Our Expanded Regime of Submarine Prior Art

by Dennis Crouch

The general rule in our new first-to-file patent system is that your effective application filing date* is of utmost importance.  In general that date is the trigger-date for prior art.  Prior Publications count as prior art (102(a)(1) prior art) as do Patents and Patent Applications filed prior to the trigger-date (102(a)(2) prior art).  To be clear, publications are generally thought of as prior art as of their date of publication, but we have a special rule for U.S. patent filings — once published or patented they become prior art as of their effective filing date. [Note, the statute stretches the date back to the “earliest [priority] application that describes the subject matter”].

I have identified these 102(a)(2) prior art patent documents as “submarine prior art” because they are kept secret (usually for 18-months) and then suddenly emerge as back-dated prior art.  Pre-AIA law included the submarine prior art under what was then known as 102(e) (2010).  However (and in my view) the new law expands the scope of submarine prior art in a few ways. First, under the new law applicants can no longer swear-behind prior art based upon their prior invention date.  Second, under the new law the submarine prior art will stretch back further in many cases to encompass (for instance) the original foreign-priority filing date. [As with the old rule, a prior application by the identical inventor does not create submarine prior art].

A Company’s Own Prior Patent Filings: One exception to the submarine-prior-art issue is 35 U.S.C. §102(b)(2)(C). That sub-section indicates that neither prior-filed patent application publications nor their resulting patents will be deemed prior art if:

(C) the subject matter disclosed [in the prior document] and the claimed invention . . . were owned by the same person or subject to an obligation of assignment to the same person [as of the claimed invention’s effective filing date].

The basic idea here is that a company’s own secret-filed applications will not serve as prior art against the company itself.**

The Pre-AIA statute included a similar exception codified in 35 U.S.C. 103(c). However, the new exception is more powerful in several ways.  Most notably from the face of the statute is that the old law only excused prior art for obviousness purposes but not for novelty purposes while the new law negates the references for seemingly any prior art purpose.  Second, the expansion of submarine prior art (noted above) makes the exception relatively more important. As more applications continue to be filed in crowded technology spaces, I expect that the exception will continue to rise in importance.

Questions that I would like to pursue is how important this expanded submarine prior art is to the system and likewise the relative importance of the intra-company-exception.***  Theoretically, the exception favors (a) larger entities who file many incremental patent applications with a variety of inventorship team combinations as well (b) as applicants who file applications under non-publication requests.  For examiners, do you consider whether an applicant can claim the exception before issuing a rejection or do you wait for the applicant to claim the exception.

This question will be part of one project that I’m planning to pursue this summer that will provide some initial studies on how the AIA has impacted patent prosecution.

On-point comments are very welcome.


* The effective filing date is defined on a claim-by-claim basis as “the filing date of the earliest application for which the patent or application is entitled, as to such invention, to a right of priority under section 119, 365(a), or 365(b) or to the benefit of an earlier filing date under section 120, 121, or 365(c).” 35 U.S.C. 100(i).

** Patents associated with inventions developed via a Joint Research Agreement (JRA) can also be avoided by parties to the Agreement.  35 U.S.C. 102(c).

*** I should note here that all countries have some form of submarine prior art, but many follow the European approach that the submarine art is not considered for obviousness (inventive step) purposes because of the reality that the secret prior art would not have been within the grasp of a person of skill in the art (since it was secret at the time).


156 thoughts on “Our Expanded Regime of Submarine Prior Art

  1. 14

    Not to be a critic to anything said here, but aren’t we now evolving to a defensive publication strategy, as in the old IBM Abstracts?

    And, is there a business plan about to publish previously unpublished abandoned applications?

    1. 14.1


      I have not heard anything re: the change to possibly publishing abandoned apps – can you tell me more?

      If this is a thought unilaterally by the Office, I don’t think this will get very far.

      1. 14.1.1

        This is not a policy consideration by the USPTO. However, right now, patent applicant’s cannot expect to receive a patent. The strategy for publishing has changed, particularly since the fee has been dropped.

        However, all those who filed before and abandoned their applications without publication may now reconsider. Publication would be by a private internet site. As I suggested, maybe this would be a potential business plan for generating revenue. Perhaps pat2pdf.org could take on this task.



          I am still uncertain as to whether or not the government is planning on this “publication” effort.

          Perhaps because anyone who themeselves have abandoned an application is (and always has been) free to publish them whenever.

          Of course, the publication date of these previous submitted, abandoned and never seen before items will have a current publication date and not be back-dated to their filing date, so the impact of such as prior art would only be forward looking…


            “Free to publish them whenever,” is true, what is needed is a private website publisher, as there is no plan to my knowledge of the government doing such. You are correct as to the effective date, but there are situations where even that can be an advantage.

    2. 14.2

      I think this has to do with professors who like to publish and only later consult patent attorneys.

  2. 13

    Is this an example of a submarine patent?

    Patent: 8,126,822
    Filed: 5/21/99
    Issued: 2/28/12

    Its emergence, 5+ years after mine has been hugely problematic. Sigh…


  3. 12

    On a more mechanical discussion of weaving the web of prior art, when you do your analysis I would suggest you may want to also explore the impact of extensive self-citations and its reach-back impact especially in light of M&A activity, real party in interest and attributable ownership issues; USPTO’s new embrace of patent families; and the use of combination sets of classifications as part of the search for prior art.

    1. 12.1

      Arleen, you made me think.

      Dennis the Tegernsee Group (the Big 5 Patent Offices) recently looked at the prior art effect of earlier unpublished filings, and how different jurisdictions handle them. The Group published a big report, which is available at least on the EPO Website.

      The non-EPC PTO Group Members were asked why their jurisdiction exempts earlier filings of the Applicant herself from having a prior art effect. The answer from Japan is revealing. For BigCorp, when drafting, it is just too difficult (in Japan) to take into account all the content of all the company’s many earlier filings.

      I didn’t see in the Report the reason the USPTO gave, but I guess it was “Because that’s the way we’ve always done it. And we are not going to change just because we are moving from FtI to FItF.

      But, as Arleen notes, there’s trouble brewing, for all those lacking a heavyweight legal team with a budget to match. That is, for all but BigCorp.

      1. 12.1.1

        A minor point, but which countries are in the Tegernsee group? You only talk of five? But which five?

        The UK, France, Germany, Japan, and the US seem to fit the bill. Are those the ones you refer to?

  4. 11

    Submarines and the quid pro quo. A scandal.

    I learn in this thread that PTO Examiners search unpublished applications and can suspend examination of later filings till the unpublished hotties are published.

    But they can’t tell the applicant what they have found in their search, till after the unpublished hottie is published.

    Doesn’t anybody have a problem with this? You can’t get a patent. the Examiner knows why but you don’t. and the Examiner can’t tell you why. A scandal.

    As to the Quid pro Quo, a lot of nonsense is writen about it. When you file, you have about 15 months to change your mind, pull the application, stop it being A published, and stay with trade secret protection. Offices like the UK, and to some extent the EPO, give you the search report early enough for you to pull your application before publication, once you see that the search reports knocks you out.

