Federal Circuit Looks for Briefing on Automatic Assignment En Banc Challenge

Following up on the Shukh v. Seagate petition for en banc rehearing, the Federal Circuit has now taken an important first step of asking Seagate for its response to the petition, due by November 13, 2015.

The issue here is the construction of an employment agreement where the patentee “hereby assigns” all future inventions (created within the scope of the employment).  The Federal Circuit rule is that the agreement serves as an effective property right transfer such that, at the moment* of invention, rights to the invention automatically transfer to the employer.

One reason why the issues in the case are interesting to me is because the current Federal Circuit rule is contrary to the traditional property law notion expounded in the UCC that  “a purported present sale of future goods or of any interest therein operates as a contract to sell.”  Although relatively new, these same limits on property transfer stretch back hundreds of years in our common law history. In the background of the case is the notion that the Federal Circuit has created a federal law of assignment agreements that seemingly operates worldwide — again going against the norm that in the patents are to be treated as personal property in the exchange context, and personal property exchanges are normally governed by local state or foreign law.

In the patent context the question the Federal Circuit’s rule serves as an important and additional thumb on the side of employers rather than their employee inventors. In addition questions of legal tradition noted above, the the case raises important innovation policy questions and questions of inventor/employee rights.  The issues here can be thought of as in parallel to other areas where the law places limits on how far employment contracts can go to strip an employee of her rights. Consider, for example, covenants-not-to-compete, inventions outside of the scope of employment, and post-employment secrecy requirements limiting what someone can do with learned skills and knowledge.  In those contexts, the courts (and legislatures) have found that it doesn’t make sense to enforce overreaching employment contracts except when very particularly bargained for.

Employers obviously prefer the current rule because it so strongly favors them.  Thus, I expect that any IPO/AIPLA (and probably PTO) commentary on this front will support the current automatic-assignment rule.


* I included the asterisk next to “moment” of invention because of the reality that invention is typically a process distributed over time.  As the courts have repeatedly written, invention begins with a full and complete conception of the invention remains inchoate until the invention is reduced to practice (either actually or constructively). Updated for spelling.


112 thoughts on “Federal Circuit Looks for Briefing on Automatic Assignment En Banc Challenge

  1. 11

    What legal right or rights does an inventor have in an invention:

    1. Without any patent statutes; or
    2. With patent statutes that require, among other formalities, that an inventor apply for a patent to obtain a patent?

    I would argue that an inventor has no legal rights in an invention absent patent statutes and has nothing to assign.

    Thus the only right an inventor has is the statutory right to apply for and, potentially, obtain a patent on an invention.

    Now assume that the right to apply for a patent by the inventor cannot be assigned. The only thing that can be assigned is the patent that is obtained upon application of the inventor.

    Now consider 35 USC 152. It requires a recorded assignment from the inventor for a patent to issue to an assignee. Assignment of what? Of the patent that will issue on a specific application.

    Does anyone disagree with this so far?

    So when one presently assigns a future invention, what does that mean in fact? In fact, it means nothing. The only right the inventor has is to apply for and potentially obtain a patent. But the right to apply cannot be assigned; and the right for the assignee to obtain title to a patent on issue is conditioned on their recording an assignment from the inventor identifying the specific patent application.

    This is not to say that the right to obtain patents on application may not be assigned without recording. But it does suggest that any assignment of title must identify a specific application or patent.

    1. 11.1

      You lean too hard on something too simple.

      As mentioned below, “choices of action” are alienable property.

  2. 10

    Thanks to Troubled at 8, for raising a good point.

    The 102 and 103 prior art effects of an earlier filing A at the USPTO on a later filing B can depend on who was it that filed A. To me it makes good sense to put the burden on B to show that A is NOT prior art effective against B.

    But I think the public needs to know whether A has a prior art effect on B, so it can come to a view on the validity of B. The reality is that B will likely get through to issue before A has been applied against B and before B has been asked to discharge its burden. Is the issued B in fact invalid? How can the public know?

    Now B of course is in a position to know whether A is in common ownership (if there was any “obligation to assign). But how about the public? In the real world, how does it find out whether A is or is not prior art to B?

    Or do such issues not matter enough to worry anybody but me?

    I ask because under the EPC the prior art effect of A on B is present regardless whether or not there is any common ownership. The founding fathers of the EPC were alive to the difficulty when they drafted the EPC in 1973 and wished to minimise legal uncertainty. Either A is earlier or it isn’t, Y/N, and ownership is irrelevant. Kudos to them!

    Asia of course follows the USA, in making common ownership a release from the prior art effect. Hence, those outside Europe who apply for a patent inside Europe are frequently surprised and dismayed when they find out that common ownership is not effective as a way to dissolve away the prior art attack.

    1. 10.1

      The Minutes of the Tegernsee Group on this issue (prior art effect of earlier filings) are revealing. Those speaking for Japan explained why common ownership is a necessary excuse. You see, when Big Corp files a heavy stream of patent applications, it is simply too burdensome and onerous to screen each one to see whether it, Big Corp, has already filed something on an earlier date that is prejudicial to the novelty of its later filing.

