Editorial Note: Each year we publish our annual IP Wishes column. I invite a panel of industry experts to tell us what they wish for in the new year. While I solicit wishes that are 1 to 3 paragraphs in length, I also let everyone know that if they care to write a full length article I am more than happy to publish their contributions. Marla Grossman took me up on that offer and drafted the following full length article. More wishes to follow…
If Gene (the “genie”) were to grant me patent and IP wishes for 2016, I would ask for (in no particular order) the passage of trade secrets legislation, resolution of the current patent reform legislation stalemate in Congress, that the USPTO consider evidence of non-preemption during its initial determination of patent eligibility; and that the USPTO prioritize accuracy, completeness and accessibility of the public record as part of its Patent Quality Review.
- Passage of Trade Secrets Legislation
One of my wishes for 2016 would be that Congress pass legislation to establish a federal civil cause of action for trade secret misappropriation to protect trade secrets from domestic and foreign theft. Hopefully such legislation would also provide for adequate safeguards to protect seized property.
On July 29, 2015, the House and Senate introduced promising, bi-partisan legislation entitled the Defend Trade Secrets Act of 2015. The legislation is based on the same standards for trade secrets protection, and remedies for misappropriation, that are found in the Uniform Trade Secrets Act and the Economic Espionage Act of 1996 (“EEA”).
American innovators would benefit greatly from a uniform federal cause of action that reflects the complex nature of trade secret misappropriation today. The current criminal statute on the books, the EEA, makes it a crime to misappropriate trade secrets for the benefit of a foreign government or for economic gain. To this day, however, there is no federal civil cause of action of general application for trade secret misappropriation. Thus, trade secrets claims must generally be litigated in state courts under state law. While most states and the District of Columbia have enacted some version of the Uniform Trade Secrets Act, there is variation among the state trade secrets acts as adopted.
Adoption of federal legislation like the Defend Trade Secrets Act of 2015 is important for two main reasons. First, despite the existence of the EEA, the Department of Justice has limited resources and will never be in a position to bring charges in all worthy cases of interstate trade secret theft. Second, the process for compensating the victim under the EEA is unwieldy, particularly when compared to relief available under civil law.
A federal trade secret cause of action could also be important for our global trade agenda. Unfortunately, many countries provide insufficient protections for trade secrets. This presents significant economic risks to American companies seeking to expand operations globally. Enhancing US trade secret protections would serve as a model for more effective deterrence of trade secret theft overseas, including in the European Union and in the ongoing Trans-Atlantic Trade and Investment Partnership (TTIP) negotiations.
There is real, bipartisan and bicameral interest in protecting American companies from trade secret theft as evidenced from the number of bipartisan co-sponsors of the Defend Trade Secrets Act in both the House and Senate. The current congressional interest makes 2016 the right time to pass strong and effective trade secret legislation, and I hope Congress makes this a reality.
- Resolution of the Patent Reform Legislation Stalemate
As we all know, patent reform legislation seems to be at an impasse. My second patent and IP wish for 2016 would be that Congress finds a path forward with this legislation, as there are important reforms that could benefit American innovators and the American economy.
Although the proposed patent reform bills could benefit from some changes before enactment, they do contain some important provisions that I hope are ultimately enacted into law. For instance, I believe that our patent system would be better off if it awarded attorney’s fees to the prevailing party unless the position and conduct of the non-prevailing party were objectively reasonable and substantially justified.
We’d also be better off with legislation that would make the high volume sending of bad faith demand letters, to end users who are not resellers, a deceptive act or practice within the meaning of the Federal Trade Commission Act, provided that such legislation is carefully structured to differentiate between abusive activity and legitimate, lawful activity. Any legislation should include a clear identification of objective acts or practices that would deceive recipients, such as false statements of patent ownership or the right to enforce or license patents. Such legislation should not be overly burdensome and require detailed disclosures, which could have the unintended result of increasing patent infringement litigation. Finally, this legislation should preempt state legislation.
This country could also benefit from patent reform legislation to limit venue in patent suits in order to curb forum shopping by patent owners. Hopefully, in a patent infringement action brought by a patent owner against an accused infringing defendant, Congress will limit venue to a judicial district in limited circumstances such as:
- where the defendant has its principal place of business or is incorporated, or if the defendant is an individual, where the defendant resides;
- where the defendant has committed acts of infringement and has a regular and established physical facility where defendant’s acts of infringement have occurred;
- where technological activities relating to the invention that led to the application for the patent(s) have taken place;
- where a plaintiff or a subsidiary has a regular place of business, not operated primarily for the purpose of creating venue; or
- where the defendant has agreed or consented to be sued.