    Some say the A publication offers no clue what claims will go to issue. I dispute this. The appln as filed has a writen description that enables the claimed subject matter over its full scope. you are supposed to include your Best Mode. If that isn’t enough to reveal what might go to issue well then the courts are not enforcing the words of the statute. But on things like “written description” they are tightening up now, aren’t they.

    I know, I know, the issued claim will lie somewhere between the original claimed envelope and the detailed description. Not knowing precisely where, in between, is not the same as not knowing what subject matter is going to get to issue.

    1. 11.1


      Solution: Deny all prior art effect to unpublished applications.

      The theory of Milburn was that the applicant did all he could to get his patent published when he filed. But, if one chooses not to publish, that is no longer true.

      We need to revisit non publication.

      1. 11.1.1

        Wrong Ned – publication alone is not the key – you too seek to take from the patentee (hmmm, just like Big Corp)

      2. 11.1.2

        That earlier filings block later filings is fundamental to FtF. So Ned, are you urging a return to FtI?


          Max, the prior art effect of a patent as of the time of its filing date is a benefit. Why not deny the applicant of any rights of priority if they choose not to publish? They would have every incentive to expedite prosecution to get their patent issued. But if they chose to keep it secret, no harm, no foul, as we say. If it issues, the patent would then be good as prior art from the date of publication of the patent.

          In this way, submarine patents cannot emerge to sink earlier-issued patents.


            More quacks like Big Corp (and an absence of the appreciation of Quid Pro Quo, and the Congress crafted balance in PTA guarantees)…


            submarine patents cannot emerge to sink earlier-issued patents

            Can you tell the 99.99% of people who will never own a patent why they should care if some newly published patent “sinks” an earlier issued one?

            This tiny expansion of prior art is not a real problem for most people, including most patentees who’ve actually invented something.



              because people who don’t care about patent law, don’t care about patent law…

              Or something

              /off sardonic bemusement


              MM’s comment set me thinking. What does Ned mean by “priority”?

              Does he mean that the 102(e) effect of his non-published application is lost? just that and no more.

              Or does he mean if the Applicant requests non-publication, his application does not have “priority” over later filed patent applications, in the queue for patent rights?

              If the latter, that is a serious sanction Ned.

              If the former (as I had up to now supposed) I’m with MM, that the sanction is not serious enough to bother anybody. The non-publishers will still want their applications held back from publication as long as possible, till the infringers are fully committed to the technology.


                Max, priority and prior art are two different things. Overcoming prior art based on one’s filing date would still be available.

                This leaves the possibility of to patents issuing covering the exact same invention unless the PTO actually searches unpublished applications for the same invention. If they did find one, they could inform the second filer that there is an unpublished application claiming the same invention but still issue the patent and let the second filer deal with the problem of his own patent’s validity if and when the first file patent emerges.

                If the first filed patent issues while the second is pending, simply issue a double-patenting rejection against the second-filed.


            Attractive idea Ned. I would not be surprised if there were unforeseen consequences but I cannot immediately shoot down your cunning plan. I like it.

  5. 10

    Soooo many things missing from the conversations here…

    The guarantees of PTA as a mechanism so that publication did not rob the Quid Pro Quo deal.

    The fact (apparently oblivious to our foreign friend MaxDrei – which might be excused – but also from examiners -inexcusable – that pre-published art very much counts as prior art, is searchable, and should be used (the “on hold” time-out mechanism is real, and I have seen it on a portfolio that we inherited.

    The aspect of obviousness is one NOT geared to any one real life person (like the inventor), but is judged as to the non-real PHOSITA, so there is no “real” fault with actual people unable to see unpublished patents. If that were a real concern, then PHOSITA would NOT have the omnipotent and omniscient ability of knowing EVERY piece of pre-existing art (and yes, in the post-AIA world, they included art published in Swahili on the island of Grand Terre (of the famous Kerguelen Islands, of course) in the public library there – one of the fifteen books catalogued and openly available.

    But hey, who cares right? After all, the Quid Pro Quo was just for “grifters” anyway, right?

    The problem with the expansion of 103 is not due to Rich, but to the Supreme Court in super-empowering the PHOSITA now to be an inventive entity, as that is what enlarges the “world of prior art” available under 103 to be the same (or nearly the same) as the “world of prior art” available under 102. It is the Supreme Court that didn’t like its “nose of wax addiction” messes with that sought to broaden 103.

    1. 10.1

      anon, it ’tis a crying shame that the worst aspects of prior art were shoved down out collective throats by the AIPLA, an organization that we had long thought was there undo the damage caused by the ’52 Act/Rich with respect to secret prior art.

      1. 10.1.1

        In your haste at chasing the anti-Rich windmill, you are aiming at the wrong entity (again).

        The “world of prior art” – under Rich was always different for 102 and 103, because PHOSITA was never meant to be the augmented version that the Supreme Court has created in its anti-patent addictions.


          The person of ordinary skill would not necessarily know of the whole of the prior art.

          It is just what he would find obvious when considering the differences between the prior art and the claimed invention.

          Now, expand that prior art to secret prior art, stuff not in the public domain, we are cutting down on the scope of the patents of later filers in favor of earlier filers. This is FTF on steroids.

          And, recall, 102(f) is repealed. Patents are not invalid for not naming the correct inventors. So, FITF is a fraud on the public, a blatant, out and out lie told to Congress.


            You’ve got this all crabbed Ned.

            You really need to ask which “world” of prior art is pertinent.

            “Known” is universal for both – and PHOSITA is omniscient (so talk about “secret” is a bit of a misnomer).

            The difference under Rich (and Congress) was that 103 was purposefully and merely a subset of 102 to which the (lesser powered) PHOSITA was concerned with, primarily turning on the “In The Art” aspect. It was NOT that any art became “unknown,” it was that typical paths of invention (like cross fertilization of one art across another) were not supposed to be patent defeating.

            Blame your beloved Supreme Court for messing that up.


              Stuff “not in the public domain” was patent defeating even prior to the AIA – think about one patentee NOT getting a patent because a later filing patentee was able to swear behind the first to file but second to invent.


                Not swear behind.

                Invented before.

                Grounds for revocation and for an interference.

                1. Invented before…?


                  Conceived before (and not lapsed in effort to reduce)…. Yes.

                  Further, you not-so-deftly miss the point that such eliminates a “regular filer” with art that he personally could not know about.

                  Do you recognize the point that I put on the table?

                2. Is this (another) case where Ned’s version of being “shouted down” is accompanied by me asking him to speak up…?

                  That seems to happen a lot.

    2. 10.2


      I’d appreciate your input here. In terms of 103, I don’t see the problem, from a policy perspective, of including unpublished patents in the prior art.

      While not published, the application does demonstrate, to some degree at least, the state of knowledge at the time the application is filed.

      Of course, the applicant isn’t aware of the application, but it still demonstrates the state of knowledge in the art at the time of filing, right?

      1. 10.2.1

        Good point Arthur. Yes, of course, what inventors wrote in their patent applications prior to the date of the claim under consideration can indeed reveal something of the state of knowledge in the technical field of the invention prior to the date of the claim and so sometimes can be probative.