      So when it comes to a balance between the public interest and that of Big Corp, which shall prevail? Is it necessary to point out to you that the AIA, in confirming that earlier filings, anywhere in the world, are good for obviousness attacks, but excusing own earlier filings, simply tilts the balance even more in favour of Big Corp. It ain’t right, is it?

    2. 10.2

      Max, I fully agree with your argument here that excusing commonly owned inventions as prior art is inexcusable from a public policy point of view. So much for that “transparency” that was shoveled in our face for so long by those who sought to impose upon us the so-called transparency of the AIA.

      We need to revise the AIA in a number of ways. One of these ways is to get rid of patents and patent applications is prior art as of their filing dates. This will almost automatically solve the problem of common ownership.

      1. 10.2.1


        I am not sure that I see the “solution” in your answer.

        Are you also including an elimination of the option of non-publication? Are you inserting an “immediate” publication requirement as well?

        Lastly, are you conflating what the legal person of PHOSITA is with actual real people in your “solution?” By this I mean, that the timing aspects reflect what the more-powerful-than-any real-person legal person can have in their repertoire and that the 102/103 effects of the legal person are meant to be more reaching than any actual real person.

      2. 10.2.2

        Eliminating any prior effect of application filing dates would introduce frequent cases of two different companies claiming ownership of the same invention, since often companies are doing R&D along the same lines. Do you really think many folks with votes [i.e., other than commentators on this blog, who have no influence] really want to scrap the AIA and go back to Byzantine interference litigations, with surprise alleged prior secret invention date invention ownership claims by others after one has made big product commitments, plus alleged secret 102(g) prior art litigation defenses against one’s patents?


          Paul, going back to interferences is not what I suggested, not at all.

          The first to file get the patent in the case when two patents or applications are claiming the same invention.


          Once again, the posts by Paul lack any touch with reality.

          a) Do you not recognize that what you call “Byzantine interference litigations” was ADOPTED fully into the AIA? You have drunk the “Kool-aid” of “ooh, that’s too difficult, let’s change ” – But not really change.

          b) Your “really want to scrap” runs the fallacy as if our “influence” here makes following the law some type of “gee, let’s VOTE on this” when it comes to whether or not Congress has explicitly made the AIA into a ALL OR NOTHING sink or swim massive too-big-to-fail-but-Ooops-WILL-fail piece of legislation by expressly NOT including the separability amendment.

          I am STILL waiting for you Paul to man up and advance an actual legal argument against my position. And yes, I just won a large wager in Vegas that you would post again without doing so – Thank$.

    3. 10.3

      Max, under all new [AIA] U.S. applications, once earlier filed application A is published at 18 months after it is filed [as something like 90% are?] A will become prior art against any B application filed after that publication date. That considerably reduces the number potentially of less clear situations.
      [As previously noted, the EPO “novelty only” prior art situation for prior UNpublished applications as of filing dates leads to LOTS of problems.]

      1. 10.3.1

        A will become prior art against any B application filed after that publication date.

        Actually, A is prior art against any B applications filed after A’s filing date. This will be different than “that publication date

        Paul, does this change the calculus with your view of “[As previously noted, the EPO “novelty only” prior art situation for prior UNpublished applications as of filing dates leads to LOTS of problems.]”…?

        I note (with some small sense of irony) that the fact that the equivalent of our 102 and the equivalent of our 103 is differently treated under the EPO for items EQUALLY not available (the UNpublished set).

        And yet, here in the States, the rationale for equal treatment – having to do with the legal person of PHOSITA, and contrasting with any “real person” of an inventor – seems “under appreciated”…


          RE: “Actually, A is prior art against any B applications filed after A’s filing date.”

          Not if they were commonly owned as of their filing date and B is filed before A is publishished, which was the subject I was responding to.

          “Novelty only” systems as in the EPO if literally applied allow any other company to claim obvious slightly different aspects or applications of the same invention in B applications if they are filed before A is published, because A is NOT 103 obviousness type prior art to B applications.
          Considering how easily electronic data storage systems are being hacked these days for competitive intelligence, even with government sponsorship or participation, there is an increasing danger that competitors may gain access to filed but unpublished patent applications.


            Paul, I will acknowledge that some see the robust prior art effect of a patent application as of its filing date to be a benefit. But if we ever seek true harmonization, we will have to end the prior art effect of applications for invention purposes.

            The larger goal is more worthy that the private interests of some.


              But if we ever seek true harmonization,

              Ned – you are preaching the mantra of Big Corp, and I am truly not sure whether or not you are even aware of that.

              What exactly is this “larger goal” of some One World Order which necessarily must come at the expense of the U.S. sovereign?


              That would NOT be “harmonization” for any of the many countries that give give “novelty” prior art effect to prior application dates.


                Paul, if we remove 102(a)(2) from 103 prior art and strike 102(c) and (d), we will be harmonized with the ROW on prior art.

                That leaves the issue of the grace period. Since we have provisional applications, I say, strike the grace period and let people file their publications as provisional applications the day before they publish them.

      2. 10.3.2

        Paul, as you say, we have been over EPC Art 54(3) before. But do me a favour could you and remind me what those problems are. I have forgotten.


