How to Patent Software in a Post Alice Era

Software technology - concept. software technology golden metallic cog gears. software technology on golden metallic cogwheels. software technology - illustration with glow effect and lens flare. 3d.Earlier this year, after a long winter of despair for patent owners on the patent eligibility front lines, hope was springing. Surely there would soon be another case from the Federal Circuit that would find at least some software patent claims to be patent eligible, right? After all, the Federal Circuit was starting to see patents on innovations of an entirely different class of sophistication. Automation of a lip synchronization process for three-dimensional characters certainly cannot be patent ineligible, can it? If the Federal Circuit did not start to find at least some software patent claims patent eligible the realization would need to set in that you could not patent software in the United States because software was de facto patent ineligible.

In May 2016 that hope was realized with a decision in Enfish LLC v. Microsoft Corporation, but then hope was quickly dashed just several days later as the court returned to its all too familiar Soup Nazi like refrain of “no patent for you” whenever software was the subject matter. But as the summer progressed we saw several more decisions that brought hope, a trend that has followed into the fall.

A great deal has been learned about what needs to be done in order to demonstrate that a computer implemented innovation (i.e., software) is patent eligible subject matter over the last six months. Despite the fact that there are some judges on the Federal Circuit that believe software is not patent eligible, the good news is that view is now a clear minority view.

First, let’s start by acknowledging that the Supreme Court has never said that software is patent ineligible. Quite to the contrary, the Supreme Court has left the door open on software patents, even finding software to be patent eligible in Diamond v. Diehr, a decision that the Robert’s Court continues to insist remains good law. The Supreme Court has also unequivocally stated that business methods are patent eligible. See Bilski v. Kappos. 

Second, after the last six months we have now seen Judges Moore, Taranto, Hughes, Chen, Newman, O’Malley, Reyna, Stoll, and Plager all sign on to decisions that found software patent claims to be patent eligible. That brings the total to nine (9) judges of the Federal Circuit indisputably in favor of patent eligibility for software in at least some instances over the last six months. Judge Wallach joined Judge Chen in the majority decision in DDR Holdings v., so it would seem safe to say that he too remains in favor of patent eligibility in at least some cases, bringing the number to ten (10). Not on that list is Judge Linn, a former patent attorney, who has throughout his career has been one of the most forceful defenders of a strong patent system and the author of previous decisions finding software claims patent eligible, including the original panel decision in Alice. Thus, the number of Federal Circuit judges that would find at least some software patent claims to be eligible conservatively sits at eleven (11).


The lone predictable “no votes,” which in all cases would be virtually certain to find any and software claims patent ineligibility are Judges Dyk and Mayer. The remainder of the court is somewhat difficult to predict at this time as the law is quickly evolving and better patentee cases are arriving at the court. Judge Lourie, also a former patent attorney, has generally throughout his career been willing to find appropriately drafted claims to be patent eligible, and would have been a predictable vote for patent eligibility in at least some cases prior to Mayo, Myriad and Alice. Chief Judge Prost authored an opinion finding method claims in a life sciences patent to be eligible, but will that carry over to software?

Even with the clear “no votes,” and those that have not yet voiced an opinion during this relatively new Federal Circuit trend, it is safe to say the tide has clearly turned at the Federal Circuit, and much has been learned in recent months. To be perfectly honest, at least as much has been learned from cases where the Federal Circuit has found the claims patent ineligible (i.e., TLI Communications and FairWarning IP) as from those cases where the claims were found to be patent eligible (i.e., EnfishBASCOM, McRo, and Amdocs). Those of us in the software space also even received some potentially insightful help from that aforementioned life sciences case — Rapid Litigation Management — which explained that while preemption is not the test for eligibility any more, the fact that a claim can be designed around confirms that the claim is not drawn to a judicial exception.

In a nutshell, if you are going to write a patent application in such a way that the reader will be left wondering what the innovation is, what the problem being solved is, or the technical particulars on how the innovation actually solves the problem and achieves the specified functionality, you should not expect a patent. In other words, if you write your patent applications without actually defining the problem, explaining the technological solution and how it is implementing the desired functionality described in the specification, and how what you are claiming is an improvement (or at least unconventional), you will not get a patent because the claims will be patent ineligible.

On the other hand, if you write your patent applications to describe (and claim) an invention that is adequately described so that someone of skill in the art will actually understand what is innovative (i.e., how and why), with a thick technical disclosure and explanation as to how computer functionality is being improved, or even generic components are working in unconventional ways, then you will get a patent because your claims will be patent eligible. [1]

What this means is if you want to patent software you simply cannot write your patent applications so that an English major or History major will be able to understand the application from start to finish. If you write your application so that an technologically unsophisticated reader can understand everything from start to finish then you have failed. That, of course, does not mean I am suggesting you hide or otherwise obfuscate the invention, but it does mean that your patent application needs to be written for the proper audience, which is a technologist skilled in the computer/software arts.

