the infirmity in using 101 as the "catch-all" it was never intended to be.

That is, if one understands what 101 was intended to be (by Congress).

[Paul F. Morgan]

Skeptical [anon], by now readers should be well aware that Alice and Mayo join earlier Sup. Ct. judicially created unpatentable subject matter exceptions, as exceptions to 101, NOT based on the language of 101.
They should also be well aware of why it is not true that "other sections of the law would EASILY work." Because they typically require an entire very costly trial and appeal, whereas unpatentable subject matter is a preliminary matter that can be raised by motion on the pleadings before any other trial expenses even start.
It does not help to confuse valid criticisms of this practice with its existing realities.

[Martin Snyder]

My modified comment from a recent thread:

An abstract invention is not the same thing as an abstract claim.

An abstract invention is an eligibility problem. An abstract claim is a patentability problem.

All information is intangible, but some information is not abstract.

Information consumed by human beings is always abstract. Without a human being, there can be no abstraction.

A method which gains its utility via the processing of some species of information and whose result is new and useful information is an information invention.

Information inventions where the utility arises by non-human consumption of information should be eligible inventions.

Information inventions where the utility arises by human consumption of specific information should be ineligible for patenting.

Abstract claims mean the claims are too abstracted from the invention, and thus do not encompass any particular invention. This abstraction can arise from obviousness, failure to demonstrate complete grasp of the invention, or the use of components in their designed capacity to produce variations, not innovations.

What did I miss?

[Skeptical]

Mr. Morgan,

You have confused ease of expediency with ease of law.

I suggest that you recognize the difference in an attempt to gain (regain) a sense of Rule Of Law.

[Paul F. Morgan]

"Skeptical," as with quite a few other commentators, on this subject you are confusing what the present law and practice is with what you would personally like it to be.
EG, yes, the Sup. Ct. left the term "abstract" in undefined confusion. But absent Congressional action, that leaves it up to the lower courts [here, the Fed. Cir.] to work that out. [Which, BTW, is not unusual for Sup. Ct. decisions.]

[Skeptical]

Mr. Morgan,

You err in presuming that, on this subject, I am confused at all, much less confusing what the present law is with what I would personally like the law to be.

You assume that a current state of what "is" IS a proper view of what the law IS. As is typical with glossing over fundamental disconnects, such is driven by an alignment with the Ends reached, and a closure of critical thinking as to the Means used to arrive at those Ends.

Such a "Ends justify the Means" shortcut view of what IS does a grave disservice to ANY understanding of the Rule of Law.

ESPECIALLY in law (and the Rule of Law), such thinking (or lack thereof) is a grievous fault, not an attribute to be ignored, much less adhered to.

This critical thinking state is "thing" that I personally want things to be.

First step: recognizing when a broken scoreboard condition exists.

It is only after such recognition is obtained, can one then think critically about that condition and how that condition affects the Rule of Law.

Let me note a particular Rule of Law evidences in the last paragraph of your most recent comment:

To the notion of "not unusual" for lower courts to "work out" a new Rule of Law written by the Supreme Court is a truism. I doubt that anyone would argue with the truism itself. But that does not address the larger Rule of Law that is implicated with any such type of "working it out."

Do you understand why?

Here is a hint: the judicial branch does not have unfettered ability to engage in the law writing that comes from the Common Law law writing approach.

That type of power is not in place in the U.S. legal system.

Is there such a thing as Common Law law writing power?

Absolutely.

Is that power unlimited?

Absolutely not.