            Thanks Paul. To my mind though, it hardly amounts to your apocalyptic “LOTS” of problems. Yes, of course, it is not fanciful to suppose that, one day, people will routinely hack in to PTO data, but I dont see it yet. And as to the 1973 EPC at the EPO, after 35+ years, I have yet to see the novelty only Art 54(3) provision cause the chaos you envisage. Not that the sky will never fall, of course.


              but I dont see it yet.

              There are none so blind as those that will not see.

              Wake up MaxDrei, Edward Snowden and the Chinese want to speak to you.

      3. 10.3.3

        Paul, you of course knew that for the first hundred and 50 years or so of this Republic that we existed without patents being prior art as of their filing dates without any problem whatsoever.

        The Supreme Court case that suggested that patents should be treated as prior art as of the filing dates did so because the patent was deemed to have been invented on the dated it was filed. This is a first to invent concept that is alien to first to file environment.

        There is no justification to have patents and published applications to be prior art as of their filing dates for any reason. None whatsoever.

        The law has always provided that two patents cannot exist on the same invention. This was prior to the creation of this Republic — it was a rule of common law. That is why we had interferences in the United States. When two patents or patent applications claimed the same invention, we allowed the first to invent to obtain the patent as against the first to file. But when first to invent is gone from the law and the first to file prevails, and when two applications, or an application and a patent are claiming the same invention, all we have to do is accord effect to the priority of the first to file an award the patent to that applicant.

        There is no requirement to make a patent or patent application prior art as of its filing date to prevent two patents on the same invention.


            The AIA only has very limited “derivation” proceedings, which are very different in both scope and other requirements from interferences. As far as I am aware, none have ever yet been instituted. That is not surprising, since in only a small percentage of interferences was “derivation” ever even asserted.


              The mechanisms were carried over – virtually unchanged.

              As I said, you don’t seem to recognize what IS.

  3. 9

    Good post, although the title of the post is not accurate. The court has not “look[ed] for brief” or asked for briefing on this issue yet. It’s asked for a response to a rehearing petition. When one says asks for briefing on an en banc issue, that sounds like it wants actual briefing on the issue. A bit technical, but an important distinction.

  4. 8

    Somewhat related to Paul’s comment at 6 is pre-AIA 103(c)(1) which disqualified prior art under 102(e), (f), and (g) developed by another inventive entity if “the subject matter and the claimed invention were, at the time the claimed invention was made, owned by the same person or subject to an obligation of assignment to the same person.” 103(c) was added by Pub. L. 98-622 (1984), omitting 102(e) which was added by Pub. L. 106-133 (1999).

    (AIA 102 carries this over to (b)(2)(C), but the critical date is the effective filing date.)

    Thomas.gov does not have legislative history before 1995, but Chisum on Patents (1995) reproduces the legislative history in Appendix 22. From it, “[T]he bill changes a complex body of case law [In re Bass and In re Clemens] which discourages members of research teams working in corporations, [etc.]”
    Summarizing further: Prior art is not considered provided “it is commonly owned at the time it is made.” The prior art can still lead to anticipation (double patenting not allowed) and only inventors can be named on patent applications. The applicant has the “burden or establishing that it was commonly owned.”

    But note, the law says “or subject to an obligation of assignment.” The legislative history states that the language is more precise than a parallel H.R. 4525 which said (according to the CRS) “is known only to applicant by virtue of his or her employment.”

    Congress apparently thought that an obligation to assign was good enough to transfer ownership.

    1. 8.1


      You may want to read again the Stanford v Roche case and notice that the particular language of that “obligation to assign” is absolutely critical.

      1. 8.1.1

        Only because Stanford v Roche was wrongly argued, ignoring California contract law [never even presented] and the Supreme Court landmark decision in Erie Railroad Co. v. Tompkins (1938).
        Federal law, that is the patent statute, does not control employment agreements or other invention assignments other that the limited effect of 35 USC 261 for patents and patent applications, and the joint ownership provision of 35 USC 261,which allows any joint owner [absent agreement to the contrary] to independently do anything they want with the entire patent with no accounting to the others. [Even worse, per case law, they cannot SUE to enforce the patent without voluntary joinder of all other joint owners.]


          Paul, agreed. The Stanford attorneys threw their client under the bus when they did not argue the point about Erie Railroad v. Tompkins and that Filmtec was wrongly decided, not only being in violation of controlling Supreme Court authority, but also wrongly decided as a matter of common law.

          Cooly of Palo Alto, and Jones Day of DC were on the brief at the Supreme Court. I don’t know who represented them at the District Court.


            Of course, Stanford self they have told their own attorneys not to assert that Filmtec was wrongly decided

    2. 8.2

      Troubled, Congress apparently thought that an obligation to assign was good enough to transfer ownership.

      No! All this says is that Congress thought that an obligation to assign was enough to bring the subject invention within the doctrine – exclusion from being prior art for obviousness purposes.

      1. 8.2.1


        At the risk of beating my observations to death, according to the legislative history, the applicant has the “burden or establishing that it was commonly owned.” It omits any mention of a burden to establish an obligation to assign. Was that an oversight? Very possible, but the history says that this version is more precise than a previous bill.


