MIT Prior Art Archive: An Overstated Solution to Patent Examination

By Steve Brachmann & Gene Quinn
November 6, 2018

On October 3rd, the Massachusetts Institute of Technology and San Jose, CA-based networking giant Cisco co-hosted a launch event for the Prior Art Archive, a database open to both the public and examiners at the U.S. Patent and Trademark Office. According to those collaborating on the project, the database is intended to provide a means through which patent-invalidating prior art can be more easily found by patent examiners during patent prosecution at the agency.

According to an MIT News article on the launch of the Prior Art Archive, Cisco has already uploaded a total of 165,000 documents into the database. A number of other major tech companies have also committed to collaborating with the project including Amazon, Microsoft, Intel, Dell, AT&T and Salesforce. Technology used to classify prior art submitted to the database has also been developed by Google.

USPTO Director Andrei Iancu attended the database’s launch event where he delivered remarks about the importance of prior art to ensuring that patents issued by the agency are of high quality. His remarks touched on a story from the 1960s about a Danish inventor who developed a way to resurface a sunken freighter off the shores of Kuwait with the use of 27 million polystyrene foam balls. The inventor went on to obtain patents from the U.K. and Germany but was denied a patent by the Dutch patent office because of a 1949 Donald Duck cartoon submitted as prior art. The Sunken Yacht, a cartoon in which Donald and his nephews Huey, Dewey and Louie raise a sunken ship by filling it with ping pong balls, disclosed the same technique claimed by the Danish inventor in his patent application.

All intriguing anecdotes aside, but will it matter? There are legitimate questions of whether or not the MIT-Cisco prior art archive will be successful in encouraging third-party prior art submissions based upon data we’ve recently received from the USPTO.

[[Advertisement]]

Third-party pre-issuance submissions of prior art can be made before the earlier of— (1) the date a notice of allowance is given or mailed in the application for patent; or (2) the later of— (i) 6 months after the date on which the application for patent is first published by the Office, or (ii) the date of the first rejection of any claim by the examiner during the examination of the application for patent. The purpose is to allow the public to submit prior art examiners should know about and consider prior to deciding whether to issue patent claims.

According to statistics provided by the USPTO, since the beginning of fiscal year 2012, the Office has received a total of only 1,584 third-party submissions of prior art for consideration by patent examiners. The high water mark occurred in 2016, when the office received a total of 329 third-party prior art submissions. This declined to 266 submissions in 2017 and in fiscal year 2018, the USPTO received a total of only 141 prior art submissions.

In both FY 2016 and FY 2017, the USPTO received in excess of 600,000 utility patent applications. In FY 2016, the USPTO received 607,753 utility patent applications and in FY 2017, the USPTO received 602, 354 utility patent applications. See Table 2, USPTO Performance and Accountability Report FY 2017.

To say third party prior art submissions are rarely made is an understatement. But why?

Large entities that supported passage of the America Invents Act (AIA) said they only wanted new ways to challenge bad patent claims and didn’t begrudge anyone from getting and keeping valid claims. The truth, however, is attorneys for these companies routinely recommend not filing third party pre-issuance submissions because it would only offer the opportunity for a patent applicant to work with the patent examiner to more narrowly tailor the claims that issue, making them stronger and harder to challenge because examiners would have considered and applied the best prior art. 

But even if patent examiners had every conceivable piece of prior art from every searchable database, other USPTO statistics suggest that patent examiners simply wouldn’t have enough time to consider every prior art reference that could be relevant during patent prosecution. According to the USPTO, during the 2018 fiscal year, patent examiners averaged 21 hours per case. Of course, the time examiners receive varies depending upon the complexity of the technology claimed in the application, but an examination period that on average equates to half a typical work week doesn’t seem as though it would provide adequate time for the full consideration of all prior art that could potentially be submitted. 

Patent examination is about a choice. If we want to make sure that no mistakes are made patent examiners will need more time per case, and the budget for the USPTO will have to be significantly increased, tens of thousands of new patent examiners will need to be hired, or the patent examination process will come to a screeching halt. Pursuing such policies would be a poor policy choice when there is simply no public interest in making patent examination better. Third party pre-issuance submissions are not being used by anyone, including Cisco and others in the infringer lobby.

 

The Author

Steve Brachmann

Steve Brachmann is a writer located in Buffalo, New York. He has worked professionally as a freelancer for more than a decade. He has become a regular contributor to IPWatchdog.com, writing about technology, innovation and is the primary author of the Companies We Follow series. His work has been published by The Buffalo News, The Hamburg Sun, USAToday.com, Chron.com, Motley Fool and OpenLettersMonthly.com. Steve also provides website copy and documents for various business clients.

Steve Brachmann

Gene Quinn is a Patent Attorney and Editor and founder of IPWatchdog.com. Gene is also a principal lecturer in the PLI Patent Bar Review Course and an attorney with Widerman Malek. Gene’s specialty is in the area of strategic patent consulting, patent application drafting and patent prosecution. He consults with attorneys facing peculiar procedural issues at the Patent Office, advises investors and executives on patent law changes and pending litigation matters, and works with start-up businesses throughout the United States and around the world, primarily dealing with software and computer related innovations. is admitted to practice law in New Hampshire, is a Registered Patent Attorney and is also admitted to practice before the United States Court of Appeals for the Federal Circuit. CLICK HERE to send Gene a message.

