The Supreme Court has consistently cautioned that the judicial exceptions to patent eligibility need to be carefully applied:

At the same time, we tread carefully in construing this exclusionary principle lest it swallow all of patent law. Mayo, 566 U. S., at ___. At some level, "all inventions . . . embody, use, reflect, rest upon, or apply laws of nature, natural phenomena, or abstract ideas.”

Alice Corp. v. CLS Bank International, 573 U.S. __, 134 S. Ct. 2347, 2352 (2014) (citations omitted).

On November 14, 2014, exception did in fact swallow the rule in the Federal Circuit’s decision Ultramercial, Inc. v. Hulu, LLC, 2010-1544, 2014 U.S. App. LEXIS 21633 (Fed. Cir. Nov. 14, 2014) (“Ultramercial III”). Judge Lourie, writing for the panel of Judges O’Malley and Mayer, found Ultramercial’s patent claims ineligible. Along the way to that ending the court observed:

[W]e do not purport to state that all claims in all software-based patents will necessarily be directed to an abstract idea. Future cases may turn out differently.

2014 U.S. App. LEXIS, at *11.

This phrasing is precisely how one would state an exception to a rule. Instead of saying that software is generally not directed to abstract ideas, and that in some cases may be, this phrasing implies the opposite: software is presumed to be an abstract idea (hence ineligible) though there may be some case in the future in which it is not. This approach turns the entire patent eligibility framework on its head. This essay explores how Ultramercial III reaches this inversion and its potential consequences.

If at First You Don’t Succeed, Try, Try, and Try Again

Ultramercial’s patent claims have been before the Federal Circuit twice before. In Ultramercial, LLC v Hulu, LLC, 657 F.3d 1323 (Fed. Cir. 2011) (“Ultramercial I”), the Federal Circuit reversed the district court’s grant of a Rule 12(b)(6) motion to dismiss in an opinion by Judge Rader joined by Judges Lourie and O’Malley. The Supreme Court vacated that decision and remanded for reconsideration (i.e., invalidation) in view of its decision in Mayo Collaborative Services v. Prometheus Laboratories, Inc., 566 U.S. __, 132 S. Ct. 1289 (2012).

Apparently, the hint from the Supreme Court was not strong enough. In Ultramercial, LLC v. Hulu, LLC, 722 F.3d 1335 (Fed. Cir. 2013) (“Ultramercial II”), Judge Rader found the claim patent eligible. Judge Lourie wrote a separate concurrence explaining why the claim was eligible under “a two-step process, derived from Mayo” in the plurality opinion he authored a few months previously in CLS Bank v. Alice Corp, 717 F.3d 1269 (Fed Cir. 2013). After its opinion in Alice, the Supreme Court, which essentially adopted Lourie’s two-step test, again vacated the Federal Circuit’s decision and remanded for reconsideration in view of Alice.

Borrowing from military doctrine, the commander’s intent was clear: Ultramercial must go down. The tactical challenge was how to achieve this outcome. Normally, reversing itself would not have created a problem for the Federal Circuit panel, since a change in controlling law typically dictates a change in outcome. However, there was no change in controlling law in Alice. The Federal Circuit was tasked with using the very same theory that Judge Lourie used to find the claim patent eligible, on the same record, and to reach the opposite outcome. Something had to give: Lourie achieved this outcome by following the Alice/Mayo two-step test at a superficial level, divorced from its substantive grounding in preemption.

Further Down the Rabbit Hole

Ultramercial’s patent claim is a lengthy affair with eleven steps describing a method of providing free media content (e.g., movies, games) to a user after 1) the user has requested a specific ad and 2) has interacted with the ad. This is ostensibly different from existing methods in which 1) the user does not choose the ad to consume, and 2) the user simply watches an ad without interacting with it. However, instead of dealing with the actual claim limitations, the court first glosses the actual steps of the claim:

(1) receiving copyrighted media from a content provider; (2)selecting an ad after consulting an activity log to determine whether the ad has been played less than a certain number of times; (3) offering the media for sale on the Internet; (4) restricting public access to the media; (5) offering the media to the consumer in exchange for watching the selected ad; (6) receiving a request to view the ad from the consumer; (7) facilitating display of the ad; (8) allowing the consumer access to the media; (9) allowing the consumer access to the media if the ad is interactive; (10) updating the activity log; and (11) receiving payment from the sponsor of the ad.

2014 U.S. App. LEXIS 21633, at *10 (emphasis added).

Step (2) is emphasized here because the court’s treatment of this step plays an interesting role in the opinion.

