Despite Misleading Question, Berkheimer Case Has Legs at the Supreme Court

by Dennis Crouch

The Supreme Court has repeatedly held that patent eligibility is a question of law. However, in Berkheimer, the Federal Circuit added some nuance — writing that the issue can include underlying issues of fact. Judge Moore explained:

Berkheimer and Aatrix stand for the unremarkable proposition that whether a claim element or combination of elements would have been well-understood, routine, and conventional to a skilled artisan in the relevant field at a particular point in time is a question of fact.

Berkheimer v. HP Inc., 890 F.3d 1369 (Fed. Cir. 2018) (en banc denial). In the lawsuit between Mr. Berkheimer and HP, the result of the underlying-material-fact revelation meant that the case was improperly dismissed on summary judgment. Rather, on remand, the court will need to be presented evidence and make appropriate factual conclusions before ruling on the eligibility question.

In its petition to the U.S. Supreme Court, HP does not directly confront the Federal Circuit holding regarding underlying facts – but instead directs the the court to a non-decided question about whether Berkheimer makes eligibility a jury question.

HP v. Berkheimer Question: The question presented is whether patent eligibility is a question of law for the court based on the scope of the claims or a question of fact for the jury based on the state of the art at the time of the patent.

In my view, the question presented by HP’s attorneys intentionally creates confusion as the Federal Circuit did not rule (1) that eligibility is a question of fact; or (2) that any aspect of the eligibility question goes to a jury.  Rather, the Federal Circuit has repeatedly ruled, including in Berkheimer, that the question of “eligible subject matter is a question of law.”  I’ll also note that no court following Berkheimer has seen eligibility itself as a question of fact or attempted to get a jury verdict on the issue.  I contacted David Salmons (counsel of record for HP) to see what I’m missing about the misleading question, but he did not respond.

Despite my misgivings about the actual question asked, the petition implicitly raises important questions of (1) whether eligibility may be based upon underlying questions of fact; and (2) the relevance of  the “state of the art” in eligibility analysis (i.e., what is an “inventive concept”).

The petition has some legs — with six amici briefs and a request from the court for a responsive brief from Berkheimer:

  • Brief amici curiae of Electronic Frontier Foundation and R Street Institute
  • Brief amici curiae of T-Mobile USA, Inc., and Sprint Spectrum L.P.
  • Brief amicus curiae of Computer & Communications Industry Association
  • Brief amicus curiae of Engine Advocacy
  • Brief amicus curiae of Askeladden LLC
  • Brief amici curiae of Check Point Software Technologies, Inc. and CableLabs

Of these briefs, Askeladden’s brief from Carter Phillips has an important section walking through jurisprudence on adjudicative facts vs legislative facts.  Engine Advocacy’s brief was filed by Stanford Law Clinic Director Phillip Malone and does a good job of explaining how Post-Alice 101 is good at ending lawsuits early and cheaply, and that Berkheimer is upsetting that system. CCIA’s brief does a good job of walking through the ways that courts have been dealing with eligibility questions on the pleadings and at summary judgment.  I have not fully read the others.

Mr. Berkheimer’s brief is due December 5, 2018 — We’ll do another post at that point talking through the various positions and the variety of issues raised in this important case. [Full Docket with Briefs].

= = = = = =

The patent at issue here is U.S. Patent No. 7,447,713, which issued back in 2008.  At that time, I hardly even talked about eligibility in my patent law class because the door was so wide open.  The ‘713 patent is broadly directed to “archiving and outputting documents or graphical items.”  The Federal Circuit found claim 1 was properly held ineligible, but ruled that there were some underlying factual disputes about whether claims 4 and 5 crossed into the eligibility threshold.

Claim 1. A method of archiving an item comprising in a computer processing system:

presenting the item to a parser;

parsing the item into a plurality of multi-part object structures wherein portions of the structures have searchable information tags associated therewith;

evaluating the object structures in accordance with object structures previously stored in an archive; [and]

presenting an evaluated object structure for manual reconciliation at least where there is a predetermined variance between the object and at least one of a predetermined standard and a user defined rule.