Warning & Disclaimer: The pages, articles and comments on IPWatchdog.com do not constitute legal advice, nor do they create any attorney-client relationship. The articles published express the personal opinion and views of the author and should not be attributed to the author’s employer, clients or the sponsors of IPWatchdog.com. Read more.

Discuss this

There are currently 10 Comments comments. Join the discussion.

  1. David Stein November 6, 2018 1:02 pm

    Yes, dumping more prior art on examiners is unproductive.

    In fact – it’s counterproductive. When an applicant traverses the current round of prior art and the examiner chooses to allow it, it’s allowed over all references of record… whether or not the examiner really reviewed it in substance. What if that pile of references happens to include an on-point teaching that anticipates the allowed claims?

    I believe that this is an underappreciated problem. Even references that an examiner reviewed and deemed inconsequential *to a first set of claims* might teach new elements in *a later set of claims* that prompted the examiner to allow them. Do examiners really go back and re-review references when the claims change?

    > But even if patent examiners had every conceivable piece of prior art from every searchable database, other USPTO statistics suggest that patent examiners simply wouldn’t have enough time to consider every prior art reference that could be relevant during patent prosecution. According to the USPTO, during the 2018 fiscal year, patent examiners averaged 21 hours per case. Of course, the time examiners receive varies depending upon the complexity of the technology claimed in the application, but an examination period that on average equates to half a typical work week doesn’t seem as though it would provide adequate time for the full consideration of all prior art that could potentially be submitted.

    Yes. This, right here.

    If we want better examination, two things must occur concurrently:

    (1) Give examiners more time per office action, and

    (2) Raise the expectations of technical accuracy in examiners’ citations of references.

    It should not be acceptable for examiners to make categorically wrong statements about the claimed invention, due to not reading the specification (ostensibly due to lack of time, but also due to tolerance by the PTO).

    It should not be acceptable for examiners to cherry-pick words from a reference out of their technical context, even *contrary to* their context, in order to stitch together a ransom-note-style rejection of the claims based on a reference that really doesn’t teach the subject matter.

    It should not be acceptable for examiners to disregard arguments about the technical content of the references, with a lazy reply like:

    “Applicants’ response has been considered but is not persuasive because [copy-and-paste the last rejection]. Applicants’ discussion of the references is immaterial because BRI and *In re Van Geuns*.”

    Not acceptable! The PTO should not tolerate this and should penalize examiners who engage in these tactics.

    I realize that what I’m suggesting – raising expectations of examiner work product, and reducing deference for vague, lazy examination practices – are difficult goals for the PTO. But if we perceive patent examination quality as problematic, that’s where the battle must be fought. Giving examiners more information and more tools (and also more deference, such as § 101) won’t address the problem.

  2. Benny November 7, 2018 6:04 am

    I’ve submitted third party observations in the past, showing lack of novelty. I think the USPTO should actively encourage the procedure, since it can genuinely help the examiner and save time. Also, the USPTO examiners should seriously consider the submitted prior art, something I’m not sure they do.

    But there is a workaround. Submit your application with overly broad claims. The third party can only respond to the claims as submitted, and after you narrow your claim down to what is in fact your invention, after the first office action telling you not to be silly, the third party no longer has an opportunity to react.

    Several times I have seen an examiner cite distant prior art, maybe a 103 rejection, when a 102 reference could be found with 10 minutes search, if you know where to look.

  3. Anon November 7, 2018 8:51 am

    For me, what resonates about this article is that time and resources are not the driving force in the clearest examples of poor examination.

    It is not a question of “more art.” (Notwithstanding that Benny may be a small part of a very small percent engaging in providing such)

    It is not even a question of “more time.” (Noting that even Benny indicates that a better focused search takes less time than apparently it takes with the providing of more art)

    It is a question of efficiency and metrics of that efficiency that simply do not drive self-improvement; coupled with an unwillingness to make deep structural differences with metrics (and enforcement) that matter.

    Additionally, far too many “band-aids” (and ease of applying those band-aids) with NO concurrent or close to anywhere near meaningful feedback mechanism into original examination from post grant reviews having been put into the system, make it simply too easy for the Office itself to “play the game,” allow (or not) — with no real consequence — and then later revoke; obtaining fees both coming and going.

    The answer is counterintuitive: make post grant review MORE difficult and force the “pain” to be taken more seriously in the pre-grant (and make it actual) examination.

    This “pain” is neither a Reject Reject Reject mechanism, making the applicant “prove” they deserve a patent, nor even (necessarily) more time [especially if more time is merely more time doing the same type of shoddy examination]. No, the “more pain” is enforcing actions TO BE on the merits and not “key word” puzzle-piece putting together. There should be consequences (and rewards) for both allowances and rejections that reflect that the first step of reading and understanding the innovation was properly completed.