        The knowledge of any one individual though, is not decisive. We need an objective standard. That comes with the notion of the PHOSITA who has a defined store of knowledge. How shall we define that store? The way the EPC does it is to postulate some person who knows all that can be known by a person skilled in the art (so think Google and Google translate, and a perfect search algorithm) but does not know what cannot be known by such a person, such as unpublished applications at the USPTO. Inventors, engineers, judges, juries, business people VC folks, investors, journalists: they can all see the sense in the EPC definition of the PHOSITA and the EPC definition of the state of the art. Simples.

        Some lawyers don’t like that. They prefer to keep their legal world mysterious, something only they can interpret, like bibles in Latin in the Middle Ages. Obfuscation. A nice little earner, if you are the only one that understands. A nice little conspiracy, by the big and powerful, against the little people


          Max, there has to be a difference between skill and the state of the relevant art.


            The state of the relevant art – by the omniscient PHOSITA (not a real person, and certainly not the real person of an inventor)


              You are both up bright and early this morning I see.

              Ned, I don’t understand. Perhaps we have a different understanding of what “the state of the art” means. For me it is what Art 54 EPC says it is. Perhaps you think of it as “the common general knowledge of the PHOSITA”? Of course I agree readily that “skill” and “CGK” are different notions. Who doesn’t?

              Can we get this clarified.


                It is amazing how difficult it is for you to recognize (or admit) that our sovereign has a different law, MaxDrei.

                I have led you to this waters of wisdom many times, yet you refuse to drink.


                Skills vary. The art is what is publicly available in the field of the invention.

                Obviousness is judged by considering the art, the differences from the art claimed, to one of ordinary skill, not to one of exceptional skill, or to below average skill.

                1. Well yes Ned, I see that. What I still don’t see is why you admonish me with your words:

                  “Max, there has to be a difference between skill and the state of the relevant art.”

                  Whatever was it that I wrote, that prompted those words of yours?

                2. Ned,

                  Why do you insist on posting something when I have presented you with a point that you refuse to see?

                  Your “publicly available” version is just a little off.

                  You do however make a valid point that level of skill for the non-real person is important, but as I also told you, this is a Congress/Rich thing that has been messed with by the Supreme Court in their attempts to augment PHOSITA.

                3. MaxDrei’s lament to Ned sounds familiar…

                  Sort of like:

                  non sequitur about “not covering improvements” has no bearing on anything that I have posted. How did you end up there?


            Max, I still think you Euros may be confusing things.

            A person of ordinary skill in any art cannot know all of the relevant art. He necessarily is a person of limited knowledge.

            The art is defined by statute. Publicly known is the ultimate standard.

            The claimed invention is compared to the art publicly known to identify differences. It is only then that we bring the person of ordinary skill in to make a judgment as to whether the differences, in the context of the claim as a whole, was within his skill.

            Here, our determination of the person of ordinary skill is made by looking at the average engineer. What describes this person. We get a feel for his capabilities. But we do not expect him to actually know all the art in the field. That would make him a fricken expert or better.


              A person of ordinary skill in any art cannot know all of the relevant art.

              Except that is the opposite of how person of ordinary skill in the art is defined.


              Well yes Ned, when it comes to obviousness, we are taliking about “differences”, a “gap” if you like, between what was known and what is claimed. Is the step over the gap inventive (PHOSITA doesn’t bridge it) or is it an obvious step to take. The one who does or does not make the step? The one of ordinary skill in the art. Not only in the USA but also in Europe.

              Here I doubt anybody is confused. Not even anon, it seems.

      2. 10.2.2


        The timing aspect is identical as to 103 as it is to 102.

        The different application appears fully grounded in misapplying the inventor as PHOSITA – hey, if foreign sovereign’s want to do this, that’s their right and I won’t knock it. But let’s not pretend that their law is our law.

    3. 10.3

      This seems to me like one of the best points made on this thread so far. Obviousness is evaluated in terms of the PHOSITA, and the PHOSITA is not a real person. Who cares that no real person can know of these various sources of secret prior art at the time of filing? The PHOSITA does know of them, because the PHOSITA (unlike any real person) knows everything in the state of the art. As a matter of public policy, why is that a bad approach to take? It seems reasonable to me.

      If the point is that this approach makes it hard to get a patent–well yeah. That is a feature, not a bug. Our common law has a presumption against monopolies.

      1. 10.3.1

        In case it is not clear, the “this” in “this seems to me…” refers to anon’s original post #10. I just realized that my response occurs so far away from the original that the antecedent of “this” might not be clear. Sorry for that.


          Well yes Greg, given. The PHOSITA is not a real person so you can ascribe to the PHOSITA any attributes you like. Of course. True. Go ahead. But is it wise?

          Why ascribe to the PHOSITA nonsense unreal attributes, like being able to read written stuff that is private and confidential, stuff deliberately shielded from the public, like applications just filed at the USPTO which the filer can pull before publication, stuff which the filer has filed as a precaution, while reserving the right to pull it before the USPTO publishes it. What’s the point?

          Would you ascribe to the PHOSITA the ability to read stuff in sealed envelopes which the sender privately mails to herself as proof of conception. For goodness’ sake, why? This is perverse nonsense, which destroys trust in the patent system, and renders it absurd.

          Your turn now.


            Obviousness must be evaluated in view of the state of the art. The reason why you might ascribe to the PHOSITA an awareness of secret to which real investigators do not have access is because these secrets are part of the state of the art. Given our common-law presumption against monopolies, it is not crazy to require that the bar for patentability be “you have to be non-obvious over even that stuff that other people working in your field would know, but which no one has yet told you.”


              “…stuff that other people working in your field would know..”

              Yes. Of course. But how about other inventors? When you write “other people” you mean “other people of ordinary skill in the art” right?

              The content of envelopes sealed by inventive people is irrelevant and misleading, to the question what the notional uninventive person of ordinary skill in the art knows.

              Or do you still dispute that proposition?


                Sound like you are OVER emphasizing this notion of “uninventive” and misapplying the “content sealed” portions there MaxDrei.

                Are you really sure you want to say that inventions (you know, by inventive people) are outside of the bounds of what the non-real person knows?

                Is that your proposition?


                I guess it is not clear to me why the content of these sealed envelopes ought not to be imputed to the PHOSITA. The PHOSITA has no creativity, but s/he knows a lot, well more than any real person knows. Why do you think that the contents of the sealed envelopes ought not to be imputed to the PHOSITA?

                That is the point I really want to understand, and no one has yet explained to me. Instead I get things like “this is not how the rest of the world does it,” or “it is really hard for the Examiner’s to search this art,” and a million and one other reasons that do not strike me as very good reasons for why the US law should be one way or another.

                1. Well I can’t speak for the USA but under the EPC the obviousness issue is explored by mapping the person of ordinary skill onto the “state of the art” (everything Google can find, if you like) and asking a TSM question: Does the skilled find in the art a hint or suggestion to put together the combination of technical features in the claim.

                  As I keep saying, this is a “real life” test question. One which every stakeholder in the patent system can relate to. When applications are refused as obvious, the judgement is manifestly fair, acceptable even to the disappointed inventor. That’s important, for nurturing confidence in the patent sytem, for widespread public acceptance that patents are actually, a GOOD THING.

                  Your turn.