          ” made, owned by the same person or subject to an obligation of assignment to the same person”

          “Owned” or “subject to an obligation….”

          The statute itself recognize that an obligation to assign is not ownership.

  5. 7

    Patents are not subject to most “other” property laws or principles (which are governed by state codes and common law). They are U.S. Constitutional grants, as determined by Congress and the federal courts, and are governed by a balance of the public interest – to promote progress with limited exclusivity. Contracts relating to chattel or goods, like the UCC, are generally not applicable to patents (the UCC does include an implied warranty provision that a seller’s products do not infringe any patents).

    Employment agreements that assign all inventions conceived or reduced to practice within the scope of employment promote progress by providing employers with incentives to hire inventors and invest in R&D, while providing inventors with employment, training, resources and motivation to invent. Since new employees usually do not have an invention when they begin their employment, it is appropriate for an employer to require an assignment of inventions conceived or reduced to practice within the scope of employment, before the inventor is hired. Also, because many inventors come up with many inventions within the scope of their employment, it is more efficient to include all inventions within the assignment provision of the employment agreement.

    While such assignments of future inventions are common in employment agreements, it should not be forgotten that they are not required by federal law and are subject to negotiation along with other terms of an employment agreement.

    1. 7.1

      Not exactly right, Jai Rho – the UCC does include mention of patents as a particular type of personal property.

      That being said, the UCC leaves things, shall we say, a little cloudy.

    2. 7.2

      “Patents are not subject to most “other” property laws or principles (which are governed by state codes and common law). They are U.S. Constitutional grants, as determined by Congress and the federal courts, and are governed by a balance of the public interest – to promote progress with limited exclusivity.”

      The SCOTUS has repeatedly held otherwise, C.J. Rho.

    3. 7.3

      JR, The problem comes up when the employed inventor files a patent application on an invention which he claims as his own. If there is a present assignment of this invention to the employer, the employer owns it. If there is only an obligation to assign, the employer still must claim its rights in that invention.

      Now there was a recent case (circa 2007) where the inventors filed a patent application on an invention that was arguably related to something he was doing at his employer; but he did it on entirely on his own time. He and others invested heavily for more than a decade in developing that invention. And, when they went to court against an infringer, the infringer obtained an assignment from the original employer of the invention. The court held that the infringer now owned the patent, and the inventor could not assert any defenses such as statue limitations or unclean hands.

      When I first read this decision, I went ballistic. Do you have any idea why I went ballistic?

      1. 7.3.1

        Ned, I don’t. Sorry.

        How important to the outcome is it, what sort of employee the inventor was at the time of making the invention? Suppose the inventor was Head of Research. Does such an employee have any “own time” at all, in which to make inventions within the field in which she is employed as Head of Research? Isn’t she a “paid to invent” person, even when she’s asleep? It would be unconscionable for her to assert that “I made the invention in my own time. So, it’s mine, and never mind whether it’s in the field of research of which I’m the Head”

        Conversely, suppose the inventor is a base level junior laboratory technician in the research labs of her employer. Now she does have some “own time” and if she makes inventions in her own time, even in a field being explored by the Research Labs, good for her, I say, and that invention ought to be her property, I agree.


          Max, the point here is not that the employer can properly arrange for assignment of inventions to the employer from the employee in the proper case. The point here is rather that the employer must prove that the invention is subject to the contract for assignment when there is a dispute and must prove that in court in a proper case. Otherwise you are stripping the employee of all rights and treating him or her little more than as cattle.


            It’s worse than that – MaxDrei’s graduated scale makes it more punishable for being more creative and climbing the corporate ladder.

            I don’t think that MaxDrei has had experience in the real world climbing that selfsame ladder. He seems far too eager to remove the personal time of those “higher up,” as if they ARE chattel of “Big Corp.”


              For the graduated scale, see the UK statute from 1977, Sections 39-43, on employee inventions.



                (with, of course, a note of admonition NOT to allow any “comparative analysis” to cloud the fact that what mayhave been “OK” for the UK should not be confused as being “OK” for the US)


          {Resubmitted in view of mystery filter]

          Max, what position the employee was hired for makes a big difference only in those countries that have statutory “inventor compensation systems” under which anyone not hired for R&D may get large financial payments from the employer in some cases for their inventions.
          [Such laws can discourage multinationals from putting R&D facilities in those countries.]


            Good point Paul. The UK provisions were indeed inspired by the German ones. At the time (1977) it was thought they might stimulate innovation in the UK, lifting it closer to the German level. It didn’t, of course, because the culture of innovation is different in UK and Germany. We continue to see loads of technological innovation in Germany and very little in England. Sophisticated multi-nationals (like GE) continue to place R&D in countries where there are statutory inventor compensation schemes, like Germany. They get the best innovators there, and the statutory compensation schemes are open to various “work arounds” that will eliminate any chance of a nasty surprise for the employer.

            The only multi-nationals the schemes discourage are the thick-headed and inflexible ones, not interested in innovating their way ahead. Tax decides where to locate. Covidien is one example. Pfizer and Allergan another.

