The Enhanced Patent Quality Initiative and Search: A Brief Time Line and Options Moving Forward
INTRODUCTION
Patent quality has been an issue for many years. It used to confound me that a patent could be cancelled in a reexamination at the PTO but upheld and deemed infringed by the courts. Famous examples of low quality patents include the sideways swing and the crust-less peanut butter and jelly sandwich.
Addressing the low quality problem Congress passed the America Invents Act. This law authorizes three new proceedings to contest issued patent before a Patent Office Board, namely:
1) Covered Business Methods,
2) Post Grant Review, and
3) Inter-Parties Review.
IPR’s are particularly interesting because they provide a window into the prior art process. IPR’s can only be brought based on grounds of new prior art; printed publications, section 102 or 103 arguments.
On February 5th 2015, the Office, through a Federal Register Notice, solicited comments for a new Enhanced Patent Quality Initiative. The EPQI is a multi-pronged effort to address a wide range of quality issues.
What I’m going to talk about today are
1) IPR’s,
2) the EPQI and search,
3) current rules/regulations that address searching,
4) comparative search models and finally
5) search reform options.
A PRESENT-DAY IPR SNAPSHOT
Inter-Parties Reviews went live on Sept 16, 2011. In the 5 ½ years since as of March 31, 2107:
6139 IPR petition have been received.
4563 of these have been completed.
2406 (52.74%) trials were instituted and of these
1277 (28%) had an adverse decision against the patent holder.
1277 represents the number of patents where at least one independent claim was cancelled. That is not a very big number but it is enough to create chaos and has affected public perceptions of the patent system.
Critics argue that the patent system has been destroyed because IPR’s are the taking of a private property right. The Presumption of Validity is established by law but low quality patents in combination with IPRs render the presumption meaningless. IPR’s and low quality patents also foster will infringement. Why invest the effort to obtain a patent, say if you are a start-up, in view of the strong possibility that the patent will be rescinded when it is infringed.
The first IPR 2012/00001 (Cuozzo 6778074) was decided on November 13, 2013. The Cuozzo patent is a great example of how the prior art process works. The patent was for a speedometer that sounded an alarm when the vehicle was speeding by accessing a GPS database. The applicant submitted seven references with the application. The references were mostly off topic. The examiner on the other hand found four references, applied two of these to reject the application, the rejection was overcome and the patent issued.
Addressing prior art, (AD LIB) there is no way to determine the amount of time it took for the applicants representative to find seven references that seemed to all have in common the words speed and GPS in them but….. As you would expect a professional searcher to do, the examiner found a patent with GPS and speedometer in the title quickly. It is clear from the file history; the examiner spent 12 minutes conducting the search. I am not attacking an examiner just stating facts. In fact, examination in this case worked just as it should. This application fits squarely within PTO quality metrics because there was negotiation between the examiner and the applicant before the patent issued. However, on a closer look, it’s pretty obvious no focused search done by either party. This patent also became a Supreme Court case but that’s another story.
The IPR was filed with 5 new references all US patents. A trained eye can interpret this. Two of the references were pre-1970 suggesting a manual classification search was done. My point here is simple; prior art is found when someone intentionally looked for it.
I’ve personally only looked at four cases and deficient searching was evident in all four. Patent 8504637, I think reflects the problem with rule 56 but I don’t know, I’m reading between the lines. This application was field with several IDS’s totaling 1300 references. The IPR was filed with two references. Both of those references were among the 1300 provided by the applicant. This tells to me the applicant didn’t even search their how art collection.
I think you will find a lack of searching in all successful IPR’s.
The peanut and jelly sandwich, patent 6004596, since I mentioned it above was pre-IPR and over tuned by both the court and the PTO but only after Smuckers had invested $20 million to build a factory in KY.
Low quality patents have very costly consequences.
THE EPQI AND SEARCH
The 2/5/15 Federal Register Notice announcing the EPQI served to initiate the public discussion on all things related to patent quality. The Office proposed two ideas, 1) an Automated Pre-Examination and 2) Crowd Searching. The Notice also announced an upcoming Patent Quality Summit.
The Patent Quality Summit was held over two days, March 25-26 2015. This was basically outreach where the Office covered the information from the earlier Notice and explained what they would do moving forward. The afternoon on the first day was devoted to small group discussions where the subject matter discussion more or less emerged organically. A representative from each table shared what was discussed with the audience. Every table had something to say about prior art.
