Peer To Patent Sequel: USPTO To Begin New Pilot Program

The initial Peer To Patent pilot program, which began in 2007, opened the patent examination process to public participation in the belief that such participation would accelerate the examination process and improve the quality of patents. Yesterday the United States Patent and Trademark Office announced a sequel to the initial pilot program and will begin a second Peer To Patent pilot program, again in coordination with New York Law School’s Center for Patent Innovations (CPI). This new Peer To Patent program will run for a one year term and will commence on October 25, 2010. This second Peer To Patent pilot program will expand on scope of the previous pilot program. You may recall that the first Peer To Patent pilot was limited to software and business methods applications, but this new pilot program will also include applications in the fields of biotechnology, bioinformatics, telecommunications and speech recognition.

Under the new Peer to Patent pilot program, which is a collaboration with the USPTO and New York Law School and is funded by corporate sponsors GE, HP, IBM, Article One Partners, Microsoft, Open Invention Network, and Red Hat, inventors can opt to have their patent applications posted on the website. Volunteer scientific and technical experts will then discuss the participating patent applications and submit prior art they think might be relevant to determining if an invention is new and non-obvious. After the review period, the prior art is transmitted to patent examiners at the USPTO for their consideration during examination. Whether to use the prior art located is up to the patent examiner assigned to the case.

In the first Peer To Patent pilot, more than 600 items of prior art were submitted for 189 applications and more than 2,700 registered peer reviewers from over 140 countries participated. In a survey of USPTO patent examiners with a Peer To Patent application, 73 percent of those who responded said they thought the program would be helpful if implemented into regular office practice. During the two years the Peer To Patent was running as a pilot, examiners used art found by peer reviewers in approximately 20 percent of the applications reviewed.

“By encouraging participation by inventors in a wider array of fields, we hope to gather information that will allow us to further test the value of peer review to patent examination,” said USPTO Commissioner for Patents Robert Stoll.

“We are excited about the opportunities presented by this second Peer To Patent pilot,” Mark Webbink, Executive Director of CPI, said. “Not only does it continue this preeminent test of open government, but it provides law students from New York Law School and other participating law schools a terrific learning laboratory. We look forward to our continued collaboration with the USPTO.”

As already mentioned, there will be some changes in this new pilot, which include:

  • the number of eligible subject matter classes increased threefold;
  • peer review time to search for prior art reduced to three months;
  • the number of eligible participating applications expanded from 400 to 1,000; and
  • the number of items of prior art forwarded to the USPTO reduced from ten to six items.

The Peer To Patent pilot does certainly further the Obama Administration’s goal of transparency, which is why it was curious that the original pilot elapsed and there was a period of time where no Peer To Patent program existed.  Perhaps that is due to the fact that it was necessary to take a step back and review what had been accomplished, its overall value and whether to expand it across the board or try a second pilot program.

I personally have nothing against Peer To Patent, and think that anytime we can engage in activities that will lead to higher quality patents and more certainty in the rights ultimately obtained it is a worthwhile endeavor, provided of course that the quest doesn’t prolong the patent prosecution process.  I have seen no evidence to suggest that Peer To Patent would slow up the examination process, and it seems logical to conclude that applications opened up to such scrutiny are likely to be vetted to a greater degree given community participation.  Thus a patent that is granted should be stronger due to the fact that prior art not located by the patent examiner or applicant has been considered.

The statistics show, however, that Peer To Patent is not the panacea anticipated by anti-patent activists, and to a large extent demonstrates that between USPTO searching and Information Disclosure Statements provided by applicants, the great majority of findable prior art is located already.  This is confirmed by the fact that only 20% of the applications reviewed incorporate Peer To Patent prior art, which means that the overwhelming majority of prior art located is not relevant to establish a proper rejection.  So while there is no harm in pursuing Peer To Patent, and good reason to make the patent process more transparent, history suggests that it does not lead to the discovery of usable prior art in the vast majority of cases, which may be why the number of references that can be forwarded to the Patent Office has been cut from 10 to 6.  Nevertheless, a patent vetted by the community and ultimately issued seems as if it ought to enjoy a particularly strong presumption of validity.


Warning & Disclaimer: The pages, articles and comments on 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 as of the time of publication and should not be attributed to the author’s employer, clients or the sponsors of Read more.

Join the Discussion

18 comments so far.

  • [Avatar for Ron Hilton]
    Ron Hilton
    October 25, 2010 09:25 am

    How profound…

  • [Avatar for Blind Dogma]
    Blind Dogma
    October 25, 2010 06:49 am

    If you have to ask the question, then the answer is “You do not see it.”

  • [Avatar for Mark Nowotarski]
    Mark Nowotarski
    October 24, 2010 06:10 pm

    Anyone else see an examination issue here?

    In what sense?

  • [Avatar for Blind Dogma]
    Blind Dogma
    October 23, 2010 09:43 pm

    And yet, after 6 – 7 office actions each

    Anyone else see an examination issue here?