                2. A couple of thoughts in response:

                  (1) It is not my experience that inventors regard it as fair that their application is denied for obviousness, even when secret prior art is not involved. If the argument here is supposed to be based on what happens in “real life,” then I am finding the factual premises on which the conclusion is based somewhat implausible.

                  (2) I suppose that if a really large percentage of applications were denied on the basis of secret prior art, then I could see the sense of your point about this eroding confidence in the patent system. If the large corporations that fund so much of our more commercial R&D said to themselves

                  What’s the point? We do all this research to develop a new product, and then we search the art to make sure it is patentable, and after all that we file and we get shot down on the basis of a document that passed through the mail in a sealed envelope one week before our filing. Why do we even bother with filing these stupid applications just to be shot down on the basis of art about which we could never have known? Mothball our research labs and give all the R&D budget to our marketing dept. From here on out we are just going to sell the same products we have always sold and count on our advertising instead of our innovation to drive customer demand.

                  that would be a bad thing.

                  On the other hand, if denials based on secret prior art are relatively rare (and as an empirical matter, they are), then this is not such a huge deal. The net effect is that the public domain is slightly larger than it would be otherwise. The risk of being denied (or later invalidated) on the basis of secret prior art is small enough that it is still worth a corporation’s while to invest in R&D, even though it will turn out that some of those investments will be spoiled by secret prior art.

                  This seems to me like a socially better outcome than the one in which there is no secret prior art.

                3. On a tangent but I wonder if tax laws are “nurtured” for acceptance by the gen er al public

                  (And I do note the assumption that MaxDrei’s view must be better in some undescribed “real life” test). Where can I obtain a copy of the results of that test?

                4. Greg thanks. I agree with you. It is my experience too, that inventors subject to the obviousness objections of the USPTO think it not fair.

                  But at the EPO, it is different. Whether an obviousness attack lies is a TSM test, and the outcome depends on whether the drafter of your patent application set forth not only the technical features of your invention but also the technical effects they bring. When an application fails for obviousness, the inventor’s ire is then more likely to be directed at the drafter than the PTO Examiner.

                  Scarey, eh?

                  But see in these columns, how much desire there is, in the USA, for a return to a TSM obviousness test.

                5. Dear Max,

                  On that point we can definitely agree. A more objective standard for assessing obviousness under U.S. law would be greatly desirable. Unfortunately, our current Supreme Court dislikes objective standards, which they disparage as “rigid.” They prefer “flexible” (translation = unpredictable) standards. More’s the pity.

                6. Greg I don’t accept your premise, that “objective” necessarily means “inflexible”. I think the EPO’s “Problem and Solution Approach” applied under its “Principle of the Free Evaluation of Evidence” is at least as “flexible” as any other way of adjudicating what’s obvious, and 100% objective.

                  SCOTUS will reject any test that is so inflexible that it is not fair. MoT for eligibility, for example.

                  But it is not the job of SCOTUS to come up with an obviousness test that works. That burden falls on the Federal Circuit, and so far (in the opinion of the Supreme Court) it has failed to discharge its burden.

  6. 9

    USPTO may “suspend” examination of later filed application Z, until earlier application X becomes public and can be cited in Office Action for Z?

    Case A: Application X direct 102 for one or more claim(s) in later filed application Z.
    Case B: Application X needed for “solid” 103 for one or more claim(s) in later filed application Z.
    Case C: Application X needed for “arguable/weaker” 103 for one or more claim(s) in later filed application Z.

    Note: Application Z may have “other” allowable claims even when Application X is considered.

      1. 9.1.1

        I don’t understand the question. Only when a patent application is A-published can it join the database which is accessed to do the PTO patentability search.

        At the EPO, they wait until 18 months has elapsed, in order then to do a “top-up” search, just to make sure they are not on the point of issuing something that has no novelty over some other bod’s application with an earlier filing date.

        But if you are going to allow applicants to suspend A publication indefinitely well then, given your precious “presumption of validity” for “duly issued” claims, you shouldn’t ever isssue any patent ever. Should you?

      2. 9.1.2

        Yes, Examiners can search unpublished applications. They show up in EAST with clear markings all over them saying that they are not published and not for citing.


          Close – but no cigar.

          While it is true that a limitation exists prior to publishing (as part of the original Quid Pro Quo), there are proper mechanisms for the examiner to not blindly allow something that has even “submarine” prior art.

          Do examiners really not know this?


          How does one reconcile the comment at 9.1.2 with the one at 5.1.2? Are unpublished utility patent applications searchable by USPTO Examiners today, or are they not?


            They are as far as I know – I was hoping for more on this idea that the Office has unilaterally stopped looking at this (US) source of prior art.

    1. 9.2

      I have had cases suspended.

      In one case, the examiner allowed the other application to issue where it turned out that they were disclosing and claiming the same invention.

      Interference baby!

      And, we lost — on the nefarious best mode grounds.

    2. 9.3

      Note: This can happen even when earlier app X is published at 18 months.

      Example: App X will publish in 18 months. Later filed App Z is Track 1 (prioritized examination) hoping for issue in 6 to 12 months. App Z may be “suspended” until app X publishes and can be cited against Z.

      Potential Track 1 quality issue if examiners are not diligent in applying the non-public apps database.

  7. 8

    Maybe practice should be changed so that six months (for example) after issuance of first action on merits an application for which non-publication was requested will be published.

    Applicant can then decide when he receives action whether or not to quickly abandon before publication to maintain secrecy

  8. 7


    You may also want to consider several “request for non-publication” scenarios.
    Where many years latter, an application may issue (become public) or not.

  9. 6

    Let me start by saying I thing pre-grant publication is a terrible idea. The inventor is forced to forfeit his potential trade secret in hopes that the Patent Office will look favorably on it in the patent examination process. The notion of quid pro quo is lost.

    That being said, if publication is such a super-fabulous policy that we should force such a deal on inventors, why wait 18 months? Why not publish ASAP after filing?

    1. 6.1

      > The inventor is forced to forfeit his potential trade secret in hopes that the Patent Office will look favorably on it in the patent examination process.

      I wouldn’t say “favorably,” but rather “fairly.”

      However, in general, I agree: applicants’ reluctance to agree to pre-grant publication is a sign of lack of trust that the process will recognize and reward their efforts as it should. The notion of disclosing key research that provides a commercial advantage, only to get it bounced out of the patent system unprotected because the examiner thought it was “basically” something trite, is a serious chilling effect on the patent system.

      This is the baggage of the Dudas “patent rejection office” era, and the unrecognized consequence of insubstantial standards of patent eligibility, and the expansion of discretion in patent examination. Unfortunately, in our current anti-patentee climate, these consequences cannot be duly appreciated; things will continue to get worse before they get better.

      1. 6.1.1

        anti-patentee climate

        This is the “anti-patentee” climate to which David Stein refers:

        link to patentlyo.com

        the USPTO has already issued more patents in calendar year 2014 than in any other prior year. I expect that the total will be around 304,000 utility patents issued this year. The number of grants has more than doubled since 2005.

        Now try to imagine the ridiculous farce we’d be staring at if we lived in the “patentee friendly” world of David Stein’s dreams.