One year later, on April 27, 2016, the Office held a Patent Quality Community Symposium at their Headquarters in Alexandria. This was a significant date because the Office provided a detailed presentation on Search and Training Enhancements. The audience was updated on the Automated Search but also three new initiatives were discussed.
- The Collaborative Search Pilot is a filing option where an applicant files simultaneously in the US and either in the JPO or KIPO. The search results and patentability opinions are shared between the Offices before a first Office Action is sent.
- The Global Dossier is a tool where the prior art from related applications filed at participating offices is collected and centralized for access by anyone.
- The Office is encouraging examiners through an “awareness” initiative, to use the services of Scientific and Technical Information Center (STIC) library. A contractor works in-house tasked with doing NPL and additional searching when requested by an examiner. They are considered Search Strategy Experts.
Two subsequent Conferences on Patent Quality were held in 2016. Both of these were in partnership with Duke and Santa Clara Law Schools.
The first Conference was held in Santa Clara on Sept. 13 2016. The second Conference was held at the USPTO on Dec. 13th.
From the Conferences we learned that Crowd Sourcing was no longer a viable option because there was too much similarity between the PTO actively encouraging the public to submit prior art in pending application and Opposition like proceedings. At the Dec. 13th meeting, Gregory Vidovich, the Associate Commissioner of Patent Quality updated the audience on the progress of the Automated Tool. He also discussed the aforementioned STIC Library.
The Conferences were particularly significant because of the panel discussions. Mr. Alfred Spigarelli, Director of Patent Quality at the EPO spoke at both of them. His told the audience that the EPO provides a high quality search product because they made a conscious decision to emphasize that a thorough search is the “cornerstone” to a patent quality. Getting a thorough search includes not putting a time constraint on the searcher. He also said that their approach engendered pride in a quality search product.
Summarizing, the EPQI is two years old and has proposed four ideas:
1. The Automated Pre-Examination Search;
2. The Collaborative Search Pilot;
3. The Global Dossier
4. STIC Awareness
I don’t see how any of these get to the core problem of inadequate and unfocused searching as exposed by IPRs.
CURRENT RULES AND/OR REGULATIONS
There is not a policy, a regulation or a rule directed to patent searching. The absence of specificity, by the patent authorities, for searching or for prior art is a driver behind low quality patents.
The term “prior art” appears in 37 CFR 1.56 the Duty to Disclose Information Material to Patentability. This rule was first implemented in 1977 to address fraud upon the Office. The rule does not require a search so the subject of searching is avoided. Hiding prior art in order to obtain a patent is not a problem today especially considering the lack of a presumption of validity. Attaching legal penalties to a lawyer for someone else’s work product results in applications untethered to prior art and are themselves a factor in low quality patents. Not focusing on finding prior art has the same results as fraud: an invalid patent.
M.P.E.P. Chapter 900, Section 904 is directed to searching. This section accurately describes the search process procedurally. This tells us what examiners are supposed to do. However since there are no quality metrics tied to this, there is no way the Office can confirm this is what they actually do. The M.P.E.P. is not reflective of a policy nor is it a rule.
37 C.F.R. § 10.7 Requirements for Registration is the regulation that designates who are a patent professionals. This regulation does not include patent information professionals. A patent professional is someone who writes patent applications. So, consequently, search professionals are not integrally part of the patent system in the eyes of the US patent office. This regulation was first promulgated in 1952 in order to implement of the Patent Act of the same year. A key concept expressed in the rule is the idea of “valuable service.” I have said elsewhere that because “valuable service” is restricted to “applicants for patents” searching is left out. This rule reflects and defines an internal PTO mindset where searching is the responsibility of the individual examiner, so for the private sector the responsibility of the individual lawyer. The idea of “valuable service” was actually introduced in 1897, not 1952 when it became official policy and who, I might ask, was thinking about thinking about prior art or searching 130 years ago? The idea was conceived to address fraud upon the public. Ironically, by not having a defined role for searching fraud upon the public is subtlety encouraged. I’m thinking here about young Mr. Cuozzo and the search he didn’t get. What constitutes “valuable service” does not include “other persons [with] other business before the office” as authorized by the regulations parent statute35 USC §2(d).
The definition of valuable service requires a 21st century definition.
It is my contention that the Office cannot reflect on search, and the fact that for 2 years we’ve seen nothing is due to long term ingrained orthodoxy.