  • [Avatar for Ron Hilton]
    Ron Hilton
    October 23, 2010 07:33 pm

    Assuming that all the prior art cited meets the necessary legal thresholds to be admissable in the case in question and is therefore “good” in that sense, then I would define one piece of prior art as being “better” if if more fully serves to narrow the scope of or reject a given claim than another piece of prior art. It may be that there exists “better” non-patent prior art that would have caused the Red hat cases to converge to a resolution more quickly than the patent prior art, but other factors can elongate the prosecution as well. For the one Red Hat patent that has issued, the important question in terms of quality is whether it would have been further narrowed or even abandoned if non-patent prior art had been brought to bear.

  • [Avatar for Mark Nowotarski]
    Mark Nowotarski
    October 23, 2010 06:52 pm

    Speaking of open source, I recently reviewed a block of seven Red Hat patent applications that are currently undergoing examination. Ironically, not a single piece of nonpatent prior art has been cited. And yet, after 6 – 7 office actions each, only one out of the seven cases has been allowed. None of the remaining six cases, has been abandoned or appealed. Red Hat just keeps narrowing the claims.

    Does this say that the patent literature being cited is effective (few allowances) or not effective (no abandons or appeals)? Would “better” prior art mean that the cases would either be allowed, abandoned or appealed earlier?

  • [Avatar for Anon As Well]
    Anon As Well
    October 23, 2010 06:07 pm



    How do you “winnow down” something that has a below-one-per-month average?

    A: Can the program.

  • [Avatar for Ron Hilton]
    Ron Hilton
    October 23, 2010 05:44 pm

    I’ve discussed the Peer to Patent programs with some open source leaders who are opposed to software patents (including Bruce Perens, most notably). I argued that the open source community should take both an offensive and defensive approach to patents, building their own patent portfolio on the one hand (e.g. the “patent commons”) and challenging questionable software patents on the other hand (that’s where Peer to Patent entered the discussion). The problem is that they do not consider software to be patentable subject matter in the first place, so for them to participate at all in a program like Peer to Patent could be viewed as a tacit admission that software may in fact be patentable. Having said that, I still think the program has great potential value, in software and in other fields as well. Most of the references cited by USPTO examiners are from the patent literature. Even a few non-patent references from subject-matter experts in the relevent fields could prove to be very valuable in many cases. It’s the quality, not the quanity of prior art that matters. Indeed, it’s very counterproductive to have too many prior art references to wade through when a few really good ones will do the job. That’s why Peer to Patent includes a voting process to winnow them down if necessary.

  • [Avatar for Anon]
    October 23, 2010 03:41 pm

    An associate of mine, light and somewhat idle, decided to crunch the numbers available on the Peer-to-Patent website.

    As if the pronouncements above were not indicative of just what a “boondoggle” this program is, consider these numbers from Version One, running just over two years (26 months) and failing to obtain even a “D” grade on the rather minuscule goal of 400 applications (57%):

    Average number of discussions per case: 3.3
    Average number of pieces of “prior art”: 2.5

    Not “thousands of”, not “hundreds of”, not even “tens of” discussions and “prior art” were generated for each application on average. You don’t even need the thumb on your left hand to show the results.

    For all of the clamoring on the anti-software patent websites about the loads of prior art available, its downright amazing that somehow no one could take the time to stop complaining and actually do something about it.

    The term “vocal minority” appears to not even be applicable, when the deafening whisper of participation makes libraries seem like rock concerts.

    For “certain websites” remaining unidentified, but beginning with Patently and ending with O, the sum total of conversations for the entire program of version one is about half of the unfinished current month.

    Someone please tell me why we are resurrecting this program?

  • [Avatar for Steve M]
    Steve M
    October 21, 2010 04:57 pm

    Thanks Gene; that approach makes sense.

    If you’re trying to “sneak” an invention through; especially if it’s no more than a slight improvement over the prior art and/or is likely of little if any commercial value (perhaps and/or if it’s for “defensive” purposes only?); one would likely not want the increased scrutiny.

    But for a broad-scope / foundational / pioneering / breakthrough invention, I can see actually embracing P2P.

    That said, Blind’s points that in the end it will matter little (how may; and for how long will; folks be willing to work for free?) to pendentcy & patent quality; and that time & $$$ resources are better spent elsewhere are, I believe, well taken.

    Frankly, I’d like to see a new computer/search system for the examiners, an immediate 25% across the board pay increase for all examiners including SPEs (for which I’d be willing to pay 25% higher fees), a 25-50% increase in time/application, allowing qualified examiners to work from other cities, a hiring requirement that all examiners be unquestionably excellent in both spoken and written English (“proficient” is not nearly good enough), the end of fee diversion, and an immediate 1 billion dollar investment/payment to the PTO so Mr. Kappos and his hard-working staff can continue doing what we all know needs to be done.

