Alice at Court: Stepping Through the Looking Glass – Part II

An illustration of Alice by Lewis Carroll circa 1887.

In Part I, I addressed many of the issues that are being considered by the Supreme Court in the Alice v. CLS oral arguments, and continue the analysis here, looking at particular issues and what various amicus briefs argue, and whether those arguments hold up or not.

The Equivalence of Software and Hardware

There is a further gulf between those who view In re Alappat as sound logic and engineering (ABL, AIPLA, Alice, Mr. Ronald Benrey, BSA, CCIA, Mr. Dale Cook, Prof. of Computer Science Lee A. Hollaar, IEEE-USA, Microsoft) and those who it as mistaken (Electronic Frontier Foundation, Prof. Robin Feldman, Red Hat) and primarily responsible for an increase in such patents (Electronic Frontier Foundation, Google, “Law, Business and Economics Scholars”). The IEEE-USA provides an excellent analysis of the relationship between software and hardware, pointing out the incontrovertible principle of equivalency, that “special-purpose programming of general-purpose hardware” is “equivalent to special-purpose hardware,” though IEEE-USA fails to mention that this is a fundamental principle of computer science, as established by Alan Turing in the 1930s. To assert, as does the EFF, that the Federal Circuit “concocted” the equivalency of hardware and software goes beyond denying the foundational work of Turing and others.  The equivalency of software and hardware is what makes it possible for Java to run on any type of computer using the Java Virtual Machine, as well the electronic design automation industry, which enables complex electronic circuits to be entirely designed in software before being implemented in hardware.

The Presumption of Validity and Whether Section 101 is a Threshold Determination

CLS and RichRelevance argue “[t]he evidentiary standard of proof applies to questions of fact and not to questions of law,” citing Microsoft Corp. v. i4i Ltd. P’ship, 131 S. Ct. 2238, 2253 (2011) (Breyer, J. concurring). Accordingly, because patent eligibility is a question of law, the presumption does not apply. CLS argues that even if the presumption of validity applies to Alice’s patents, it has been overcome. The Solicitor General takes a different stance. Relying on the majority opinion in Microsoft, rather than Breyer’s concurrence, the Solicitor General expressly refutes CLS and argues that “the presumption of validity applies to all invalidity defenses,” and that under Mayo, the judicial exceptions to Section 101 are an invalidity defense. That said, the Solicitor General agrees with CLS that the presumption is overcome.

ABL argues for a presumption for validity on different grounds. “The § 101 preemption analysis centers on the practical, real-world effects of the claim,” according to Judge Lourie. If so, then real evidence of whether the claims preempt “all practical application” in the “real world” is necessary. Thus, like the questions of law of enablement under §112 and obviousness under §103, patent eligibility is a question of law based of factual predicates, and therefore the presumption of validity must apply as well. Professors Menell and Lefstin (while supporting CLS) make a similar point: “Courts would likely need to hear from expert witnesses to determine what is ‘well-understood, routine, conventional activity,’ Mayo, 132 S.Ct. at 1299, presumably as of the time that an invention was made.”

The amicus briefs spar over whether eligibility must or should be decided first in a litigation context as some sort of “threshold” test. Here, we have a great example of interpretation being in the eye of the beholder. CLS argues, and many of its supports agree, that “Section 101, as construed by this Court, performs a ‘screening function’ (Mayo, 132 S. Ct. at 1303) that is an important ‘threshold test’ for patentability.” Of all of the briefs, the brief for The Clearing House Association has the most extensive focus on this question, arguing that Section 101 “should be enforced rigorously and assessed early in litigation,” “before other issues, such as claim construction or obviousness.” The Retailers’ brief, filed on behalf of over twenty national retail companies, argues that §101 should be decided on a motion to dismiss before claim construction, discovery or any other question of fact, to “provide some measure of relief to Retailers because they will not have to spend millions of dollars and wait until all appeals are exhausted to resolve this gateway issue,” a perfect example of special pleading if there ever was one. Google, RichRelevance and EFF among others agree with these arguments.

Apart from the pure interest in having a means for getting out of litigation quickly, it’s not entirely clear that the “early” interpretation of “threshold” is correct. Both sides of the debate cite Bilski for the proposition that “the § 101 patent-eligibility inquiry is only a threshold test,” Bilski, 130 S. Ct. at 3225, and Mayo, 132 S. Ct. at 1303, that §101 provides a “screening function.” However, being a “threshold” or “screening function” does not imply that this must be decided first before other sections of the patent act. Section 102 imposes a “threshold” of novelty, and certainly Section 112 imposes a “threshold” of definiteness, and both “screen” out defective patents in unique ways. Further, if there is no presumption of validity, and if §101 is indeed a threshold that must be crossed, then this would imply that in every patent litigation the parties must either brief and argue patent eligibility or stipulate to it, before anything else. That seems contrary to the expressed goal of using §101 as a fast and less expensive way to terminate patent litigation.

Interesting Ideas

A number of amicus briefs proposed interesting and perhaps radical ideas. IBM suggested getting rid of the abstract idea analysis entirely, and relying on Section 103 obviousness. IBM argues that “Federal Circuit’s varied attempts to apply abstract idea precedent in this case demonstrate the doctrine’s failure in the computer-implemented invention context.” This is because “[u]nlike laws of nature and natural phenomena, which exist in nature and are discovered, abstract ideas do not exist in nature and are instead created.” “The end result is that any application of the abstract idea exception involves an undisciplined parsing and rewriting of the claims such that the court’s ultimate analysis of the patent eligibility of an invention…quickly devolves into a subjective and free-floating inquiry that defies predictable results.” While IBM has perfectly captured the present state of dysfunction and offered a fresh start, I doubt that the Court is willing to go so far as to junk Section 101 entirely. I do applaud IBM for not being shy about the source of the problem: “The blame for that failure does not fall on the Federal Circuit alone. The “guideposts” developed by this Court over the years, and supported through recommendations by well-intended amici such as IBM…have after decades of struggling failed to yield a formulation of the abstract idea exception that can be administered to computer-implemented inventions with any degree of predictability.”