            Remember that saying, from your President General Dwight Eisenhauer wasn’t it: Our technical progress? Why, it’s because our Germans are better than their Germans. The Employee Inventor Law of Germany was introduced by a guy called Adolf, to foster a culture of innovation and to declare that innovation is “a good thing” for the country. It has worked pretty well, over the years, for Germany but perhaps doesn’t work at all, in English common law countries.


              Having strong patents used to be OUR sovereign advantage (your much-maligned presumption of validity, as a matter of fact was one big part).

              Then people actually started believing the anti-patent propaganda.


              Max, there is no doubt that the ability to enforce a patent was a prime reason for America’s inventive progress for such a very long time. That is now being crippled by big business. It is my understanding that investments in startups in Silicon Valley have fallen off dramatically because of cases like eBay, and because of IPRs. The American Golden goose has been slain.

              It is guys like Paul Morgan who are responsible.


              Max, regarding the UK and innovation, you have to ask yourself whether the EPO or European patent system is really working to foster invention in startups. It appears to me that the European patent systems are enormously expensive, and patents are subject to oppositions which are themselves expensive. Then there’s a matter of enforcement. It does appear that the patents are enforced in Germany but not so much in England.

              I would think that Europe might want to look at ways of reducing costs to make the system more affordable for small businesses, and more enforceable by also looking at European wide remedies.


              Max, we once investigated the UK as a possible location for R&D. We decided not to go there because of the “class” problem. That is never going to be fixed by a patent system. The Brits have to figure out that that they are their own worst enemies.

  6. 6

    As some of the commentators have noted, patent applications are increasingly joint inventions with plural named inventors, usually from the same company. Contrary to the views of some employed inventors who would like to have their names on everything, companies have no legal, financial, or other incentive to leave any actual joint inventor of a patent application. Rather, a strong incentive to put too many names on a patent application and to leave them on even after the claims are changed in prosecution. Furthermore the AIA has made it even easier to correct erroneous inventorship designations.
    As another commentator notes, there is no good reason to add an additional old layer of legal complexity to the process of employed inventor invention assignments. As an analogy. you may recall that medieval English property law required manual transfer of a clod of earth from the land to complete a land transfer.

    1. 6.1

      PM: there is no good reason to add an additional old layer of legal complexity to the process of employed inventor invention assignments.


    2. 6.2


      As I indicated to Prof. Crouch, the machinations in the drive towards the AIA are definitely worth some study.

      It is no accident that the AIA was heavily influenced by the “it takes a village” mindset which coincides with the Large Corp multinationalism “concerns.”

      But let’s be perfectly clear: what is in the interest of Large Corp multinationalism is NOT in the interest of the US sovereign.

      Respect of sovereign does tend to get in the way of a One World Order with our constitutional basis of patent law.

      Yet another “inconvenience” to those wanting to make the patent right a lesser “efficient breach” infringement business “contract” matter rather than a “real person” personal property matter. – as witness by the comment of Fred Rule: “The current rule is a hack to get us in to the same end that other countries more sensibly provide for, ownership vesting in the company in the first instance.” Note Fred’s inclusion of the subjective “sensibly,” which is ENTIRELY debatable – and in fact – is NOT sensible when one considers the history of our sovereign: we PURPOSEFULLY wanted this to be a real person thing, with that real person having power to then fully and freely alienate that property right.

      We (the US patent system, based on established views of that system reflected in award winning historical journalism) do NOT want the Corporate “person” to have that power, and it is only by Us (the royal Us) falling asleep to the “slow boil” that we are in danger of losing that very precious piece of our sovereign identity. We really so need to be aware of (and beware of) the encroaching power of multinationalism and corporatism, as such entities may want “rights” in all of the sovereigns, but truly owe NO allegiance to ANY of the sovereigns.

    3. 6.3

      Paul, as a rule of thumb, there is no vice to include as named inventors those who are not inventors. But there is a severe problem if an of the subject matter claimed is not named.

      1. 6.3.1


        I have to disagree, if not on principle alone.

        The oath being sworn to by ALL inventors should not be “cheapened” by such “catch-all” manner of including everyone and “meh-so-that-person-rally-was-not-an-inventor>”

        Also, on a recent thread there was a discussion of the danger of individual contributors being able to “defeat” the group by sub-licensing and the difficulties of having to bring all “owners” in for any enforcement actions.

        Believe it or not, there to this day remains many smaller entities gobbled up into larger entities that have a wide variety (including no) of employment agreements – with and without clauses assigning inventions to the corporation.

        I do believe that your post exhibits a needlessly cavalier attitude in this regard.

  7. 5

    Looks like I’m wrong about the panel judges. It is very likely that this request for a response, made just a day after the filing of the petition, came from a panel judge. En banc petitions first circulate to the three-judge panel.

    So, there might be one or more en banc yes votes among the panel judges.

      1. 5.1.1

        I don’t think a rehearing petition (whether en banc, panel or combined) would ever go to a motions panel:

        Action on a petition for rehearing en banc that is part of a combined petition for panel rehearing and rehearing en banc will be deferred until the panel has acted on the petition for rehearing. A petition for rehearing en banc that is not combined with a petition for panel rehearing will be presumed to request relief that can be granted by the
        panel that heard the appeal; consequently, the clerk will send
        the petition for rehearing en banc promptly upon filing first to the panel in accordance with IOP #12, paragraph 1(b), and action on the petition for rehearing en banc will be deferred until the panel has had
        the opportunity to grant the relief requested.