  • [Avatar for Mark Nowotarski]
    Mark Nowotarski
    October 21, 2010 04:35 pm

    We (Tom Bakos – insurance actuary and myself, patent agent) were very active participants in the first Peer to Patent evaluation. We reviewed applications, had our submitted art cited by examiners, and drafted two “test” patent applications specifically designed for Peer to Patent review. Our experiences are described in “Canaries in a Coal Mine”

    We had great participation from peer reviewers. Tom had several Fellows of the Society of Actuaries leave useful comments on his. I had an entire class of MBA students leave comments on mine. There wasn’t anything particularly new in the submitted prior art, however, which is what we were hoping for.

    So the bottom line is that we didn’t see anything particularly bad about the process, but we didn’t see much benefit either.

  • [Avatar for Blind Dogma]
    Blind Dogma
    October 21, 2010 02:32 pm

    But if Peer to Patent can substantially reduce bad allowances, say from 20% to 5%

    Chose again (carefully). This stat is not what is being proclaimed. This is not a blind-study test where the art would not have been made of record or considered by the examiner lacking the peer to patent. Further there is zero study of coorelation of supplied art to the prevention of a “bad allowance.” This is merely another channel of providing art to the examiner on the front end. That’s it.

    There are far more worthier avenues of developing far better capabilities of providing art to examiners than this boondoggle and public relations ploy.

    I agree with you that the status quo in terms of patent quality examination is not really acceptable. This “Peer to Patent” effort is hardly the goal to that quest, and as far as it diverts resources that could be better spent, it is a detriment. Now if the resources spent are minimal, as they appear to be, then sure, as a public relations ploy, this works fine. But make no mistake, this will not substantially reduce bad allowances by any stretch of the imagination. You may chose to think carefully here.

  • [Avatar for Malcolm Scoon]
    Malcolm Scoon
    October 21, 2010 02:11 pm


    Great incite on Peer to Patent. Thanks.

  • [Avatar for Gene Quinn]
    Gene Quinn
    October 21, 2010 01:35 pm

    Steve M-

    Sorry to sound like a lawyer… but “it depends.”

    I would NEVER suggest a client opt into Peer To Patent unless we had done what I believed to be a thorough and complete patentability search and assessment and wrote the application accordingly.

    I would also NEVER suggest a client opt into Peer To Patent if they were looking for a bargain basement patent application. In that situation what you do is write the claims, then write the specification to match the claims. The specification doesn’t get nearly the work, time and breadth it deserves to be a foundational document for a possible entire patent portfolio.

    If a client had a real innovation that seemed exciting and was going to be investing substantial sums of money into the innovation then I would certainly talk to them about this process, but the call would be theirs. I think in the right circumstance, where an enormously broad Specification is written with an eye toward many facets of the overall innovation, it would be appropriate. The thing I like about Peer To Patent is this: (1) only 20% of the time is relevant prior art located; and (2) it would be quite difficult for a defendant to convince a district court judge the claims are invalid if the patent owner could argue that the application was willing laid open for public inspection and nothing was found that prevented the issuance of the claims. In essence, I think at the district court level Peer to Patent would have about the same effect on the judge as would a reexamination certificate.

    I think Peer To Patent results in more certainty in the claims obtained, which is definitely positive.

    What do you think?


  • [Avatar for Ron Hilton]
    Ron Hilton
    October 21, 2010 12:10 pm

    I chose my words carefully. I said that one bad patent “_may_potentially_” cause significant economic harm. I did not say that every bad patent inevitably will cause such harm. The majority of patents sit on the shelf without doing any good or harm economically. But if Peer to Patent can substantially reduce bad allowances, say from 20% to 5%, then the substantial back-end cost of litigating bad patents will also be reduced. An ounce of prevention is worth a pound of cure. Of course perfect quality is not obtainable in reality, but I think it is widely acknowledged that the status quo in terms of patent quality is not really acceptable. Anyone who questions that is the one who needs a reality check.

  • [Avatar for Blind Dogma]
    Blind Dogma
    October 21, 2010 11:53 am

    Even one bad patent that gets issued over valid prior art may potentially cause significant economic harm.

    That must really suck then, that an Examiner error rate of 4.5% is considered top line outstanding.

    Time to reset your expectations to reality, Ron.

  • [Avatar for Ron Hilton]
    Ron Hilton
    October 21, 2010 09:48 am

    Even one bad patent that gets issued over valid prior art may potentially cause significant economic harm. So I think the fact that the first pilot program uncovered prior art that otherwise would have been overlooked in 20% of the cases is not a bad result at all. It shows that the 80/20 rule is alive and well (80% of the benefit can be derived with about 20% of the effort). But the corollary is also true, that is often takes 80% more effort to get the last 20% of the benefit. And if that last 20% included patent applications that would have wrongly been granted, with the potential harm that such would have caused, then the 80% extra effort from the program was worth it. In some fields of endeavor, 80% quality is “good enough” but patent examination is not one of those fields.

  • [Avatar for Steve M]
    Steve M
    October 20, 2010 08:46 pm

    Gene–so would you want any of your clients putting their inventions through this process?