        Nobody is forcing either of you to file anything. There are plenty of worthy innovators out there happy to take advantage of a patent system that isn’t polluted with the junk that you two love.

      2. 6.1.2

        Once published, why bother hurrying or even giving a dutiful examination?

        What is the applicant going to do? Walk away?

    2. 6.2

      SVG: I thing pre-grant publication is a terrible idea.

      Please show us the applications you’re filing and we’ll talk about “terrible ideas.”

      1. 6.2.1

        What is the reasoning behind such negative comments?

        I’m dealing with a close family member who has severe depression. Your comments make me think you have something similar.

        Do you sleep well at night? Do you think “everyone is out to get you”? Do you isolate yourself from all of your friends and family? Do you think of suicide?

        If you have any of these symptoms (and there are many more), I implore you to see a doctor.


          What is the reasoning behind such negative comments?

          The “reasoning” is my 100% certainty that if submarine patenting was still permitted in the US that “silicon valley girl” would be filing as many of them as possible with the dream of popping up at just the right moment to extort a deep-pocketed company with said submarine patent.

          And there’s lots of other people just like “silicon valley girl” who lament all these (trying not to laugh) “anti-patentee” changes to the patent system that nearly everybody on earth understands were absolutely essential to maintaining some semblance of sanity to the system. But here’s the thing about these complainers: they are invariably quite well off and, in most cases, vastly better off than nearly everybody else in the country. Yet they never stop complaining about how they can’t get more patents more quickly and how nobody understands how important their “innovations” out there.

          Let’s see the claims that svg is filing and then we’ll see which idea is more “terrible”: the idea of publication before grant, or the idea of granting patents on functionally descriptions of information processing logic.


            Malcolm, submarine patents a la Lemelson were disincented by the patent term of 20 years from filing. No need to add pre-grant trade secret forfeiture to the disincentive.


              No need to add pre-grant trade secret forfeiture to the disincentive.

              Because we don’t have enough patents?

              Seriously, if you have a worthwhile trade secret, it’s silly to disclose it in a patent application.

              If you think you deserve a patent on some innovation, then do your homework, hire a decent lawyer to double-check your work, do a search, and write a decent application that isn’t going to go up in smoke two seconds after somebody with half a brain reads your claims.

              If you don’t like your chances, then don’t play the game. Most likely you don’t deserve a patent and the PTO doesn’t need to waste its resources listening to you whine.


                Nice strawman – of course, the actual point has nothing to do with your “patents are bad” mantra.



          Malcolm suffers from a severe case of cognitive dissonance – as supposedly he is a patent prosecuting attorney who job it is to obtain patent rights, something he clearly disdains.

          I have suggested that he seek employment in a field of which he can believe in the work product produced.

    3. 6.3

      Another comment that reveals how dysfunctional US patent law can be.

      The restraint on trade that patent rights deliver is defensible only with its promotion of progress in the useful arts, brought about by the early publication of enabling disclosures of new and non-obvous solutions to technical problems. 18 month A publication puts competitors on notice and scares them off “your” patch. You see an A publication. You see issues. You trade or design around. Progress is promoted.

      In a dysfunctional jurisdiction nobody watches the stream of emerging A publications because i) there is so much rubbish in them and ii) if you do you might get done for wilful infringement. No watching, no promotion of the progress.

      Further, there is complete uncertainty, whether your subject matter, if filed, will be deemed eligible, and how it will be decided whether it is obvious. So you don’t know whether or not to keep it a trade secret.

      Instead, ludicrous transaction costs, submarine patents, ambushes, hijacks of an entire industry, Gilbert Hyatt,the intermittent windscreen wiper guy and assorted other chancers, investors, banksters, litigators and speculators.

      1. 6.3.1

        You make completely unwarranted assumption that the “type A” publication will have any relation to what may eventually yield as a patent.

        As I mention above, there is NO impetus to move once the Quo of publication has been taken – the Office has at that point obtained the maximum that it can obtain, so what is the driver for giving the applicants their Quid?


          (Add in the “fruits” of the AIA: longer wait, more uncertainty, more difficulty to enforce, and the “deal” has lost lots of it’s worth – exactly what the sponsors of the AIA wanted)

    4. 6.4

      Silicon, we talked about that at the time — that we still needed some means to tell whether the scope of allowable claims justified the publication. We wanted at least a FAOM before publication.

      Note, the ROW has a search conducted before publication. It would be good for the US to adopt the same approach.

      1. 6.4.1

        ROW has search before publication? Really? Do you know how many JP and KR publications I’ve seen where they were still waiting for a request to start examination?


          anoy, yeah — deferred examination. Another innovation that we opposed but which now seems like a very good idea to save expensive examination resources.


            How do you square the non-publication option with deferred examination? Or would you rip that (true) reflection of the Quid Pro Quo away?


              anon, good point.

              NP would be an option in exchange for immediate, and expedited (Track I) examination only.

                1. The 18 months is not (necessarily) arbitrary – and anyone looking into the history would readily recongize where it comes from – and its relation to an attempt not to wreck the Quid Pro Quo while still reflecting some Big Corp multi-national concerns.

                2. Silly Cone, anon is right. The idea is to conduct a search before publication for two reasons. One, is to publish with the search report. The second is to allow the applicant to withdraw the application and not allow publication.

                  Now, if publication takes place 18 months after priority to, for example, a provisional that is not published, that gives only 6 months from filing to get everything done. Cut it down by even a month or two, and there would be insufficient time.

  10. 5

    Most notably from the face of the statute is that the old law only excused prior art for obviousness purposes but not for novelty purposes while the new law negates the references for seemingly any prior art purpose.

    This isn’t much of a substantive change. The number of cases where 1) the reference wasn’t old enough to publish to avoid the rule and 2) the Applicant couldn’t have claimed priority anyway and 3) the applicant admitting that an anticipatory reference was co-owned doesn’t generate a double patenting rejection, are small.

    For examiners, do you consider whether an applicant can claim the exception before issuing a rejection or do you wait for the applicant to claim the exception.

    You always assume that any claim is valid, because them perfecting their claim is grounds for a second non-final. If it’s a short document (one time the provisional was just some power point slides) one would be inclined to decide the issue before it comes up, but rarely do most examiners do that.

    Questions that I would like to pursue is how important this expanded submarine prior art is to the system and likewise the relative importance of the intra-company-exception.

    If you’ve noticed the outsized allowance rate considering the increasing number of filings (and therefore the increased number of submarines) plus the AIA increasing the extent of said submarining, you’d see that it makes no difference at all. When the number of filings jumps up and the allowance rate increases, it’s pretty clear evidence the system can’t handle the search at all.

    And the system, in fact, cannot handle the search at all. If you filed a PPH case today and it was assigned to me tomorrow, not only can I not pull up my own case in a standard EAST search, I couldn’t pull up most of the filings in 2014 or 2015, all of which is (or will be) prior art.

    Even if, for example, two cases were to come to me directly both covering the same subject matter, there is no policy I know of by which I can delay the later case from a docket clock despite the fact I know of an anticipatory reference with no non-publication order which will, by the normal course of time, become prior art. If the latter case comes up on my docket before the publication time runs, it would have to be allowed. I would have to essentially ask my SPE to break the rules so that I could properly decide the case.