It is perfectly accurate to say, as I’m sure some will, that there is no technical standard in U.S. law. While that is true, if you are not writing your applications to satisfy such a technical standard you are not only going to have problems in the U.S., but you are going to have problems elsewhere, which is why Microsoft has settled on the European technical standard as the guide for how to draft software patent applications. If it is good enough for a top patenting company like Microsoft then shooting for a technical standard reminiscent of the European standard should be good enough for you, and in fact you should consider it to be a best practice.

To accomplish the goal of satisfactorily describing a computer implemented invention with enough specificity to overcome the patent eligibility threshold it is imperative for inventors and practitioners give serious thought to describing the overall architecture of the system that will perform the desired function. Merely reciting the process steps in a way that is disassociated from the overall architecture of the system will simply not be enough. Thus, the first thing you need to do is not focus on the software, but rather, focus on that hardware and how the hardware will be connected. Some will think this harkens back to the machine-or-transformation test, or perhaps longer ago when computer software was not patentable at all and these innovations were protected by describing the machine performing the functions. Think what you want, but if you want to patent software your description must be of a concrete and tangible technological innovation, and that concrete and tangible technological innovation must come through in the claims.

Conceptually it is critical to understand that while the machine-or-transformation test is not the test for patent eligibility, it is exceptionally difficult to characterize something as an abstract idea when tangible hardware is present, particularly if that hardware is described to be working in unconventional ways. Further, even if the the claim is characterized as an abstract idea the presence of tangible components can and does add “significantly more” than the abstract idea so as to render the claim patent eligible. This has lead many to observe that the easiest path to patent software is to make sure your software is adequately supported by hardware components.

It is also essential to guard against the tendency to want to describe the software from the perspective of the end user. In truth, descriptions that focus almost exclusively on what the end user perceives are responsible for many of the problems facing software patents over the last 5-10 years. While this information certainly should be in any application, it is not sufficient to describe things purely from the user’s perspective. The description that is essential is an explanation of how the software operates from the perspective of the computer, not the perspective of the user. This is a subtle but extremely important distinction. If you describe things only from the user’s perspective you are not describing what happens on a technical level. How things happen might as well be black magic. What you will need to do is explain how the software operates to achieve the desired results. In order to do this it will be absolutely necessary for you to break down the software step by step so that a computer programmer will be able to create the code necessary. What this means is that you must describe the logic that the computer programmer needs to follow.

Will all of this in mind, it would be useful to review the patent eligibility cases decided by the Federal Circuit over the last six months.

CLICK HERE to CONTINUE READING… Up next, a summary of significant Federal Circuit patent eligibility decisions over the past six months. 



[1] Of course, that assumes patent examiners actually start following the law and the below summarized Federal Circuit cases, which I understand is a mighty big “if” given so many in this space openly profess that they work for the patent denial authority and are all too proud to tell applicants and patent practitioners that they have not issued a patent since the Supreme Court decided Alice v. CLS Bank and they simply will not start with your application regardless of what senior management says or the Federal Circuit commands.


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

13 comments so far.

  • [Avatar for Edward Heller]
    Edward Heller
    December 22, 2016 11:59 am

    bruce, somewhat agree. The rush to the patent office needs to be avoided in favor of reduction to practice. Real solutions that work need to be described.

    Also, we need to keep in mind that the purpose of the US patent system is to promote invention in the United States.

  • [Avatar for bruce]
    December 21, 2016 04:56 pm

    I think its time for a major reform of the patent system. It doesn’t seem like it is serving its intended purpose: “To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries.”

    Somehow the system has morphed to the point where progress in useful arts is being impeded by patent trolls and actual authors and inventors can’t untangle the legal minefields surrounding patent disclosure. It is a world where one wrong word can doom an otherwise brilliant patent while carefully crafted legalese can protect stupidly obvious ideas.

    The degree of patent protection incurred by the system should strive to be proportional to the amount of Science and Art that it promotes, so a patent that clearly shows the Inventor identified a real problem and undertook real effort to understand it, and solve it should incur more protection than a patent that deals with hypotheticals and vagaries and unsupported promises. Most of the time the devil is in the details, so we need a system that rewards killing these devils.

  • [Avatar for Independent Inventor]
    Independent Inventor
    November 19, 2016 08:39 pm

    Night Writer — first let me thank you for your years of support for American inventors and their inventions.

    Next; I see great value in such a “Lemley’s Lies” paper being made available to all. I know I’d like to read and have a copy of it.