        — IOP #14(2)(a)

  8. 4

    The current rule is a hack to get us in to the same end that other countries more sensibly provide for, ownership vesting in the company in the first instance. If a person is employed to invent, that person shouldn’t have any ownership interest to start with (which is a separate issue from rewarding invention financially on a patent-by-patent basis).

    And does it really make a difference, other than creating more friction for the employer to perfect its ownership? If employed to invent the employee has a duty to assign, as also expressed in the employee agreement, so the company has equitable ownership. Without the ability to take an assignmeet of future rights, the employer will, I suppose, instead have to go to court to have the court grant legal ownership of the invention. But you get to the same place, just at a greater cost, which isn’t in our societal interest.

    1. 4.2

      Fred, I think the Supreme Court in Stanford v. Roche held that inventions are initially owned by their inventors – and this appears to be a matter of federal law as opposed to state law.

      The assignment of inventions can be a matter of state law, but not its initial ownership.

      Do you agree or disagree?

    2. 4.3

      This was exactly my reaction, too.

      We currently have System (A): An employer and employee sign one agreement, and it automatically assigns the future creation of IP. It’s convenient, certain, and relatively painless.

      Under consideration is System (B): The employer and employee have to re-execute the contract every single time, for every single patent. This system raises questions like: Does the conveyance of a particular patent also cover continuations, or continuations-in-part? – or do the employer and employee have to re-sign the contracts again for every such follow-on application?

      System (B) creates all kinds of “gotcha!” loopholes, like where an employee has assigned 47 patents to his employer but not the 48th – and the employee can then:

      (1) Convey the rights to a competitor as a get-out-of-infringement card; or

      (2) Sue the employer for infringing the employee’s patent – and the employee certainly has intimate knowledge of the infringement, because the employee built all of the employer’s products to centrally rely on that particular invention, and the employee now basically has a gun to the employer’s head and can demand $$$$$$$$$$$$$$$$.

      Why in the world would anyone advocate for System (B)?… I mean, for any purpose other than further weakening patents?

      1. 4.3.1

        David, I am somewhat shocked by your statement. It seems to say that many companies right now do not obtain assignments of their patent applications from their inventors when they file patent applications. That is contrary to my experience. Please confirm what you just said.



          Sadly, I can confirm what David says.

          As I posted above – many smaller entities lack employment agreements (at all) or have agreements with a wide variety of requirements to assign.

          Criminy, the Stanford case itself showed an entity with such a poorly drafted agreement. Stanford. The very same school associated with Lemley. If they did not take care “to do it right,” why would you ever assume that the vast number of small (or even medium or even large) organizations “do it right” – or for the large “gobble them up” corporations can even force redo’s of employment agreements on those smaller entities that they so gobble up?


            Anon, what I am talking about is patent attorneys filing patent applications and not obtaining assignments of the patent applications from their inventors without either the serial number of the application or the title of the application in the assignment, the latter when the assignment is filed with the application.

            I understood David to say that this is not being done today by law firms. I am incredulous.


              I think that you are re-phrasing what David is saying to a much much more narrow set of circumstances, that is quite divorced from the larger topic at hand.

              Move the goalposts back, Ned.


                anon, I see a small problem. Under 152, as construed, the inventor must sign the application filed by the assignee (equitable or not) in order for a proper conveyance of the patent (patent application) from the inventor to the assignee.

                If the inventor has an equitable obligation, and signs the application filed by the equitable assignee, I think that should be sufficient to convey legal title – but we have no case that actually holds this.

                As to an inventor who has a equitable obligation to assign, he can file or convey to another, but he cannot avoid the equitable obligation. It will bind him or his putative assignee.

                1. Did you mistype….?

                  Did you mean:

                  If the inventor has an equitable obligation, and the corporation signs the application filed by the equitable assignee, I think that should be sufficient to convey legal title – but we have no case that actually holds this.

                2. anon, inventor->(equitable)corp.

                  Corp files.

                  Inventor signs application.

                  This may be sufficient to convey legal title to corp. under 35 USC 152.



          No, my experience comports with yours: companies routinely obtain and file assignment documents for each application. (And it does routinely create headaches.)

          However, some of these companies also include the automatic assignment clause in their employment agreements. It’s a belt-and-suspenders approach, designed to create a safety net in case something slips or an inventor tries to game the system.

          But I’m not describing my experience, or even general practice. I’m comparing these two systems from a top-level approach: a system that enables automated assignment (either as a safety net, or as the default mechanism) vs. a system that restricts them in all circumstances. My argument is that the latter imposes significant costs without any significant advantage.


            David, “efficiency” is in the eye of the beholder. After all, a guy named J.C. was once arrested around midnight, tried at dawn, beat to a pulp with scourges by drunk guards having a good time, was crucified at noon, and was dead by 3.

            Now, such is efficiency. But only if one views it from one side.