    1. 5.1

      “And the system, in fact, cannot handle the search at all. If you filed a PPH case today and it was assigned to me tomorrow, not only can I not pull up my own case in a standard EAST search, I couldn’t pull up most of the filings in 2014 or 2015, all of which is (or will be) prior art.”

      The PTO claims that this art is searchable and findable by the examiners.

      1. 5.1.2

        The USPTO let the UPAD (unpublished patent application database) die around 2010. Nothing filed since then is searchable internally until the 18 month publication date.


          While I definitely appreciate your posts PS DIP, do you have anything in the public record to confirm this?



            There is absolutely nothing on the record about any internal databases at the USPTO. It would be too embarrassing if there was…

            I’ve just verified with folks working at the PTO and at the contractor that used to handle the contract: Nothing since 2009 has been scanned and entered into the examiner searchable database of unpublished applications known as UPAD.


              Thanks PS DIP, do those folks recall the “stop order?” Was it in writing? It is difficult to fathom the process stopping “on a whim,” as I would think somebody would say, “Hey, did the legal status of this stuff stop being prior art?”

    2. 5.2

      I think you would have to go to your SPE and get him to get a Quas to put the case in the program formally known as (but which still exists in various forms) as Saws.


          I’m literally looking at the TC Saws training right now. Submarine type applications are bullet 17 of 20.


              Applications with long pendency which have not published. That’s what the bullet says at least.

  11. 4

    many follow the European approach that the submarine art is not considered for obviousness (inventive step) purposes because of the reality that the secret prior art would not have been within the grasp of a person of skill in the art (since it was secret at the time)

    There are good arguments for this rule but those arguments seem to me to be diminished in circumstances (unique to the US) where applicants are filing their applications with non-publication requests and where those applications may remain unpublished for several years or longer.

    1. 4.2

      MM, it would seem more logical to have patents have no prior art effect from their priority date if the applicant chooses not to publish.

      The prior art effect is a benefit to the patent applicant. But that benefit should be forfeit if he chooses non publication.

  12. 3

    The term “submarine prior art” sounds like it’s being used pejoratively here (in the same vein as “submarine patent”), which is odd considering that 102(a)(2) is the keystone of first-to-file. It’s right there in the name. If someone files before you, then their filing is prior art against your later filing. The only reason it has this so-called “submarine” quality is to protect filers from having their inventions published prematurely, which I’m sure most filers would consider to be a worthwhile tradeoff.

    1. 3.1

      The term *should* be used pejoratively. See MaxDrei’s comments 2 and 1.1 below. We’re no longer in the first to invent world, we’re in the first to file world. In that world, you need novelty to ensure that two people don’t get claims on the same thing, but obviousness should be judged on the basis of what was publicly known at the time you filed.

      1. 3.1.1

        Your rationalization, if followed to the letter, allows situations where the policy’s objectives are not met. For instance:

        Inventor A files an application on day 1 disclosing a highly complicated machine. Inventor B files an application on day 2 claiming that same highly complicated machine in every detail, such that Inventor A’s application anticipates it, but adds a dependent claim reciting a “tree-shaped air freshener”.

        Under your rationalization, that dependent claim would be allowable because Inventor A’s prior disclosure should not count as prior art in an obviousness rejection, even though Inventor A actually invented the real invention first.

        It’s a safer alternative for the public to treat such references as valid prior art for both anticipation and obviousness.


          (Edit: should have read “Inventor A actually filed the real invention first”)


            Can you tell us (because I’m not sure) whether the tree-shaped air freshener as such is old. I suspect you cite it as the epitome of old-ness, but you don’t say. If so, who claims it anyway. And if they do, where’s the harm in letting them take it to issue. They still can’t practise it without a licence for the earlier filer. And if they claim “Any car with a tree-shaped widget” ie a claim not limited to the complex new machine of the claim filed the day before, that wider claim is hopeless anyway.

            When it came out, I dubbed the AIA Winner Take All First to file because if you are first to file, your filing destroys the patentability of everything filed as from the next day, that is obvious relative to your filing. And unlike other jurisdictions, this prior art effect is unrestricted as to where the prior art was filed. A WO that doesn’t extend to the USA is still prior art for obviousness in the USPTO. This is bonkers.

            This Winner take All mentality is a peculiarly American thing. It suits Big Corp, to stamp on eferything else filed, as from the next day.

            But in the real world, people don’t claim uber-old tree-shaped air fresheners. Rather, when for an invention its time has come, filers file, all over the world, on much the same subject matter, at much the same time. A rule is needed, to divvy up patent rights amongst them. The ROW thinks it fair to let every filer have what she enables in her patent application and written description, minus that which just isn’t new over the content of earlier filings in that specific jurisdiction. Mark my words, the USA will come to regret this turbo-FtF, in disharmony with the rest of the world.


              My example was made somewhat in jest, in part because I didn’t want to take the time to think of a better example. The tree-shaped air fresheners I was talking about are these. I guess these aren’t popular in Europe? Not that they’re necessarily popular in the US, either, but they’re enough of a cultural icon here that most Americans probably know what they are.

              But it seems my example may have been misleading, because the example item that I used was irrelevant to the invention. A better example would have been some detail that is almost always found along with the true invention, but that is well-known, not critical or inherent to its functionality, and likely to be omitted by someone else in their disclosure.

              I guess a slightly better example would have been a claim to a braking system that bothers to mention that the brake rotors are each connected to a wheel and tire, in a case where the earlier filed invention didn’t disclose any specific use for the braking system.

              The problem isn’t that Inventor B would have to pay a royalty to also make the patented invention. The problem is this: In the example, Inventor A licenses their braking system to a manufacturer who plans to use it in a car. Inventor B is then able to extract a royalty from that manufacturer for the exact same invention, simply because they had the foresight to claim a specific embodiment where the braking system was used in a conventionally well-known setting (the car). So there definitely is harm in disallowing Inventor A’s disclosure from being used as prior art in an obviousness rejection.


                “Inventor B is then able to extract a royalty from that manufacturer for the exact same invention, simply because they had the foresight to claim a specific embodiment where the braking system was used in a conventionally well-known setting (the car).”

                How can it be foresight when Inventor B had no knowledge of Inventor A’s invention (as Inventor A’s app wasn’t published at the time)?

                “It’s right there in the name. If someone files before you, then their filing is prior art against your later filing.”

                True. But the question is: should it be? It’s not in, for example, Europe where such “secret prior art” can only be used novelty rejections, but not inventive step rejections. I haven’t heard of any dire consequences for the public from that regime. Maybe MaxDrei can provide some insight, but I’m not aware of any real world instances of the hypothetical you’ve expressed concern about.

                1. Under the EPC, the “disclosure” of a document, that which takes away novelty of later filings, includes that which is “implicit” in its teaching. There is a rich line of EPO caselaw on the notion that that which the reader of the doc “seriously contemplates” is within the disclosure of that document.

                  So for a “hard case” under Art 54(3) EPC you need something lacking from the disclosure that is also not implicit and yet is something which anybody wanting to implement the teaching has to use.