Claim 4. The method as in claim 1 which includes storing a reconciled object structure in the archive without substantial redundancy.

Claim 5. The method as in claim 4 which includes selectively editing an object structure, linked to other structures to thereby effect a one-to-many change in a plurality of archived items.

42 thoughts on “Despite Misleading Question, Berkheimer Case Has Legs at the Supreme Court

  1. 7

    Yeah, the question is misleading in that avoids the ONLY question anyone really cares about–WHO decides the “factual” issues under the “inventive concept” (second) prong of Alice? Judge or jury?

    HP had to phrase the question misleadingly because the Berkheimer case didn’t actually address the judge-or-jury question, so it’s actually not ripe for Supreme Court review in this case. The question desperately needs an answer, so HP obviously felt, let’s see if this case can become the vehicle to resolve that through a false fact vs. law dichotomy.

    The truth is, whether the second prong of Alice can raise “factual” issues is a bogus question. OF COURSE IT CAN.

    Even claim construction, the most paradigmatic “question of law” in the patent world, can involve underlying questions of fact. But those questions are still decided by the judge (and subject to “clear error” review under Teva), as part of the overall legal evaluation of claim construction. Those questions do not go to the jury.

    But the Berkheimer line of cases has created a real crisis at the district court level because no one knows who decides disputed “factual questions” under Alice Step 2. Bench trial before the district court (like claim construction), or do they go to a jury (like the factual questions underlying obviousness, another “question of law”)?

    The few district courts confronting this question seem to have taken the latter view, and there have been cases where juries are literally asked to decide whether the technology is routine, conventional, etc., under Alice step two. In the few cases where this has happened, this becomes totally unworkable, as the question utterly devolves into a legal debate. There are no jury instructions on this issue, so juries are fed rote and sentences from Alice and a few contradictory decisions from post-Alice CAFC decision as their purported jury instructions.

    Personally, I think the second step of Alice needs to be resolved by the district court, like claim construction. I don’t think Alice step two is really like the three Graham factors for obviousness, which are themselves neatly presentable as factual questions. Obviousness is cleaner because the underlying Graham factors are entirely factual — only the conclusion is legal. The Alice step two analysis is too interwoven with the first prong of Alice (which is a legal question), and policy considerations like preemption, etc. It’s not really suitable for resolution by a jury, as the few cases where this has been tried, have demonstrated.

    1. 7.1

      the question is misleading in that avoids the ONLY question anyone really cares about–WHO decides the “factual” issues under the “inventive concept” (second) prong of Alice? Judge or jury?

      It’s not binary. Judges resolve obviousness issues all the time, appropriately. Sometimes it’s appropriate because there isn’t really an issue at all (this is true for many “do it on a computer” cases); sometimes it’s appropriate because no reasonable jury could find differently.

      1. 7.1.1

        This is not the point, MM. If there are disputed factual questions, the judge cannot resolve it on summary judgment or under the “no reasonable jury” standard. Berkheimer, in fact, held that the “inventive concept” issue could not be resolved on summary judgment for about half of the asserted claims because of factual issues. From that starting point, it is completely binary — does the factual question get resolved by the court or the jury.

    2. 7.2

      the question is misleading in that avoids the ONLY question anyone really cares about–WHO decides the “factual” issues under the “inventive concept” (second) prong of Alice? Judge or jury?

      I would also differ here and note that the only question anyone really cares about is whether the “inventive concept” can, itself, include ineligible subject matter (e.g., information identified by its content) or can be completely devoid of any physical/structural (non-abstract) distinctions with respect to the prior art. If the answer to those questions are “yes”, then the so-called “eligibility analysis” has already gone of the rails.

      The truth of the matter is that, in spite of the PTO and the CAFC’s silly efforts to create an over-arching one-size-fits-all test for eligibility that’s just not how it works, or is ever going to work, in practice.