COMPARATIVE MODELS
The Government Accountability Office (GAO) released a pair of reports last June on Patent Quality and Patent Searching:
1. Patent Office Should Define Quality, Reassess Incentives, and Improve Clarity and
2. Patent Office Should Strengthen Search Capabilities and Better Monitor Examiners’ Work;
The first report tells us many things but two stand out in particular: 1) most examiners don’t feel they have enough time to search and 2) that the incentive system is geared to quantity over quality.
The second report compares the USPTO with other Offices but particularly to the EPO. We learn that the EPO, employs a three member panel to examine an application. A first panel member searches, a second panel member does substantive examination and the last member oversees the case administratively. By separating search from examination the person doing the search focuses on that search task only and gives the search task time attention.
A first comparative model for searching therefore is a committee approach as practiced by the EPO.
A second comparative model is the private sector. In the private sector searchers and lawyers do different things as they should. Searching in the private sector is an independent full time activity. Applying references, negotiating with the examiner, and comporting with a multitude of case history, is not finding prior art. At the USPTO expecting one person to simultaneously become both a skilled searcher, and learned in patent law is misguided. They should encourage the development of experts in both functions.
Similarities between the EPO and private sector include:
1. EPO and private sector searchers produce a body of art representative of the applications technical teachings. In the US examiners specifically looks for art to apply to the claims. This approach enables a rolling search where the search is directed to claim language and is expanded in response to amended claims.
2. Both the EPO and private sector searcher spend equivalent amounts of time 8 – 12 hours on a search project. In the US the estimate for is 5 hours although a review of select IPR’s suggests 3 hours or less.
3. Searching full time engenders pride in the product.
Searching is an activity that requires full time attention. When full time attention is made to search, patent quality goes up.
OPTIONS
I see three options regarding patent quality and searching.
Stay the course.
The PTO is a production driven agency. The more applications an examiner processes the greater the reward. The further you go up the ladder the more you have to produce and the less time you have to do it. It is in this environment that new examiners are conditioned.
The Office holds monthly webinars under the EPQI. Two months ago the subject was Training for the Examination of High Quality Patents. This presentation made it clear that the Office focuses on legal aspects of examination. When it was specifically asked how examiners are taught to search, we were told that new examiners are taught how a search is structured given a handful of hands-on trials and then sent to their art units where it is expected the SPE will fill that raining need. There is no additional follow-up through career development.
Last month the topic was Examination Time Analysis. An audience participant directly asked if the Office has any plans to separate search from examination in view of EPO procedure. The answer was no. This person providing this answer was a group director. He referenced an early 90’s pilot and used the phrase “two people searching the claims.” Interestingly, I have heard this exact comment from other PTO officials. Searching the claims is part of the problem considering the objective of a search is to produce a body of art.
Stay the course means that the Office continues the status quo and waits for the automated pre-examination search to develop. Examiners currently receive minuscule training on searching so that will continue. The haphazard rolling search that is evolving will continue. Low quality patents will continue to issue.
Adopt a Committee Approach
This is the EPO model. A patent examiner will wear one of three hats during any single examination. This of course will require a mass conversion of 8-9k examiners. Considering that it takes two years to train a searcher, a reasonable transition doesn’t seem possible
Start New
The primary motivation to start new is accepting the maxim: A quality search product is produced by a full time professional searcher. The only way to strengthen the search process is to visibly declare that searching is valuable service to persons with other business before the Office.
The Office has the ability to form an internal cadre who then become search experts. Search would ideally be embedded at the Tech Center level so they could work on a broad range of technologies thereby strengthening their specialists’ skills. Alternatively the cadre could be centralized similar to the STIC arrangement.
Starting new will require the PTO to publicly assert the value of the search function.
CONCLUSION
Before the Office can do anything they must understand what search is.
In closing I’m going to list several open ended questions. A reasonable person I think would expect the PTO to know the answers. Considering that their four search related initiatives avoid any hint of self-assessment and instead clearly imply that either technology or a foreign patent office are the solutions tells me that they can’t answer the questions.
- How long does it take to train a searcher?
- Should a search be conducted before examination or during examination?
- What is the objective of a search?
- What is the nature of the skill?
- Can the PTO fully search 600k applications?
- What is the purpose of the cited reference section?
- Does Rule 56 impact search quality?
- Should there be an Invention Statement?
- Etc. etc. etc.
A cogent search policy that can address 21st century demands is required so the PTO can get beyond the current crisis period. If the PTO separates search in-house, a corresponding policy directed to the private sector would have to follow.