On Wednesday afternoon, I attended the USPTO’s public forum about the latest Interim Guidance on 101 Subject-Matter Eligibility. This session had the distinction of providing a genuine, open dialogue between the patent bar and the USPTO. Practitioners expressed their praise, criticism, and concerns for the guidance without much padding. USPTO representatives spoke with exceptional candor about the background of the latest draft of the guidance, and their constraints and hopes for future drafts.
On the whole, the session was richly informative about the background of the Interim Guidance; the different perspectives of the USPTO and the patent bar; and the issues that future versions of the guidance may address.
The public forum began with an introductory review of the Interim Guidance from a panel of USPTO officials – including Drew Hirshfeld, Deputy Commissioner of Patent Examination Policy (and author of the Interim Guidance). The session then featured brief remarks about the guidance from a variety of presenters: law firm partners; executives of IP consulting companies and practice groups, including AIPLA, BIO, and the IP section of the American Bar Asosciation; and in-house counsel for leading biotech and computing technology organizations – quite a collection of deeply knowledgeable practitioners.
Much of the public feedback discussed the practical effects of the Alice case and the Interim Guidance. Recurring themes in these comments included:
- Broad concern about the uncertainty created by new questions of patent validity, and the variance and unpredictability of prosecution.
- The effects of the sweeping changes of patent law on the rate of innovation and the health of the economy. (One presenter related a story from the 1970’s, where a small company asserted a single patent to protect its market lead in the field of check printing from a large, well-funded competitor – and noted that the small company grew into the 8,500-employee, $4.5-billion-annual-revenue Intuit.)
- The “deharmonization” of international patent protection, due to growing discrepancy between patent practice in the U.S. and elsewhere.
- Metrics of the diminished allowance rates, particularly in business art units. (One presenter offered a chart of rejection rates, showing a massive spike that reached nearly 100% in business-method-oriented art units.)
- The stark contrast between the restraints placed on patent law by Alice and the historic basis of patenting: i.e., earlier versions of the U.S. patent system and its predecessors widely recognized the patent-eligibility of all sorts of “abstract” ideas.
- A broad feeling that the USPTO and courts have over-applied the intent of Alice, and that heightened patent-eligibility requirements are “swallowing all of patent law,” despite the Alice Court’s cautionary note. Hope for moderation and balance (“the pendulum swinging back,” etc.) was frequently expressed.
While valid and urgent, these comments seemed misdirected to this forum. The USPTO has no power over patent law – it is strictly an administrative agency. Its responsibility is to take all of patent law (35 USC, 37 CFR, the policy statements of the executive and legislative branches, and the massive body of case law passed by the judiciary), and to set forth rules for examiners that express those sources as clearly and coherently as possible. Policy-based arguments, including the effects of the changes and exhortations for adjustments, are to be directed to Congress and the judiciary. 1
Other comments, from both attendees and USPTO officials, were directed at the content of the guidance – and many of these remarks were both new and valuable. Some of the more prescient remarks include:
- Raul Tamayo, of the Office of Patent Legal Administration, acknowledged – and apologized for! – the skewed focus of the Interim Guidance toward life sciences technologies and away from computing technologies. Future versions of the guidance are likely to flesh out the details of 101 practice for computer-related technologies.
- Several commenters noted the broad trend in recent case law to create a single, comprehensive test for 101 patent-eligibility that is applicable to all technologies. These commenters expressed concern that despite this trend, the USPTO’s policies are creating divergent patent-eligibility rules for life sciences and computing technologies. One commenter noted that the guidance explicitly promotes this divergence with distinct examples and clarification that are specific to “nature-based products.”
- Several commenters noted the abundance of examples in the guidance of patent-ineligible claims, and the comparative lack of examples of patent-eligible claims.
In response to these comments, the USPTO officials generally noted that their mission in preparing the guidance was to synthesize the case law as faithfully as possible. For example, they would liked to have included more examples of patent-eligible claims, but the general absence of available court decisions is prohibitive.
However, the most provocative and well-targeted comment of the forum addressed this argument by noting the degree of latitude in the USPTO’s synthesis of case law into the Interim Guidance. The commenter noted that some cases are interpreted as broadly stating a general rule, but other cases are interpreted narrowly as only applicable to the case at bar.
It is appreciated that any attempt to assimilate the compendious jumble of contradictory case law into a coherent statement of 101 patent-eligibility must require a heavy degree of artfulness and tact. However, the USPTO is reluctant to acknowledge that it has used any interpretive latitude in this task.