Since the Supreme Court declined to offer a definition of an abstract idea in Alice, Lourie offers that an abstract idea is “an abstraction--an idea having no particular concrete or tangible form.”¹ Starting with the claim gloss then, the court very carefully selects which features to include in the abstract idea:

The process of receiving copyrighted media, selecting an ad, offering the media in exchange for watching the selected ad, displaying the ad, allowing the consumer access to the media, and receiving payment from the sponsor of the ad all describe an abstract idea, devoid of a concrete or tangible application.

2014 U.S. App. LEXIS 21633, at *10-11.

Accepting the court’s definition of an abstract idea as something devoid of concrete and tangible form or application, it is immediately apparent that even the court’s selective reading of the claim fails that test. If by “devoid” of application the court means that this characterization is inherently non-concrete or intangible, and incapable of being made concrete, that is clearly a false statement. Short of beaming ads and media into someone’s brain directly, I can’t think of any way in which a consumer can select an ad, the ad is presented, and the consumer then accesses media that is not ultimately “concrete” and “tangible”; whether an ad is displayed on a billboard or smartphone, whether the media is accessed on a laptop or a TV, it’s inherently a physical, tangible implementation that requires physical commitments. By contrast, the steps in Bilski v. Kappos, 561 U.S. 593 (2010) and Alice are first and foremost notional commitments: financial transactions to exchange currency or establish hedges are per se creatures of legal and regulatory regimes governing purely conceptual rights and obligations. If I break your nose in a fight, the primary act is that I’ve had a concrete and tangible impact on your face; it’s only secondary that I’ve broken a legal rule. But if I break a contract to exchange one set of rights for another, the only thing I’ve broken is an agreed upon legal fiction, nothing more. And so it is with displaying ads and accessing media: first and foremost, the ads are always presented physically, interacted with physically, and media is always accessed physically by the user. There’s nothing notional or conceptual about these acts.

If, on the other hand, the court is saying the gloss itself is lacking in concreteness, then that simply begs the question because it was the court that constructed the gloss to begin with. Notice that even though Ultramercial’s claim requires selecting an ad based on a specific and very concrete requirement, the number of times the ad has been presented (and as specifically claimed, that number is less than a contracted number of impressions already presented), this limitation has been left out, and the step is characterized as simply “selecting an ad.” As to the specific requirement, the court explains away this detail:

Although certain additional limitations, such as consulting an activity log, add a degree of particularity, the concept embodied by the majority of the limitations describes only the abstract idea of showing an advertisement before delivering free content.

2014 U.S. App. LEXIS 21633, at *11.

On what basis is the court permitted to decide what is the abstract idea in the claim based on the “majority” of the glossed-claim steps? That’s not done in Bilski, Mayo, or Alice. The court offers no principled way to decide which elements are included in the majority and which are not. This selective filtering is precisely what Judge Rader warned against: that a court “cannot go hunting for abstractions by ignoring the concrete, palpable, tangible limitations of the invention the patentee actually claims.” CLS, 717 F.3d at 1298.

Another consideration missing from the court’s analysis is showing that the abstract idea is in any sense fundamental. Unlike the economic concepts in Bilski and Alice—hedging and risk intermediation—providing online access to media after the user is presented with an interactive ad selected by the user is not a “fundamental building block” of science, technology, or modern commerce. Interactive ads on the Internet are no more than twenty years old; one cannot cite a hundred-year-old reference to show this is fundamental like the Court did in Alice.

It’s worth noting that the abstract idea identified by Lourie in Ultramercial III is different from that identified in Ultramercial II. There, Lourie states:

The abstract idea at the heart of the ’545 patent, which the district court properly identified, is “us[ing] advertising as an exchange or currency.”

722 F.3d at 1355.

Alice, coming after Ultramercial II, did not make any change in the definition of what constitutes an abstract idea, and yet somehow the abstract idea is now “showing an advertisement before delivering free content.” 2014 U.S. App. LEXIS 21633, at 11. These two characterizations are related, but they are definitely not the same. The court gives no explanation as to why there is a change in content of the abstract idea in the absence of a change in the law or facts of the case.