                  Consider now the well-established black letter English law on novelty under the EPC. What a doc discloses includes anything which is the inevitable result of putting the enabling disclosure of the doc into effect.

                  As AAA surmises, we haven’t (as far as I know) had any such imagined dire consequences yet. Since 1978. But hey, who knows. Anything’s possible.

                2. How can it be foresight when Inventor B had no knowledge of Inventor A’s invention (as Inventor A’s app wasn’t published at the time)?

                  That’s exactly why it is foresight. If he knew what was in Inventor A’s application, then (a) it would be hindsight and (b) it would trigger one of the 102(b)(2) exceptions.

                  In my example, I envision it as being Inventor B’s attorney’s standard operating procedure to throw in a bunch of those sorts of dependent claims (adding in well-known limitations that are frequently used but not always disclosed) just in case there’s an anticipatory reference filed within the grace period. In other words, this hinges more on the attorney’s draftsmanship rather than the inventor actually being the first to file.

                3. “That’s exactly why it is foresight.”

                  I still don’t see it as foresight. It’s part of Inventor B’s “invention” or “inventions” or whatever you want to call them. If Inventor A didn’t disclose the applicability of his/her braking system to a car, then Inventor A didn’t “invent” that. There’s no reason to deny Inventor B a claim to that. Part of the “invention” is a consideration of the uses and applicability of the “invention,” even to “a conventionally well-known setting.” The notion, particularly amongst the examining corps, that all applications and uses of an “invention” are de facto obvious is just wrong. Every invention, and every application and use of it, are “obvious” once somebody discloses it/them.


              “Mark my words, the USA will come to regret this turbo-FtF, in disharmony with the rest of the world.”

              Already regretting. Have had several instances of 102(a)(2) prior art used against us that was filed less than a month before us, in once case only 6 days before us.



                Why in the world did the AIPLA back this? Except for FTF, it went against AIPLA policy against secret prior art, a policy of very long standing.


                Wait a minute, an example of you regretting the AIA rules is not the same thing as the USA regretting the AIA rules. I have no strong feelings one way or the other about secret prior art, but if I am to be convinced that the “USA” will regret our policy of allowing secret prior art, I will need to see more evidence of the alleged policy failure than just “this or that inventor is annoyed to discover the existence of art of which he was not aware at the time of filing.”

                This is as much a question for Max Drei as for AAA JJ, but I would like to hear why the USA as a whole will regret our allowance for secret prior art. The dissatisfaction of any given inventor/applicant/attorney does not count as a reason.

                1. Greg, “WE” never had a policy of using secret prior art to invalidate a patent. From 1790, known or used (publicly) was the policy.

                  The way 102 was restructured to include both known or used prior art with the defenses that the inventor was not the first and true inventor, in combination with the new 103, and in combination with Rich ruling as he did on the bench, resulted in something that no one bargained for. What happened was not intended. Had anyone any inkling that the basic policy that something had to be public to be prior art was going to be trashed by the “careful” drafting of the ’52 Act by Federico and Rich, the bar would have been in an uproar.

                  The AIPLA long made it a goal to undo the damage. That is why it is almost incomprehensible that the AIPLA, the people who brought you the AIA, would have codified what they so long opposed.

                2. “WE” never had a policy of using secret prior art to invalidate a patent. From 1790, known or used (publicly) was the policy.

                  Clearly you cannot mean that “never” honestly. We have allowed for all sorts of secret prior art under the 1952 act, and the years since that act came into effect have been some of the most scientifically and technologically glorious ever seen in human history.

                  Meanwhile, in 1790 we also had no law regarding obviousness. I would like to think that we have learned somethings since then and improved the patent laws accordingly. Certainly by the time of Milburn we allowed for secret prior art, and have been doing so ever since. It is far from clear to me that this approach has impeded innovation.

                3. Greg, trust me. The ’52 Act was not intended to change the law regarding prior art. That was the work of Judge Rich and the SC in Hazeltine v. Brenner.

                4. Dear Ned,

                  Whether or not the ’52 act was intended to allow for secret prior art is irrelevant to the point that I am making. U.S. law has allowed for secret prior art—for both novelty and obviousness purposes—for years now, and it is not clear to me that it has had any deleterious societal effects.

                  I am still waiting for someone to explain to me why the US will regret allowing for secret prior art. Hazeltine (and other such law) has not been any impediment to innovation, as near as I can tell. Why am I supposed to regard a Hazeltine system as problematic?

                5. Well, Greg, it caused the Big Boys and Congress to pass numerous exceptions to 103 prior art — common ownership, joint R&D, just to contain the damage. I don’t think everyone would agree there was no problem.

                  But now, making secret prior art available for 103 purposes keeps is out of step with the Europeans for example when the whole point of going to first-to-file was harmonization.

                6. it caused the Big Boys and Congress to pass numerous exceptions to 103 prior art — common ownership, joint R&D, just to contain the damage.

                  What is this “damage”? I am still waiting for a clear explanation here. I agree that large corporations did not like the secret prior art effect of their own patent filings, and they convinced Congress to relieve them of this annoyance, but I am still not clear as to how their annoyance was any problem to society as a whole.

                  Mind you, as an attorney, I love 103(c). It makes my job easier. But it is not clear to me that the patent laws should be drafted with my convenience (or the convenience of my clients) in mind. Patent law should be drafted to advance the public good, and I am still waiting for an explanation of why the public good is better served by excluding unpublished filings from the prior art.

                7. Well, Greg, for one, how does one properly deal with a duty of disclosure if everything a company is doing internally is prior art to everything else the company is doing?

                8. Dear Ned,

                  I keep asking “what is this ‘damage’?” and you keep offering examples that boil down to “it is annoying to inventors and their attorneys.” I am just as much in favor as you are of not annoying patent attorneys, but avoiding annoyance of attorneys and applicants cannot be a basis for the patent laws. The patent laws should be crafted to serve the public good, not our convenience. This duty-of-disclosure issue you raise is still not a real example of damage.

                9. Well, Greg, in the abstract, the difference is that the first filer gets full 103 benefit — creating a larger zone within which he might practice his exclusive rights.

                  This could actually be view as a good thing if the problem of big companies is really a non issue.

                10. trust me. The ’52 Act was not intended to change the law regarding prior art.

                  Absolutely B$.

                  I can give you at least 103 reasons.


                Wow….for real?….also, in the words of Mosslinghoff, not ftf, fitf (first inventor to file)……not defending…..
                6 days is tough.


          Inventor A files an application on day 1 disclosing a highly complicated machine. Inventor B files an application on day 2 claiming that same highly complicated machine in every detail, such that Inventor A’s application anticipates it, but adds a dependent claim reciting a “tree-shaped air freshener”. Under your rationalization, that dependent claim would be allowable

          “Clearly you have a bias against the fresh air arts.”


            Dennis: “On-point comments are very welcome.”

            Examiner Mooney: “Clearly you have a bias against the fresh air arts.”

            Heckuva job, Dennis.

      2. 3.1.2

        In that world, you need novelty to ensure that two people don’t get claims on the same thing, but obviousness should be judged on the basis of what was publicly known at the time you filed.