    Have you thought about seeing if Gene would be willing to publish it here for the benefit of all? You could of course share it under a pseudonym; as others have been permitted to do.

    Would be great to have all of Lemley’s baloney in one place.

  • [Avatar for Night Writer]
    Night Writer
    November 19, 2016 02:11 pm

    @9 Anon. I absolutely agree. And, I believe that Mark Lemley should be brought up on ethics charges before Stanford for his intentional misrepresentations, which includes cites that do not support what he claims they do.

    And, if the coward ever wants to really debate me about it and not hide behind the privileges he enjoys as a professor at Stanford, then I have a paper all ready to go about his intentional misrepresentations.

  • [Avatar for Anon]
    November 19, 2016 07:43 am

    The use of terms sounding in function was fully allowed by Congress in the Act of 1952.

    The option of using 112(f) is for when a claim element is presented in TOTALLY functional language.

    The use of terms sounding in function happens – and is permitted to happen outside of 112(f) when the claim element has some structure.

    It is utterly reprehensible that this distinguishing fact is repeatedly attempted to be obfuscated. with some sort of antiquated (and not supported in law) use of a “Point of Novelty” to dissect a claim and attack terms sounding in function when such terms when used in a greater claim element are fully allowed under the law as written by Congress.

  • [Avatar for Night Writer]
    Night Writer
    November 19, 2016 06:08 am

    1) Scope of enablement.

    2) Functional language is what people skilled in the art use to describe a set of solutions that are known or enabled by the disclosure. That is reality and Lemley should be disbarred for intentionally misrepresenting this in his vanity publication about functional claiming.

  • [Avatar for Simon Smith]
    Simon Smith
    November 19, 2016 05:28 am

    Well, those of you who know about Australia and RPL Central – the panel of judges hear the word computer, and that’s the end of it. The Full Federal Court erred so much that a 3 year old could look at the patent specification and read the ruling they handed down and see how they completely misordered the way the patent was applied and where. They mistakenly created their own view despite a simple data flow diagram showing preprocessed data, yet in their findings claiming the data was processed via a browser via the user and that this was the method of manufacturer. The donut eating useless judges could not be overturned because the other donut eating useless judges in the High Court exercised our wonderful constitution to ignore the law and not hear an appeal, because it was not in the public interest. How’s that for justice?

  • [Avatar for Edward Heller]
    Edward Heller
    November 18, 2016 05:46 pm

    anon, it will remain true for awhile that anyone who has views on 101 will have opposition. Hopefully, though, we can get a Director and advisors who will act to dismantle post-grant revocations, and will do something to reduce the cost and delays of obtaining a patent. As Gene has pointed out, it is expensive to make every issued patent bullet proof, and trying to make every patent bullet proof simply makes the patent system affordable to many.

  • [Avatar for Anon]
    November 18, 2016 07:17 am

    With no intended sleight, the only people from your list that I would recommend are Ron Katznelson and David Boundy. Most definitely you would not make that list, Mr. Heller, and certainly not Mr. Armitage. Gene and Curious are close, but I think that Mr. Katznelson and Mr. Boundy would capture their viewpoints more or less.

  • [Avatar for Edward Heller]
    Edward Heller
    November 17, 2016 07:44 pm

    And, might I add, any new Director should call a summit where the likes of Gene, and perhaps Curious, Ron Kaztnelson, David Boundy, me and others of our thinking might discuss ways of restructuring the patenting/examination process as a whole with a view to reducing costs and making the whole system more reliably pro-patent and pro-inventor. We might reserve one chair for Armitage, as his views are always welcome.

  • [Avatar for Edward Heller]
    Edward Heller
    November 17, 2016 06:19 pm

    While he is overqualified for the job, I recommend Adam Mossoff as the next PTO Director. He supported the petitions of Cooper and MCM with briefs arguing that patents were property and not public rights. This is the kind of guy I want in the government advising on patent policy.

  • [Avatar for adsfsdfds]
    November 17, 2016 04:35 pm

    The patents might be valid if they describe something more than:
    “A method for analyzing voice recordings, consisting of: recording someones voice; determining parameters related to the recording; based on the parameters, generate a list of words associated with the recording”

  • [Avatar for angry dude]
    angry dude
    November 17, 2016 01:54 pm

    I remember reading many patents and patent apps from Microsoft many years ago, mostly voice recognition related

    For some reason all of them included detailed description of a general purpose PC computer (yes, with nice pictures of keyboard, mouse, display, peripherals etc – as ridiculous as it looks )

    What about today ?

    Are those patents invalid regardless of tech (mostly algorithmic techniques) they describe (and claim as part of software running on general purpose PC)?

    Does everybody realize just how stupid this whole situation became ?