After recasting the abstract idea to which the claim is directed, the court turns to the second step of Mayo: whether the claim recites substantially more than the abstract idea. Here, the court quickly dispatches the remaining features of claim, including the specific requirement of step (2):

The majority of those steps comprise the abstract concept of offering media content in exchange for viewing an advertisement. Adding routine additional steps such as updating an activity log, requiring a request from the consumer to view the ad, restrictions on public access, and use of the Internet does not transform an otherwise abstract idea into patent-eligible subject matter. Instead, the claimed sequence of steps comprises only “conventional steps, specified at a high level of generality,” which is insufficient to supply an “inventive concept.” Id. at 2357 (quoting Mayo, 132 S. Ct. at 1294, 1297, 1300). Indeed, the steps of consulting and updating an activity log represent insignificant “data-gathering steps,” CyberSource Corp. v. Retail Decisions, Inc., 654 F.3d 1366, 1370 (Fed. Cir. 2011), and thus add nothing of practical significance to the underlying abstract idea.

2014 U.S. App. LEXIS 21633, at *13.

In CLS, Lourie stated that preemption was the Supreme Court’s primary concern: “the Supreme Court’s foundational § 101 jurisprudence offers the guideposts to such a system, one that turns primarily on the practical likelihood of a claim preempting a fundamental concept.” CLS, 717 F.3d at 1277. Relying on this framework in Ultramercial II, Lourie found that the “additional claim limitations reciting how that idea is implemented “narrow, confine, or otherwise tie down the claim so that, in practical terms, it does not cover the full abstract idea itself,”” and “the added limitations in these claims represent significantly more than the underlying abstract idea of using advertising as an exchange or currency and, as a consequence, do not preempt the use of that idea in all fields.” Ultramercial II, 722 F.3d at 1355. Now, in Ultramercial III, these very same limitations, including selecting ads based on prior number of impressions, have become “insignificant” data gathering.

Why Preemption Matters for Patent Eligibility

In Alice, the Supreme Court uses preemption to rein in the judicial exceptions to patent eligibility. Patent claims that “integrate the building blocks into something more … pose no comparable risk of pre-emption, and therefore remain eligible for the monopoly granted under our patent laws.” 134 S.Ct. at 2354. But in Ultramercial III, there is no mention whatsoever of preemption, and no explanation for why it no longer matters.

Removing the preemption consideration from patent eligibility is a dangerous maneuver, because what is “significant” enough to provide an inventive concept is intimately connected with the preemptive impact of a patent claim. As noted by the Supreme Court, all patent claims preempt to some degree. Accordingly, a limitation is significant if it prevents the claim from covering the “full scope of the abstract idea itself,” so that “in practical terms,” it does not “preempt the use of that idea in all fields.” Bilski, 593 U.S. at 604. Whether a particular limitation is significant “in practice” in a particular technology domain, and whether it prevents preemption, are ultimately questions of fact that can only be answered with real, objective evidence, not subjective judicial speculation. The patent law already uses the objective standard of POSITA, the person of ordinary skill in the art, when deciding other “questions of law,” such as obviousness, enablement, and claim construction. The failure to provide a similar objective grounding for patent eligibility is the fundamental problem in current § 101 jurisprudence.

The preemption analysis here is crucial: Ultramercial’s claim is before the court on a Rule 12(b)(6) motion, and thus, there is nothing before the court outside of the pleadings; how then does the court know which limitations are of “practical significance” in the field of interactive advertising and which are not if there is no evidence in the record? If the abstract idea in Ultramercial’s claim is “showing an advertisement before delivering free content,” as the court suggests, than the limitations of selecting ads based on the number of times they have been shown, requiring that the user request the advertisement, or making the user interact with the ad before receiving the free content all arguably narrow the claim so that in actual practice it does not cover the “full scope” of this idea. That is, there are numerous practical, non-infringing alternative ways of showing advertisements before delivering free content that are not preempted by Ultramercial’s claims; consider, for example, pop-up ads that commonly appear on videos. Consideration of the existence of practical non-infringing alternatives as evidence of patent eligibility was suggested by Judge Rader in Ultramercial II as “common sense,” since there were “myriad ways to accomplish that abstract concept that do not infringe these claims.” 722 F.3d at 1353. Expert testimony could be easily obtained to support such facts, but such testimony is generally not allowed on a Rule 12(b)(6) motion. And if such facts would be relevant and in dispute, then a Rule 12(b)(6) motion for ineligibility should be denied by a court that is rigorously relying on preemption.