- USPTO Considers Evidence of Non-Preemption During Its Initial Determination of Patent Eligibility
My third wish would be that the USPTO consider evidence of non-preemption during its initial determination of patent eligibility. The USPTO requested comments on its examination guidance on patent subject matter eligibility on July 30, 2015. Unfortunately, the July 2015 examination guidance devalued the role of non-preemption evidence, stating that “the courts do not use preemption as a standalone test for eligibility. Instead, questions of preemption are inherent in the two-part framework from Alice Corp. and Mayo…and are resolved by using this framework to distinguish between preemptive claims.” The July 2015 examination guidance stated that “while a preemptive claim may be ineligible, the absence of complete preemption does not guarantee that a claim is eligible.”
This analysis is not in line with the express holdings of the Supreme Court and the actual practice of the district courts applying the Mayo test. Evaluating preemption is fundamental to distinguish between claims that preempt fundamental building blocks and those that “pose no comparable risk of preemption, and therefore remain eligible for the monopoly granted under our patent laws.” Without consideration of the preemption one cannot determine whether the risk of preemption is disproportionate; the mere possibility of preemption is not sufficient.
- The USPTO Prioritizes Accuracy, Completeness and Accessibility of the Public Record as part of its Patent Quality Review
Accuracy, completeness and accessibility of the public record, and consistency in applicant experience across patent examiners and art units are important dimensions of patent quality that I hope USPTO will not overlook in 2016 as the agency defines and pursues its quality objectives.
In introducing its new quality initiative, the USPTO has quite correctly said that “[e]ffectively promoting…innovation requires that issued patents fully comply with all statutory requirements and, of equal importance, that the patent examination process advance quickly, transparently and accurately.” The agency’s goal is “[h]igh quality patents” that “permit certainty and clarity of rights, which in turn fuels innovation and reduces needless litigation.” Such goals are laudable. However, the issue of patent quality has many dimensions and those dimensions have varying degrees of importance depending on one’s perspective. The most obvious dimension of quality relates specifically to the accuracy or correctness of patent examiners’ decisions, but this is by no means the only dimension of quality. For example, consistency of the examination process across examiners in an art unit, and across art units, is an important measure of quality to applicants and patent attorneys. And accuracy and comprehensiveness of published patent content is highly important to users of public patent information and is very relevant to the USPTO’s goals concerning excellence in work products and excellence in measuring patent quality.
More specifically, there are several aspects of quality that can easily be overlooked when considering excellence in work products and excellence in measuring patent quality but are of utmost importance to the long-run integrity and quality of the US patent system:
- The Quality of Inputs to Examination – The quality, suitability and accuracy of the information provided by applicants to the USPTO have direct bearing on an examiner’s activity and consequently the outcome of the examination process. Analysis of applicant-submitted documents to the USPTO shows that applicants make literally millions of procedural, categorization, and other errors in their submissions to the USPTO each year. Currently the USPTO has contractor-operated processes in place that filter and correct documentation submitted by applicants. This includes, for example, correcting applicant-supplied doccodes, which in turn enables examiners to efficiently locate documents contained in the file wrapper. This filtering is critical to effective and efficient work on the part of patent examiners.
In addition to maintaining existing quality checks on applicant inputs, filtering for applicant-generated errors can and should be expanded and enhanced by using advanced text analytics to identify structural flaws and other document shortcomings. A wide range of anomalies and error conditions can be detected through automation, and the USPTO should be applying such technologies before the start of examination, thus creating opportunity to improve efficiency during examination. Relatively straightforward examples of common detectable errors include claim numbering errors (particularly when claims are revised by inserting or deleting whole claims) and lack of antecedent basis. Identifying such problems up front could provide significant benefits to patent examiners and improve the efficiency and quality of examination.
- The Quality of the Examination Process in Relation to Structural/Procedural Accuracy – The examination process, which commonly takes place over the course of multiple months (sometimes years), may involve replacement of the application specifications or entire claim sets. The risk of structural errors becomes much greater with each submission. Review of allowed patent applications shows that well over 100,000 flaws relating to policy or statutory requirements and/or agency business rules persist up to the point of allowance each year. These are identified post-allowance today by independent review in the publication process – a key pre-issuance quality filter. It is critical to the quality of published USPTO content that this quality check and correction step be retained. If it is not, the clarity of the record will be diminished and users of public patent information will be negatively impacted. Additionally, technology exists to apply text analytics upon introduction of new content, as well as at the point of allowance, that could help to avoid the persistence of an additional range of structural errors in allowed patents. The USPTO should keep existing post-allowance filters in place, but in addition to those filters, it should also implement text analytics that are already available in the private sector to catch errors and anomalies in the examination cycle with a view to further reducing the persistence of errors to the point of allowance or beyond.