      Until there is some clear test for which types of logic are eligible for patenting and which types of logic are not eligible for patenting (and don’t hold your breath on that one! LOL), the ultimate issue for claims that implicate data processing logic/transmission/storage will always be whether there is non-obvious structure or non-obvious physical transformation recited in the claims. This isn’t an “opinion” because I’m not making any assumptions. It’s the necessary logical result of the system we have for protecting innovation where we use words to describe those innovations. If you don’t like this result, the answer isn’t to hand-wave and create artificial “fake” structures for lawyers to argue about (ins @nity). The answer is to devise a separate scheme entirely for protecting logic “innovations”. The primary reason the maximalists haven’t made any effort in this direction is because the use of the patent system to protect logic results in an incredible grifting opportunity. It’s a farce, and it’s a farce that is easily exploited if you’ve got the time, money and inclination to exploit it. It’s less easy to do that now (thanks to the AIA and IPRs, which the maximalists fought tooth and nail against) but still way way way easier than actually doing some science, or writing bug-free software and marketing it.

      1. 7.2.1

        will always be whether there is non-obvious structure or non-obvious physical transformation recited in the claims.

        Your own beloved broken scoreboard rejected that (9-0) in Bilski.


          If Bilski said you could have an eligible claim where the claim fails to distinguish itself from the prior art on non-abstract grounds, then Bilski was wrong.


          The Court makes mistakes sometimes, as we all know. And sometimes the mistake the Court makes is not thinking deeply enough (even when very little thought is required).

          In Bilski, what was unanimously agreed upon was that Bilski’s claim was ineligible, and the easiest way to see that (also agreed upon, and very reasonably so) was to recognize that there was no recitation of a machine and no transformation of matter recited in the claim.

          To fail to mention this is a prime example of your intellectual dishonesty. But there are many such examples, as you and everyone else is very much aware.


            Your “easiest” and “very reasonable” was what the Supreme Court denied — by a 9-0 vote.

            You should know this, not in the least because you have been told this many times now.

            How Trump of you to pretend otherwise.


              …so yet again the number one Malcom meme of Accuse Others and inte11ectual dishonesty is on full display.

              It’s as if you cannot help yourself

      1. 7.3.1

        “policy considerations like preemption”
        preemption is policy…?

        Rules for eligibility that would result in the mass granting of patent claims to fundamental building blocks of knowledge/science (e.g., information, facts, correlations, logic, naturally-occurring products, etc.) exist, in part, because of underlying policy considerations. Some of those policy considerations also went into the writing of the Constitution.


          You mean Constituoonal policy considerations like separation of powers, allocation of particular statutory law writing authority, Void for Vagueness, need for present case or controversy (no advisory opinions), and the like?

          You are WAY behind on this topic.


            You mean Constituoonal policy considerations like separation of powers, allocation of particular statutory law writing authority, Void for Vagueness, need for present case or controversy (no advisory opinions), and the like?

            No, I mean the First Amendment and the patent clause itself. The First dominates the patent clause, which merely grants Congress a limited power to set up a rights-granting system to promote progress in the “useful arts” (arts that never included “writing instructions”, by the way). The First also prevents the government from directly or indirectly telling me what I can think or how I can think (or maybe you believe otherwise? — let everyone know).

            But, sure, let’s talk about “void for vagueness” because that’s s00per d00per important to you. Sure it is. We all believe that. “Void for vagueness” is a real serious genuine concern of the Be-Giled Richsters. Riiiiiiiight.


              Again with your vacuous reference to First Amendment…?

              Maybe you should actually elucidate that “First Amendment” position of yours with that “direct English language” that you always seem to want others to use.

              As it is, you crow “First Amendment,” as if the protection of utility is no different than the protection of expression (funny, how you never seem to address either the differences or similarities of the different intellectual property protections that come from the same Constitutional clause…

              As for Void for Vagueness, your expression of feelings really is not a discussion of that concept. All you do is attempt to disparage and you have not provided any cogent point on the matter. Your fixation with using Rich as a scape goat shows that you merely lack an historical understanding of how the current 35 USC 101 came about.

              Your disparage while lacking understanding is so very Trump of you.