            We did have this case where the inventor filed his on application on work he did on his own time — received a waiver from his company — left the company —developed his business with others for more than ten years — sued only to have his case dismissed when the infringer “bought” the patent rights from his former company because of such a present assignment clause, where an action to specifically enforce the contract to assign would have long been barred by a statute of limitations.


              Ned: That’s a very specific and peculiar set of anecdotal circumstances. Not generally the type of thing that motivates generalized policy changes.

              And even in the case you describe, I question the circumstances for at least the following reasons:

              (1) I’d question the scope of the employment agreement – most such agreements specify that inventions are only subject to assignment if they are in the scope of employment. (And the reason that most contracts stipulate this limitation is that they’re likely to be so limited by contract law, whether or not they do.)

              (2) I’d question why, if the inventor invested so much time and energy into embarking on this project, he didn’t get the waiver in writing – or at least have an email exchange on hand, to evidence their mutual understanding. If he’d had any type of patent counsel to secure the patent, why didn’t they advise him of that?


                He did get the waiver in writing.

                The court that the waiver did not apply because the employee had not given the employee a copy of the application or patent, as the case may be (I don’t recall.)

                There was a remand to determine whether the invention was in fact within the scope.



          I find once again that you simply assume too much with your views of “equitable assignments.”


          What is your point? Are you suggesting equitable assignment solves anything? The owner of the equitable interest really has nothing until it is converted to legal ownership because it doesn’t have standing for a patent infringement lawsuit. link to patentlyo.com.


            Fred, are you talking to me? I fully understand what an equitable assignment is. I question whether David does.


          Ned: Sure, but equitable assignment relies on a balancing test, the discretion of the court, etc. With the patent values at stake here – ranging up to $1B – who wants to leave any piece of the puzzle unresolved?

          Let’s say Company A tries to assert a patent against Company B, which says, we don’t infringe because we licensed the patent from Q, one of the inventors and your ex-employee. Consider these scenarios:

          (1) Company A produces a written assignment document from Q to A covering all of Q’s interest in the patented invention.

          (2) Company A produces Q’s employment agreement, which has a legally enforceable clause that automatically assigns, from Q to A, all of Q’s interest in any patents during Q’s employment and in the employment scope, including this invention.

          (3) Company A produces Q’s employment agreement, points to Q’s declaration, and argues for an equitable assignment of Q’s interest to A.

          Are you really telling me that you perceive these three situations as equivalent? If you were Company B’s counsel, would you prefer any of these situations to the others?


            David, the employee cannot grant more than he has. If he has an obligation to assign to the employer, the putative assignee only obtains legal title in trust subject to an obligation to assign. The employer can still obtain specific performance from the employee/infringer and record the legal title when conveyed.

            The real problems with the present assignment of a future invention arises when there is a genuine dispute as to whether the invention falls within the employment agreement in the first place. It strips the employee of the statute of limitations, equitable defenses of unclean hand, and such things as that the deal with the employee was unfair to begin with. In order to get specific performance, the party requesting it must show that it is not in breach itself, that all conditions precedent have been satisfied, and that the contract itself was just and reasonable.


              It strips the employee of the statute of limitations, equitable defenses of unclean hand, and such things as that the deal with the employee was unfair to begin with.

              Our contract law presumes that people have the right to contract away their rights, and only asserts “unfair deals” in truly exceptional circumstances where it’s manifestly unfair – stuff like universal non-competes (e.g.: “you agree not to work in this business for at least 50 years following the termination of your employment with us”).

              You seem to presume that “inventions within the scope of your employment belong to us” is an unfair contractual arrangement. That strains credulity.

              Indeed, you don’t even seem to be protesting the obligation to assign – only the automatic assignment. This is an extremely minor incremental tweak, and it provides significant value to the employer (in preventing its IP rights from leaking down the drain due to a paperwork glitch or employee gamesmanship).

              What is it about this arrangement that pushes your buttons? Is it just an inclination to shoehorn IP law into the law of contracts for the sale of physical goods, coupled with that one peculiar anecdote that you posted above? Not very persuasive, Ned.


                Let us agree to disagree their David. The difference between a legal assignment and an equitable assignment to me is not minor for the reasons previously stated.

    3. 4.4

      Fred, it is common practice when filing a patent application to obtain of assignments of the patent application an invention naming the title of the patent application or the serial number of the application and then record that assignment. In the ordinary case there is no problem.

      The problem arises when the inventor neither wants to sign an assignment and also refuses to sign the application.

      Now is my understanding that historically the patent office would allow multiple inventors to proceed without the cooperation of one inventor. It would also allow the equitable assignee to proceed on proof of an obligation to assign. But nevertheless, in the end, without a legal assignment the putative assignee does not get the legal title from the nonsigning employee/inventor.

      1. 4.4.1

        Now is my understanding that historically the patent office would allow multiple inventors to proceed without the cooperation of one inventor.


        Not at least until the AIA.

        **( I am not talking about the ability to proceed if the inventor could not be found with some due diligence)


          Actually, anon, they could. We used to have the problem all the time. One had only to prove that one proffered the application for his signature and that he refused to cooperate.