- Consistency of Performance across Examiners and Art Units – Currently a very wide range of examiner performance exists both within individual art units and across art units. This variation in performance is often overlooked as a quality issue, but it can be a source of significant frustration for patent applicants and a cause (as well as a symptom) of inefficiency in the patent prosecution/examination cycle. Automated tools exist that quantify such variation, and quantification is a necessary component of an overall improvement plan. For example, technology exists in the private sector that can predict with a high degree of accuracy how long it will take for an RCE to be considered on an Art-Unit-by-Art-Unit basis, or what the likelihood of allowance is on an examiner-by-examiner basis. Such tools are already beginning to change applicant behavior. The USPTO should consider implementing such tools within the agency in an effort to reduce unwarranted and undesirable variation and improve the quality of applicant experience as well as the efficiency of the examination process.
- Accuracy of Searchable Published Patent Content – Accuracy of the text and other searchable components of published patent documents is absolutely fundamental to the clarity and accessibility of the public record. This is especially true in a data set as large as the entirety of published U.S. patent content. It is easy to take this for granted, but it would be a major mistake to do so. An extraordinarily large number of documents are processed by the USPTO and USPTO contractors today using Optical Character Recognition (OCR) technologies. OCR’d text is known to contain a significant number of errors. Human review overcomes this problem by correcting an estimated 200,000,000 text errors each year in OCR’d patent documents prior to publication. These errors are corrected by highly skilled human proofreaders rather than relying exclusively on automated correction processes. Such automated processes alone would not produce text that is accurate enough for patent research. Algorithms and spelling dictionaries used by automated processes are notoriously lagging behind the technology fields represented by patents. These automated processes alone would inevitably introduce deviations from the patent source documents. Deviations cannot be allowed because the application submitted must be faithfully recorded. Failure to correct OCR errors would have immense negative consequences for all stakeholders who rely on the accuracy of this data. The USPTO’s Patent Application Text Initiative (PATI) – which relies on OCR technology alone – produces outputs that are sufficient for some purposes, but not for publication, and not for the detailed analysis applied to such data by commercial entities and researchers. Thus, the USPTO should maintain its current contractor-operated text correction processes and patent text quality checks to ensure the continuing high quality of published patent content at 99.995% accuracy rates or better.
- Richness of Metadata / Depth of Content Tagging – U.S. patent content today has a very rich layer of metadata that facilitates utilization of the content for a wide range of purposes across the IP stakeholder community. The richness of this metadata is itself a significant measure of quality. The USPTO should not consider reducing the richness of the metadata produced today as a tradeoff to achieve higher degrees of automation and lower data capture / content processing costs. Nor should it do so because of a desire to shift the responsibility for tagging to applicants, knowing that applicants will often get the tags wrong if the requirements are too complex. Reducing the richness of the tagging for any such reason would be a significant mistake. Users of patent information in the US and around the world rely on the rich metadata produced today by the USPTO and consider it the gold standard for published patent information globally. The USPTO should not consider any simplification of current metadata / tagging requirements. On the contrary, enriching the metadata further – for example, to allow linking of part numbers to drawings or to cross reference different chemical names for identical chemical entities – would increase the value and perceived quality of U.S. patent content, and would foster even greater innovation.
- Consistency of Published Content – Patent applications are filed in the U.S. by a wide and diverse range of over 30,000 patent agents / attorneys. In combination with the novel nature of the content contained in patent applications, this fact would yield significant variation in the way similar material is presented to the public (e.g. in the way in which tables or chemical structures are published) without a processing step to introduce consistency across all applications. The agency currently utilizes a contractor to ensure such consistency. Failure to continue this practice would likely result in a reduction in perceived quality of published content across the stakeholder community, and would certainly result in an inferior product compared to today’s published patent products. Therefore the USPTO should retain this processing step prior to publication.
Accuracy, completeness and accessibility of the public record, and consistency in applicant experience across patent examiners and art units are important dimensions of quality that must not be overlooked as the USPTO defines and pursues its quality objectives. It would be a significant mistake to take these dimensions for granted; they are essential to an efficient patent and trademark system that promotes US innovation, economic growth and the public good. The agency should explore each proposed change thoroughly to ensure that any changes to current processes will indeed enhance the quality (and the perception of the quality by stakeholders) of USPTO processes and published content, and not diminish it.
In summary, I hope the USPTO will indeed take these points into consideration as it makes decisions about how to make the best patent system in the world even better. May 2016 bring health, happiness and innovation to all!