By removing preemption from the analysis, the court can decide for itself what counts as significant, even though there are no facts in the record to support this conclusion. When preemption no longer matters, a court can reach any patent eligibility outcome it desires, even if there is no risk of preemption, let alone the disproportionate level of risk that is the Supreme Court’s concern. And indeed, this is precisely what is happening. For example, in McRO, Inc. v. Activision Publishing, Inc., 2014 U.S. Dist. LEXIS 135152 (C.D. Cal. Sept. 22, 2014), the court quoted from Alice that preemption is the guiding concern of Section 101, the defendants admitted that the patent claims did not cover the animation methods they used, and the court states that “It is hard to show that an abstract idea has been preempted if there are non-infringing ways to use it in the same field.” Id. at *23. Yet the court abandons these concerns with preemption and goes on to find the patent claims ineligible—an exercise in pure, subjective discretion.

The failure to consider objective evidence of the relevant technology domain also arises in the court’s use of the “data gathering” exclusion to wipe out a necessary feature of computers: All computer-implemented algorithms are combinations of “data gathering” steps, data processing steps, and “post-solution” steps. As correctly noted by Judge Pfaelzer recently in Cal. Inst. of Tech. v. Hughes Communs., Inc., 2014 U.S. Dist. LEXIS 156763 (C.D. Cal. Nov. 3, 2014), “The essence of software is manipulating existing data and generating additional data through algorithms. See Oplus Techs. Ltd. v. Sears Holding Corp., No. 2:12-cv-5707, 2013 WL 1003632, at *12 (C.D. Cal. Mar. 4, 2013) (“All software only ‘receives data,’ ‘applies algorithms,’ and ‘ends with decisions.’ That is the only thing software does. Software does nothing more.”). Judge Lourie’s logic allows a court to throw out whatever steps it cannot call abstract as being insignificant pre- or post-solution activity.

Judge Lourie also applies the machine or transformation test to Ultramercial’s claims, and in doing so he again implies a default rule that all software is patent ineligible. The court states that “any transformation from the use of computers or the transfer of content between computers is merely what computers do and does not change the analysis.” 2014 U.S. App. LEXIS 21633, at *16. Read literally, this asserts that any transformation of data by computers is “merely” what they do, and thus can never provide support for patent eligibility. This is completely inconsistent with what the Supreme Court suggested in Alice: that improvements in the “functioning of the computer itself” would evidence patent eligibility. Alice, 134 S.Ct. at 2359. Yet all such improvements must necessarily result from “merely what computers do.” This is because the court’s combination of any and merely turns the whole world inside out—Alan Turing, John von Neumann, Claude Shannon, and others merely defined the architecture of an amazing machine that merely allows for the flexible specification of any type of information, and from that, merely computes useful, valuable results in the form of new information, using a relatively simple set of operations merely defined in software. Complexity is achieved from simplicity by creating layers upon layers of software instructions, by defining high-order operations and data in terms of lower-order operations and data, all which is necessary to effect any improvement in the function of the computer, or any other computer-implemented invention. If Lourie is correct, then the Supreme Court’s apparent safe haven counts for nothing precisely because computers were merely built to operate this very way.

Ultramercial III is another indication that the Federal Circuit is firmly swinging against software and business method patents, without providing a coherent and consistent articulation of how to distinguish between abstract ideas and technological implementations. The Federal Circuit’s revised patent eligibility approach is boundless in the first step of the Mayo test in deciding what is abstract and unmoored from preemption in the second step. Anything can be described as an “abstract idea” when generalized sufficiently, and what’s left over can be thrown out as either routine, conventional, or pre- or post-solution activity, or “merely what computers do,” regardless of whether it actually preempts anything at all. This is a perfectly self-contained, almost logically irrefutable approach to invalidating any process claim, software or otherwise.

The Supreme Court has consistently warned that patent eligibility should not depend on the skill of the patent draftsperson. The great irony is that without the grounding principle of preemption, patent eligibility is now completely dependent on the drafting skill of the court in characterizing the invention as some abstract idea and selecting from the judicial toolbox which limitations will be insignificant. Instead, the courts must respect and give full weight to the Supreme Court’s treatment of preemption as the primary basis for the judicial exclusions to Section 101.

1. It’s important to note that Lourie conflates “abstraction” with “abstract ideas.” As I have explained, abstractions are different than abstract ideas. An abstraction is a generalization; it is a term or definition that identifies the principle aspects or features of the concept that are relevant to a given context, while removing features that are not important. All patent claims necessarily make use of abstractions. However, the limitation on §101 is focused on abstract intellectual ideas, not on abstractions generally: “Phenomena of nature, though just discovered, mental processes, abstract intellectual concepts are not patentable, as they are the basic tools of scientific and technological work.” Gottshalk v. Benson, 409 U.S. 63, 67 (1972). Over time, the use of the modifier “intellectual” has been dropped, and now the Supreme Court simply refers to “abstract ideas.”