        Except it would be totally possible to publish everything on the day it was filed in the office. You’re using the fact that the law gave a benefit to an applicant (18months to determine whether they want their information disclosed) to suggest that we have to give another benefit to the applicant.

    2. 3.2

      APoTU, I think the issue is not 102 prior art, but 103 prior art.

      Assume double-patenting is being enforced.

      Two inventors filing on the very same invention. The first to file gets the patent — 102(a)(2).

      The second filer has variant B not anticipated by first’s disclosure. Why not give second a patent on B. Nothing is withdrawn from the public domain, and the first filer did not invent B.

  13. 2

    Can anybody give me another country besides the USA that allows submarine art to be the basis of obviousness attacks? Or is the USA out on its own here? Dennis writes that “many” countries follow the EPC and do not allow such obviousness attacks. But are there “any” exceptions? Is “many” in fact “all”?

    It reminds me of those pre-AIA days when American patent attorneys wrote that “many” other countries had a First to File patent law. In choosing not to write “all” it was just a case of attorneys being cautious I suppose.

    As to own earlier filings being Prior art, I believe Japan takes the American approach. But what does China do? Anybody know?

    I think the “own earlier filing” shield is something bulk filers want and which helps them prevail over less prolific filers. But that’s what Japan and the AIA is all about, right?

    1. 2.1

      Max – As you guessed here, I was just being cautious in my approach (for once) because I have not taken the time to inspect this particular rule in non-U.S. patent regimes (other than Europe).

    2. 2.2

      Max, Israel follows the EPO approach on this, and as far as I know, the post-AIA USA is the only country where unpublished documents can be cited as the basis for an obviousness/lack of inventive step rejection.

      Some harmonization!

      1. 2.2.1

        Dan, and we really didn’t have to do this to ourselves, did we? We could have excluded from 103, 102(a)(2) prior art, reversing even Milburn. But the attempt was not even made.


        The AIPLA has historically been opposed to secret prior art being the basis of 103, and has historically been in favor of harmonization. But, it was also the AIPLA, in the form of Bob Armitage and folks over at IBM as I understand it (Kappos), that made at least the initial drafts the AIA.

        What happened? Did the PTO take control of the drafting?

        It would be interesting for those who know the history of the drafting to tell us who did what and when.

    3. 2.3

      Yeah, Max. I explained the origin of the problem below.

      1. Federico putting defenses into 102.
      2. Federico/Rich having the new 103 refer to 102.
      3. Rich, now judge, making the whole of 102 available for 103 purposes.

      Rich — was at the center of all this. What he did was unnecessary and a mistake. Some what like the mistake he made in drafting 112(f) and 271(b) and (c) the way he did. Lots and lots of unforeseen collateral damage.

      1. 2.3.1

        Agree with you that Judge Rich got this one wrong when he expanded “secret prior art” to the obviousness analysis. IIRC his rationale was “If the patent had issued on the day it was filed, it would be public”. That’s like saying if pigs had wings they could fly. Never thought its application in the obviousness context made sense, even pre-AIA, and even less so post-AIA.

  14. 1

    1. The expansion of submarine prior art to foreign language applications is going to be a problem. The PTO needs to require translations.

    2. The use of secret prior art for obviousness purposes is indeed unfortunate. We should adopt the European approach.

    1. 1.1

      Ned, I should appreciate your thoughts on something.

      The EPC approach sees novelty and obviousnes as two distinct concepts with different but complementary roles in any First to File system. You take the date of the claim and examine obviousness relative to everything the PHOSITA had available, one day before that. The entire prior art universe, as it was, the day before the date of the claim

      So why bother with novelty at all? Who needs it?

      It is needed to adjudicate between rival filers, none of who tells the public their invention before they file at the PTO. Which of them gets a patent? What is fair, reasonable, workable and pragmatic? Shall it be all of them or none of them?

      Answer: the first to file gets all she asks for, all that is new, enabled and not obvious.

      The later filers get the same, except that they are cut back to that which is novel over the earlier filers.

      Why should bulk filers be spared the prior art effect of their own earlier filings? It makes them sloppy. It makes no sense.

      Or does it make sense to you Ned. In a First to File system. If so,why?

      1. 1.1.1

        After the ’52 Act, a lot of what had been defenses to infringement, that the inventor was not the first inventor, was put into 102 as 102(f) and (g).

        As well, 102(e) codified Alexander Milburn link to scholar.google.com

        What many had not expected was that the CCPA/Federal Circuit would count prior invention (f) and (g) as prior art for 103 purposes. See, e.g., Application of Bass, 474 F.2d 1276, 177 U.S.P.Q. 178 (C.C.P.A. 1973)(prior invention); OddzOn Products, Inc. v. Just Toys, Inc., 122 F.3d 1396 (Fed. Cir. 1997)(102(f) is 103 prior art).

        Even though Milburn was premised on the theoretical ability of an applicant to have his patent issued on the day he filed it, the courts extended the prior art effect of patents back to 120 benefit cases filed in United States. In other words, by placing into 102 the codification of Milburn, and the defenses of prior invention, all this became 103 prior art as of the earliest date of invention or filing, and nobody expected that to be the case. Nobody was pleased. This was not expected. The courts got it wrong — but who is to question Judge Rich?

        Big business got together to see how they can limit the collateral damage caused by the new 103 and 102. One of the things they did was remove from 103 prior art art that is 102(e), (f) artor (g), but only if the inventions were subject to assignment to common owners. Obviously, the legislation was proposed by big business to protect themselves only.

        So long comes the new AIA, and big business is still trying to protect itself by excluding as prior art commonly owned patent applications. Obviously this will permit a lot of double patenting, but the expectation is that double patenting will be enforced by the courts.

        As you might be aware, double patenting applies to common inventors/owners under circumstances where two different patents are claiming the same invention or obvious variations and where the two different patents are not prior art to each other. So, in point of fact, the movers and shakers are simply wasting everyone’s time by moving the consideration of commonly owned applications from 102/103 to double patenting. The overall result is the same.


          Check that, with double patenting, one can get obvious variation issued using a terminal disclaimer. But, given that both patents must expire at the same time, what is the point?

          In my experience as well, courts will not permit a patent owner to assert multiple claims of essential the same scope. I assume also that even if one prove infringement of claims from two different patents subject to a terminal disclaimer that one can get only one royalty.

    2. 1.2

      “1. The expansion of submarine prior art to foreign language applications is going to be a problem. The PTO needs to require translations.”

      Not going to happen. Much easier for the PTO to presume the U.S. version is entitled to the foreign filing date and then force applicants to bear the expense of establishing otherwise.

      “2. The use of secret prior art for obviousness purposes is indeed unfortunate. We should adopt the European approach.”

      Hal Wegner beat that drum endlessly before passage of the AIA and didn’t get any traction.

      Of course the use of secret prior art in the determination of obviousness is absurd, but that’s where we are. Maybe one could successfully argue against a secret prior art reference used in an obviousness rejection on the grounds that the subject matter disclosed in the secret prior art reference was not within the scope and content of the prior art at the time the claimed invention was made, or before the effective filing date of the application. I’ve never seen that argued, and would be interested to know if it’s ever been argued and what the outcome was.

Comments are closed.