While I do not believe that the USPTO is taking liberties with its interpretation, I do feel that the USPTO may not have chosen the best solution. The guidance attempts to gloss over the areas of 101 case law that are most vague, contradictory, and problematic – like forcing together the poorly-fitting pieces of an unworkable puzzle, and minimizing the gaps in between. Instead, the Interim Guidance could acknowledge the specific questions of case law where a clear answer is currently unavailable, and where clarification is needed – thus prompting the CAFC and SCOTUS to address them. This aspect is completely consistent with the “Interim” status of the guidance. 2
Other comments involved how examiners are applying the guidance:
- The most common request, by far, involved requiring examiners to set forth their reasoning in Alice rejections with more specificity. Many commenters noted that many such rejections are conclusory, and impossible to rebut. The USPTO officials noted the abundance of this request throughout all public comments on this topic: “When even the competitors in a field are submitting the same request, it’s clear that we need to act.”
- One well-received comment urged coordination among technology center directors and SPEs, in order to provide a more consistent experience across art units and technology areas.
- Some discussion involved whether the examiner’s prima facie case for an Alice rejection required evidence – i.e., citation to the specification or a reference – and whether it could be rebutted with evidence, such as an affidavit by an inventor or expert. Not much resolution here, but the questions are interesting.
For my part, I offered three specific observations and recommendations – largely because I haven’t read or heard them expressed to date:
- Recommendation #1: Boilerplate language for 101 rejections.
For years, the USPTO’s office actions have adhered to a particular format for standard rejections: consistent boilerplate is used to assert an anticipation rejection under 102, an obviousness rejection under 103, an ambiguity rejection under 112p2, etc. The boilerplate serves the important function of specifying the legal requirements that the examiner is invoking – i.e., how the examiner’s interpretation of the claims and references are mounted onto the framework of patent law.
However, for 101 rejections that assert patent-ineligibility, every rejection looks completely different. Some are brief; some are extensive. Some recite explicit case law; others cite the guidance (or earlier USPTO memoranda); some don’t cite anything at all. As the legal test of 101 patent-eligibility, such rejections may paraphrase Alice and Mayo (“fundamental economic practices” / “methods of organizing human behavior” / “lacking ‘significantly more’ than an abstract idea” / claim elements identified as “conventional and routine”); the Bilski CAFC opinion (neither “tied to a specific computer,” nor “transforming an article into something different”); even remnants of Freeman-Walter-Abele and the “printed matter exception.” Some rejections are so paraphrased and inaccurately summarized that they seem like brand-new variations on 101 patent-eligibility.
Given the extreme variance is rationale for rejecting claims under Alice, it is frequently difficult to determine the examiner’s precise complaint about the claims. For practitioners, this is a major obstacle to preparing a response – and even deciding how to respond. Is the examiner using the right standard of law? How is the examiner applying the flowchart from the Interim Guidance to the review of the claims? Is the examiner even rejecting the claims as written, or is the rejection asserting patent-ineligibility of the invention at large? Indeed, if the rejection only asserts that the claims recite “patent-ineligible subject matter,” is this even a rejection under Alice or just under In re Nuijten?
My recommendation, for a future version of the guidance, is to provide boilerplate that examiners are directed to use for Alice rejections. I noted that different types of boilerplate could be provided: one for “the field of the invention is patent-ineligible”; one for “the claims are overbroad and seek preemption of an ‘abstract idea,’ and therefore need ‘significantly more’ to garner eligibility”; one for “the claim lacks utility, as per Mayo“; etc.
- Recommendation #2: Resolution of Nuijten 101 issues.
It’s interesting that the Interim Guidance provides a flowchart, where step 2 (providing the complex framework of Alice and Mayo) is preceded by step 1, with the simple threshold test: “Do the claims recite a proper statutory class?”
Of course, this threshold question invokes In re Nuijten (2006) – and it does so because today, in 2015, examiners and practitioners continue to grapple with statutory class issues, with continuing disputes over questions like:
- Is a “system” a proper synonym for “apparatus?”
- Are terms like “computer” and “processor” limited to a physical device, or does it include the human brain? Do we need to specifically recite: “a hardware processor?”
- Does the terms “memory” and/or “memory device” include the example of propagating electromagnetic signals? Do we have to distinguish signals with an explicit disavowal (“a memory not comprising an electromagnetic signal“), or with specific descriptors (“a nonvolatile memory” / “a physical memory” / “a hardware memory” / a “nontransitory memory”)?
- Is it sufficient to distinguish the claims with a legal disclaimer (“a computer-readable medium only including variants that satisfy 35 U.S.C. 101, comprising instructions…”)?
- Do the instructions embedded on the medium even provide meaningful limitations, or can they be rejected under the printed matter exception?