  2. 6

    Patent eligibility inherently is a mixed fact/law question, like obviousness, where both eligibility and obviousness are legal questions that can only be answered by analyzing the underlying claimed technology. Close questions of patent eligibility require more factual investigation of the claimed technology, just like obviousness that requires a factual analysis of the prior art compared to the claimed technology based on the knowledge, skill, and experience of a person of ordinary skill in the art (POSITA). I’ve rarely met a judge or law clerk that is a bona fide POSITA that does not need any assistance deciding close factual questions of patent eligibilty and obviousness.

    1. 6.1

      PC: Patent eligibility inherently is a mixed fact/law question, like obviousness, where both eligibility and obviousness are legal questions that can only be answered by analyzing the underlying claimed technology.

      Certainly some truth to this but, as a matter fact, not every eligibility question needs an “analysis of the underlying claimed technology” because in some instances there is no “underlying technology.” Presumably these sorts of cases aren’t the “close cases” that you are referring to in your comment.

      My next point is not directed specifically at your comment but it’s an important point: treating all eligibility questions identically, or treating eligibility issues as if they are identical to each other is a very, very silly idea with bad consequences for everybody.

      For instance, cases where the recited statutory categories are plainly not met by a claim are very easy to deal with — so easy, in fact, that they are rarely present themselves.

      Cases where the recited statutory categories are met (e.g., claims that begin with the recitation of a “A device, wherein …” or “A process, wherein”) present “closer” questions but even those claims are often not “close” in the sense that a jury should ever get close to them. For example, a claim reciting (a) a prior art data gathering step, where the method is admittedly in the prior art followed by (b) a step of “determining” some “non-obvious” fact implicated by the data is not a “close” case. That’s a perfect case for tanking on the pleadings under 101 (and there are plenty of junk claims like that out there).

      Then there are the cases similar to that case where step (a) is not admitted to be in the prior art but nevertheless — beyond doubt — still is in the prior art as evidenced by the defendant’s unrebutted (or unrebuttable) presentation of evidence unequivocally showing that the case is in the prior art. Such cases may also be tankable on the pleadings. Certainly they should be tanked easily on summary judgment.

      The “closer” cases are where the data gathering step (a) is, reasonably arguably, non-obvious. Note that in these cases it is always a bit of an oddity that the patentee has even bothered to recite a step (b) because it adds no patentable weight to the claim and, if step (a) is non-obvious it should be (or should have been) claimed on its own terms. Of course, if step (a) is non-obvious then there is no harm in including a downstream “correlating” step, i.e., nobody freely practicing the prior art or an obvious variation of that art will be turned into an infringer merely for thinking about a correlation.

      The eligibility issues raised by certain “do it on a computer” claims parallel these issues with some distinct twists that arise from the manner in which “do it on a computer” claims are scrivened (very often a manner in which the nature of the claim elements is intentionally obscured or given a fake “structural” veneer that is easily wiped away by trained eyes).

      Unfortunately for Berkheimer, these claims are pure junk. It’s just logic being carried out a computer that was built to carry out logic. There are no “structures” or “objects” other than the virtual (abstract) ones that stand as proxies for ineligible data and/or ineligible logic. You can’t turn an ineligible subject matter (e.g., “parsing data”) into eligible subject matter merely by reciting the prior art (e.g., a “parser” <–LOL).


          You made your point well enough. I made a few distinct but equally important points in response.

          For the record, there’s no doubt that I’m very bullish on 101 (relatively speaking) for reasons that I’ve explained here hundreds of times. But I’ve consistently maintained the position that the best way to avoid eligibility questions outside of the domain of “new” information processing methods carried out by prior art computing devices (which will always be plagued with eligibility issues because … information processing) is to recite non-obvious physical structure in the claim, or to recite non-obvious transformations of matter.

          The only (somewhat) common outlier to this approach are claims that detect “non-obvious” natural phenomenon using prior art detection means. That’s the Sequenom eligibility problem, which is the same eligibility issue implicated by Myriad’s claims. Again, the larger point being: different flavors of eligibility, different analyses brought to bear. None of this especially difficult, by the way.