Tony Dutra, a former lawyer, software executive, and now legal Editor for Bloomberg BNA’s Patent, Trademark, and Copyright Journal, proposed a “useful arts” test, based on Section 101’s “useful” requirement. “A “new and useful art” should then be read to mean either (a) an invented system or new rules for applying an existing trade, or (b) a system of rules specifying a new trade.” Business, law, medicine, teaching and politics are at best “learned arts,” not trades subject to patents. Computing itself would be a trade (along with ancillary technologies, such as robotics), but if a claim is “directed to another art that is other-than-useful (e.g., business), then it is not patent eligible: “logic processing” that “searches a dictionary of words to suggest alternatives for alliteration,” or “allows a user to fill in a tax return” is not “useful” in the status sense and would not be patent eligible. It would seem that this approach would wipe out a large number of patents in a wide variety of fields, such as teaching, healthcare management, finance, medicine, logistics, advertising, games, amusement devices and so forth that do not meet the historical definition of a trade.   This another species of the technological arts test, and unlikely to be adopted by the Court.

The Computer & Communications Industry Association suggests that to be patent eligible, a claim for a computer-implemented invention must be “tied to particular computer hardware, such as a particular type of desktop computer, a particular configuration of computer servers, or a computer system embedded in a device (like a car).” Though they do not give specific examples, it would seem that they want to see specific operating systems (e.g. Microsoft Windows, Linux) and other brand or type details in the claims themselves: “Some of the relevant characteristics of computer hardware include: processing speed, number of central processing units (CPUs), size of computer memory, non-volatile data storage, distribution of software components and/or data among computers connected by a network, operating system, programming language used, and/or required software that must be installed for the invention to work, e.g., a web browser, a particular word  processing program, etc.” This specificity is necessary to “support the goal of limiting patent claims to what the inventors actually invented.” If this is the goal, then the rule should apply for all inventions, requiring a mechanical or electrical claims to recite dimensions of parts, size and speed of motors; the types of fuels; exact electrical parameters in circuits (e.g., resistance, voltage, capacitance values) and so forth. Clearly, this is something we don’t do and is entirely contrary to the idea that claims are not limited to the exact embodiments disclosed.

Favorite Quotes

A few of the briefs had some very artful statements which deserve special mention. Charles Hiekken writing for Judge Michel, has this great remark: “If claim can be sliced-and-diced, however, it is likely that each element of the claim is, by itself, old and conventional, and is preemptive of a truly broad range of implementations, as no one works from scratch but God.”

Steven Sereboff, on behalf of the Conejo Valley Bar Association, gets my vote for the most colorful metaphor: “Patents are not diamonds; they do not last forever. They are more like cabbage, which takes some time to grow, but will eventually return to base organic matter.”  May the Court’s own opinion be as vivid and memorable.


Warning & Disclaimer: The pages, articles and comments on do not constitute legal advice, nor do they create any attorney-client relationship. The articles published express the personal opinion and views of the author as of the time of publication and should not be attributed to the author’s employer, clients or the sponsors of

Join the Discussion

2 comments so far.

  • [Avatar for step back]
    step back
    March 29, 2014 01:38 pm


    One would think that a “machine” is a machine.

    But by the time the word-spinning laureates finish doing their kabuki dance, you would start thinking that day is night and under the bridge there hides a fright. (The mythical T-word creature)

  • [Avatar for NWPA]
    March 29, 2014 01:15 pm

    I think the problem with this analysis is that the justices are not educated in science or scientific thinking. So many of these arguments made by those that want to broaden abstract rely on the reader being ignorant. For example, software and hardware are equivalent, and yet the anti crowd vigorously deny this or just ignore that this renders their arguments absurd. Another example, is Lourie discussing a paper and pencil test when the Church-Turing Thesis renders this argument absurd. Another example, is J. Tarantula trying to hold that building a computer system to simulate human thought is per se ineligible. That last one is an offense to science and our society.

    Another big picture item that illustrates that the justices are not qualified to render a decision is that none of them can even conceive of the fact that what they do in their jobs will one day be done with software. How could it be that a machine to perform their functions would be ineligible for patentability? That last fact is shameful. Each of the justices should be deeply ashamed of themselves for their ignorance. And, then there are the Louries who know they are ignorant and don’t care. They just want to legislate and don’t care about the law. Lourie should be impeached.

    I am getting closer to a prediction. One point: if the machine is sitting there and performing useful work, then what set of claims would be OK? Another point, is that not only is the Turing machine absent from these arguments, but the conservation of information is as well. Basically, information processing is best understood as a transformation of represented information that takes time, energy, and space. The conservation of information is the most important law in physics.

    One final point: if these functional claims are not OK, then what claim language is OK to include the 100’s of ways that one skilled in the art knows how to solve one function of a claim? Again, we have outrageous behavior from Lemley in not offering any real analysis or solution. Lemley is no better than an inquisitor and his vanity paper (published in the vanity press the Stanford Law Journal.)