    There is no real “easy” answer in part because it will not be easy to admit that far too much current examination is wasteful (of the time already given) and far too many Examiners are merely marginally competent (at best).

    This will not be easy because the job itself is not easy, and certainly, pay should be greatly increased to reflect that difficulty (and along with that pay must come greater accountability).

    Either make the examination better than the possible monied after grant review by third parties or eliminate it completely.

    This current “hybrid” of “well, we gave it a shot, and those interested third parties with money to spend will spend it on post grant mechanisms” is fatally flawed and the worst of either option of meaningful or no pre-grant examination.

  4. Benny November 7, 2018 10:50 am

    Anon,
    It looks like you wrote “Government employee (examiner)”, “consequences” and “accountability” in the same post. In trade union speak, that is a capital offence.

  5. CP in DC November 7, 2018 12:19 pm

    The problem with examination is the reward system. It’s the count that matters, how you achieve it is irrelevant. More time, more art, more anything doesn’t change the ultimate goal, more counts. The less effort and less time spent on achieving the count is what currently matters. Change that metric and examination may change.

    It’s not that the examiners don’t have the resources to find the relevant art, it’s that they don’t care to find it. The Japanese patent office uses the same prior art databases that the USPTO uses and the Japanese Office Actions consistently find the relevant art. I’m always impressed with the cited references. I spoke to the person who set it up, and he confirmed that the JPO used the same databases.

    You can give the examiners more time, but that won’t matter. In fact every time you file an RCE the examiner gets more time on the SAME application. Ok, now a show of hands of those who file an RCE and get new rejections or new art cited? Yeah, no one…. You often get a copy and paste job (with the typos) of the last final. So much for more time.

    When the PTO was implementing all the “quality measures” to improve examination I asked the simple question: why not study the PTAB opinions that overturn examiners? It turns out the PTAB and the examining corp are separate and not allowed to interact. So you never get PTAB judges (no matter what you think of them) telling the examining corp what they often get wrong or right. No feedback loop from the one source that often reviews examiner’s work.

    I agree with the proposition that we should make examination the center of the debate. Make examination difficult so we can obtain the best examined patents when they issue. As recognized, that is a difficult job that you just can’t hire anyone to do. Also you may need to fire a fare share of examiners that are not up to the job. Judge Paul Michel spoke to congress about the need to easily fire bad examiners and REWARD the good ones. If we made examination the focus, then applicants would have a fair chance of getting an easily defended patent (that means no IPR implementation), because the junk that gets through bad examination would cease to exist.

    Of course, the examination has to be fair to applicants. Current examination is anything but fair. An examiner formulates some junk rejection with conclusory rejections and then sits back waiting for the applicant to prove why they deserve a patent (read the statute to see how burdens work). Persist, wear the examiner down, don’t amend to meet description or enablement (examiners rarely reject on those grounds), and eventually you can get patents. I know prosecutors who do this, of course, the result is more junk. This junk only serves as examples for the anti-patent crowd. I can do without that.

  6. David Stein November 7, 2018 12:23 pm

    @anon: I’m right there with you.

  7. Anon November 7, 2018 1:39 pm

    Sadly, Benny – that is very much part of the problem.

    Pretending that it does not exist serves no one.
    Not the Patent Office.
    Not the examiners.
    Not the public.
    And certainly not the innovators.

  8. Stephen Harper November 7, 2018 4:29 pm

    I’m a little confused by the statistics mentioned in the article:

    “According to statistics provided by the USPTO, since the beginning of fiscal year 2012, the Office has received a total of only 1,584 third-party submissions of prior art for consideration by patent examiners. The high water mark occurred in 2016, when the office received a total of 329 third-party prior art submissions. This declined to 266 submissions in 2017 and in fiscal year 2018, the USPTO received a total of only 141 prior art submissions.”

    According to the USPTO-posted information I was able to access, the numbers of third party submissions have been significantly higher. Still not a large % of the total applications being filed, however.

    Could the exact source of the statistics be cited?

  9. Gene Quinn November 7, 2018 5:30 pm

    Stephen-

    I asked the USPTO how many third party pre-issuance submissions they received. The information they provided me via e-mail is what we published. I was shocked to see the numbers were so low myself.

  10. LazyCubicleMonkey November 10, 2018 4:38 pm

    Patent examiners definitely need more time to examine patent applications. It’s telling that who patent examiner is greatly influences whether a patent gets published.

    As such, this study estimates that as as 28% of the patents would be invalidated by the courts (from 2012, so how well this study would hold up in today’s legal climate is anyone’s guess): http://mason.gmu.edu/~smille9/ShawnMillerPublication2.pdf

    Perhaps because there’s some mis-aligned incentives?
    https://scholarship.law.duke.edu/cgi/viewcontent.cgi?article=6327&context=faculty_scholarship

    A few modest proposals:
    https://www.brookings.edu/research/decreasing-the-patent-offices-incentives-to-grant-invalid-patents/

    Nice to have some empirical evidence instead of just anecdotal evidence.

Post a Comment

Respectfully add to the discussion.

Name *
Email *
Website

Our website uses cookies to provide you with a better experience. Read our privacy policy for more information.Accept and Close