    2. 6.2

      The unpatentable subject matter tests required by the Sup. Ct. or any Fed. Cir. decisions thereon do not include any “close factual questions” or a 103 obviousness to a POSITA test. Even if step 2 of the test is required any asserted fact has to be something “well-understood, routine, and conventional.”


          Yes, but any assertions of a negative you don’t want to hear can be easily disproved by anyone citing any Fed. Cir. decision that uses a a 103 obviousness to a POSITA test in an Alice or Mayo 101 decision. Looking forward to seeing it.

      1. 6.2.2

        Paul, you are correct.

        The “new” requrement scrivened in by the Supreme Court is NOT a “103” obviousness item.

        It is an item that demands higher proof than that (from those that would deny patent protection).

        That proof is not presence, or even possibility of being known or used (were it known), but instead is proof that the item [including the particular arrangement of items vis a vis ordered combination] are widespread and in use.

        To prevail on that factual underpinning, “prior art” alone does not suffice.


          “anon”: The “new” requrement scrivened in by the Supreme Court is NOT a “103” obviousness item. It is an item that demands higher proof than that (from those that would deny patent protection).

          Please show everyone the Supreme Court language where they “demand” that “conventionality” of a prior art element be proven before a claim that recites “Prior art plus ineligible abstraction” is deemed ineligible.

          Let’s see it.

          And please recognize that even if you manage to find this citation, it will never hold up scrutiny because it turns 101 into a “dead letter” with respect to newly disclosed prior art. In other words, it creates the same absurdities and corrodes the system in the same way that the claims already addressed (and tanked under 101) by the Supreme Court did.


                Your request is not on point — that’s why you need to move the goal post back and read what I actually wrote.

                1. That “legend in your own mind” thing of yours is showing again.

                  So very Trump of you — another “win” in your own mind, untethered to reality.

      2. 6.2.3

        The Supreme Court has never held that in order to show ineligibility of, e.g., a claim in the form

        A method comprising (a) [data gathering step] then (b) determining “meaning” of data collected in step (a).

        that one must prove that step (a) was “well understood, routine and conventional” (a showing that goes far beyond mere obviousness). The CAFC has taken dicta from a case in which step (a) happened to be (admittedly) “routine” and has created an unconstitutional “test” out of that dicta that will never stand up to scrutiny. Why is that the case? Because the CAFC’s test would permit someone to, e.g., pull a newly disclosed (and therefore not “conventional”) display device from a patent application and obtain granted claims to the use of that device for displaying non-obvious (and ineligible) information. Such a claim, of course, would have the impermissible effect of protecting the information itself, at least in that prior art context.

        The CAFC’s test has zero chance of standing up to scrutiny for this reason alone. The Supreme’s invocation of “routine and conventional” was not a floor for a showing but rather something closer to the ceiling. It was invoked because the facts of the case before them were such that the only “statutorily eligible” subject matter in the claim was “routine and conventional.”

        Friendly reminder: programming programmable computers to process data has been a routine and conventional activity for a time period beginning before most patent attorneys were born. Nobody has ever believed or suggested that data processing computers are incapable of processing data that differs from prior art data by virtue of the “content” of the data. Keep that in mind.


          created an unconstitutional “test” out of that dicta

          LOL – who by chance did they learn that trick from?

          Hint: The Supreme Court and the repeated brow beatings.

          The CAFC’s test has zero chance of standing up to scrutiny…

          Just like the Supreme Court.

          Friendly reminder: programming programmable computers to process datausing protons, neutrons, and electrons to configure things has been a routine and conventional activity for a time period beginning before most patent attorneys were born since the first few moments of the universe.


            Last paragraph (strike through did not take):

            “Friendly reminder: using protons, neutrons, and electrons to configure things has been a routine and conventional activity for a time period since the first few moments of the universe.”

            Same “logic”

  3. 5

    HP is smart in raising this question, because the Supreme Court has conflated eligibility and obviousness and as a result has changed what should have been a question of law into a question of law that requires factual development before it can be decided. I hope they take it.