These issues continue to arise daily. Every examiner and art unit handles them differently, and the standards continue shifting. No one is happy with this situation. Nine years after Nuijten, this topic is majorly unresolved.
However, in stark contrast with Alice-type issues, recent case law over the sufficiency of statutory classes is practically nil. The only recent remarks on this topic are casual and tangential: if an invention is held patent-ineligible under Alice, the courts tack on some remarks about why the statutory class of the claims doesn’t save it from ineligibility.
I believe that the Interim Guidance is an excellent opportunity for the USPTO to formalize its policy on Nuijten and statutory classes. The guidance could provide clear examples of eligible and ineligible statutory classes – and an instruction that, absent clear language in the specification that provides otherwise, these definitions are controlling for 101 Nuijten issues. Indeed, this is a fair tradeoff: providing an expedient resolution of 101 Nuijten issues enables examiners and applicants to focus attention on hashing out the 101 Alice issues.
- Is a “system” a proper synonym for “apparatus?”
- Recommendation #3: Explicitly state that a technical solution to a technical problem requires evaluation on its own merit, regardless of the broader field of the invention.
Consider the following (completely hypothetical) invention:
Within the field of language translation, computers are often configured to translate words between language. To do so efficiently, computers often use an index that enables a rapid lookup of an entry for a word in a first language, where the entry specifies the equivalent word in a second language.
In some cases, an expression of a multilingual individual may include words from multiple languages. In order to translate these words, it may be helpful to generate an index containing the words of multiple languages. Where the languages are represented by different symbols, such as the English alphabet, the French alphabet, and Cantonese pictograms, the indexing mechanism may utilize a multi-language encoding format, such as Unicode; however, indexing may fail if the words of each language are represented in a discrete alphabet. For example, the French word “café” may be equivalent to the anglicized version “cafe,” yet a primitive indexing technique will feature two unrelated index entries encoded in different Unicode symbol sets, which may give rise to divergent results based on the selection of different symbol sets.
The present invention is an indexing algorithm that enables the indexing of words in different languages, but where equivalent symbols in different languages are consolidated to a common letter. The algorithm is: (reciting an algorithm that achieves this result). Configuring a computer to generate an index through this algorithm, and to use such an index during language translation, may improve the efficiency and accuracy of the machine translation.
This type of invention is squarely within the realm of patent-eligible subject matter: it raises a specific technical problem, and provides a specific technical solution that directly improves the performance of the device. Nevertheless, in many cases, these types of applications are being rejected as follows:
This invention is directed to the field of language translation, which dates back to the dawn of spoken languages. Additionally, the invention is directed to indexing techniques, which are conventional and well-studied in the field of computing. Irrespective of its details, an invention in these fields necessarily involves an abstract idea that is not patent-eligible under 35 U.S.C. 101 (see Alice Corp. v. CLS Bank).
The hypothetical examiner in this case didn’t even make it through the summary of the invention to the actual description of the problem. The examiner just saw “language translation” and “indexing techniques,” and began writing up a 101 rejection. The problem is that the examiner hasn’t appreciated that the application recites a specific technical problem arising within those fields, and a specific technical solution that is a patent-eligible invention. That is, the “significantly more” step of the guidance requires that the examiner consider the invention, even if the field of use of the invention seems abstract.
At present, this examination scenario is arising frequently, and I would like to see the Interim Guidance provide an example along these lines in the “significantly more” section. DDR Holdings, Inc. provides a solid example: within the fields of e-commerce and marketing, a specific problem is recognized (featuring a third-party product in a way that does not clash with the vendor’s web page), and a specific technical solution is provided (insert elements into the web page that present the third-party product ad with the “look and feel” of the vendor’s page).
- Of course, the USPTO Director plays a vital role in the policy process: the director can convey deep institutional knowledge and recommendations to policymakers and judges. But neither Acting Director Lee nor any representative from her office was in attendance. And it must be noted that the last attempt by a USPTO Director to take an active role in setting patent policy was not well-received. ↩
- The Supreme Court should also be applying this principle, by holding 35 USC 101 hopelessly void for vagueness under the basic principles of statutory construction. Instead, the Court has attempted, at length, to fix the severe problems with 35 USC 101 – thus causing 30+ years of chaos over one of the most basic issues of patent law.
The problems that the Court is trying to address with its 101 tinkering are, perhaps, not best resolved by the efforts of nine unelected justices who have, altogether, no technical credentials; no experience in patent practice; and no claim to the business of technology. I reserve hope that the Court will someday reach this conclusion, and prompt Congress to resolve these critical issues through the ordinary legislative process. ↩