    1. 5.1

      the Supreme Court has conflated eligibility and obviousness

      No, they have not done that and, if you believe this, you have been played.

      The “conflation” story is g@ rb@ge propagated by a tiny group of whining attorneys who never understood subject matter eligibility and never will.

      The truth is that in a system where we claim inventions using words and where attorneys are permitted to include prior art elements in their claims without identifying those elements as prior art, in becomes necessary in most cases where eligibility is an issue to understand which elements in a claim are in the prior art and which are not. The reasons for this are crystal clear to anybody who has been paying even slight attention to the issue for the past ten years. I’m happy to provide very easy-to-understand examples if you are confused.

      But this necessary and fundamental analysis is not “conflation” of eligibility and obviousness. Obviousness is still obviousness; eligibility is still eligibility. The fact that the eligibility of a claim may hinge on the novelty or non-obviousness of a physical structure recited in that claim is not “conflation” of the two statutes. Again, if you can’t understand this, take a breath, think about it, and ask an intelligent question and someone will explain things to you at a level in which you will certainly understand, or you will revealed to be a w@ nker.

      Good luck.

      1. 5.1.1

        No, they have not done that and, if you believe this, you have been played.

        You are full of your own propaganda.

        Wake up son.


          Billy, your efforts to spread your silly propaganda were identified by me (and others) and called out as such years ago. It’s why you and your fellow patent maximalists are pretty much the laughing stocks of the patent bar. And it’s only going to get worse.

          The fact that most of you are wingnerts or glibertarian g00 fb@lls was also apparent a long time ago, as was the fact that you are total hypocrites and not very intelligent. I’m not even going to get started on all the R e p u k e sickness associated with your crowd except to say that Iancu is a lying fr@ ud just like the criminal who appointed him.


            Your Accuse Others meme has it exactly backwards.

            As is typical.

            (that laughter you hear is NOT laughter with you)

  4. 4

    Dennis, no question that this issue needs clarification, but does the fact that the subject decision was a remand trip up the “legs” here? That is, could the Supremes simply hold this case not yet ripe for cert because it is not final and may change depending on what happens in the District court on remand plus the likely second appeal therefrom to a different Fed. Cir. panel? Was that argued in any of these briefs?
    Also, there is the usual tendency of the Sup. Ct. to not take up cases from a recent decision [like Alice] until after the lower courts have had time to work out its application.

  5. 3


    But I do not think that to be the pertinent point.

    The pertinent point is whether or not the Supreme Court has the stomach to better tune its own scrivinings.

    What here in the questions would invite the Court to revisit its own Gordian Knot of undefined terms and tangled principles of eligibility and patentability?

    1. 3.1

      Even IF the Court has the stomach to lay out its scrivinings with more care, would they dare invite further scrutiny with what is arguably merely an invitation to engage in an advisory opinion lacking justiciability?

      To wit: “but instead directs the [ ] court to a non-decided question

      Implicitly (or otherwise) raising “interesting” questions that do not bear on the merits of the decision below (and thus, cannot affect the decision below) smack of the type of thing that WOULD invite more scrutiny for the Court.

      IF cert is granted on this non-decided question, would the decided question be impacted?

      That’s a double big “IF” coupled with a conjectural estimation, all in order to invite the Court into an area of law that true experts have viciously panned the decisions of the Court (even prompting the bar organizations to call for Congress to explicitly over rule the horrid Supreme Court decisions).

      As for the pedigree of the Court’s “Friends” who are writing briefs encouraging the further disenchantment from facts (who needs fact, right?), it’s a veritable laundry list of Efficient Infringer encouragers, including the thinly veiled “Big Bank” mouthpiece of Askeladden.

  6. 2

    Asking the wrong question is a feeble amateur trick by attorneys who know they’ve already lost.

    1. 2.2

      If the “question presented” to the Sup. Ct. in a cert petition is an issue never raised, briefed, or discussed below, and miss-represented as it if was, it ought to be sanctioned, not just rejected.

Comments are closed.