            We used to have the problem all the time. One had only to prove that one proffered the application for his signature and that he refused to cooperate.

            Ned, was that because you also provided a copy of the employment agreement showing an assignment of future patents? Exactly the document that will no longer be any good if the Shukh ruling is reversed?


              Fred, no — it was because all inventors have to sign the application (declaration).

              My point is this: if the inventor signs the application of his equitable assignee, that should be enough to convey title to the assignee. This might be a good avenue for the en banc court to pursue.


            Not so Ned – I think that perhaps you “got lucky**”

            **lucky in the sense that it was processed without a signature – much like the present situation.

            There can be plenty of reasons why an “supposed” inventor refuses to sign – and there is NO way that you can safely “assume” that you [the corporation] are “safe” without the signature.

            One example that comes to mind is that the “supposed” inventor cannot sign in good faith. Perhaps he KNOWS (or even merely thinks) that someone other than himself is the true inventor, but some other situation prevents the “supposed” inventor from revealing this “truth.”


              (again, the “not safe” comment should be considered outside of the provided “can’t locate or find” type excuses available)


              Well, for some reason, we always had lots of inventorship problems at Seagate. We always treated theses as within the exclusive jurisdiction of the patent department, and not a political decision by the executive. From the smell of things emanating from this case, it appears that something happened this time at Seagate that was not on the up and up.

      2. 4.4.2

        But nevertheless, in the end, without a legal assignment the putative assignee does not get the legal title from the nonsigning employee/inventor.

        Ned, exactly the problem – so what did you mean to suggest when you said “David, you have any concept, truly, of what an equitable assignment is?” Equitable assignment doesn’t solve the problem, so I don’t understand why you suggest (as I’m reading it) that it is a solution to the issue that David raised.


          Fred, EQ Asn + Inventor signing the application should be enough to convey legal title per 35 USC 152.

          If the inventor refuses to sign, the employer would have to go to court to enforce by specific performance. The inventor could defend on any number of grounds, including statute of limitation, unclean hands, breach by the employer of the employment contract, etc., etc., etc.


            If the inventor refuses to sign, the employer would have to go to court to enforce by specific performance. The inventor could defend on any number of grounds, including statute of limitation, unclean hands, breach by the employer of the employment contract, etc., etc., etc.

            That’s the problem that Shukh, as it currently stands, solves. With an assignment of a future invention the assignee is the legal owner and doesn’t have to go to court.


              Shukh or Filmtec?

              Regardless, if the inventor does not sign the application, under 35 USC 152, the corporation does not have title.



                I think that I understand what you are saying Ned (you are not using one of the “outs” in pre-AIA law).

                Are you aware that the AIA changed that…?

                1. 35 USC 152 has not been amended, anon.

                  IIRC, a record, legal assignee can file declarations for inventors under some circumstances. But what we are talking about here is an equitable assignee only and an inventor not willing to sign. That is different.

  9. 3

    I’m not trying to be picky but I just had to laugh when I saw “asterix” above, where “asterisk” was intended. Are there any examples of a Spell-Checker inserting a Registered Trademark as a default, when it detects an apparent typing error?

    Dennis’s topic is interesting also to an international readership. In the UK, the Statute declares the invention of an employee in a “paid to invent” position to be the property of the employer, from the moment of its conception. The UK modelled its 1978 law on that of Germany. But in Germany, still, the employer has to claim ownership away from the employee inventor, within a tight timetable. Woe betide it if it does not. Innovative industry, making inventions all over the world, finds a different ownership situation in each legal jurisdiction. Problematic, when inventions are increasingly made in the ether, in the course of internet dialogue between those individuals who will in due course be named as inventors, sitting in a plethora of different jurisdictions.

    1. 3.2

      While it does look, Max, like the Federal Circuit was trying to harmonize American common law with English statutory law.

      1. 3.2.1

        Importation of foreign law by the judicial branch…

        (Ned appears “OK” with this, if done by the Supremes, but…)

  10. 2

    AIPLA are you listening? Everyone thinks today that you are a tool of big business.

    Prove them wrong, just once.

  11. 1

    Not sure that this is “just” a CAFC “Federal” item, as the Supreme Court in Stanford v Roche did the very same thing…

    (see also the background in the AIA, a la Lemley…)

    1. 1.1

      In Stanford v. Roche, the court restated the longstanding principle that title in a patented invention vests first in the inventor. In my view, that is different from determining the meaning of a written document purporting to assign ownership to the invention.

      Can you remind me what Lemley has to do with this?

      1. 1.1.1

        Lemley campaigned hard to insert the “it takes a village” aspect into the AIA and the widespread easing (both procedural and substantive penalty aspects) of the inventor/owner transitions both during and at filing of applications.

        You might recall that part of that issue reflected in Stanford v Roche was the turning of the inchoate right into a viable property right and who owns what when.

        Yes – it is critical to remember that title must always first vest in the individual (since the juristic person of the corporation is not YET recognized as a type of person that can be an inventor). But it is NOT outside the trends that we see that a change to this power of the juristic person may be brought into the inventor/owner discussion. I grant that we are not there yet.


          …and I would add that the notion of inchoate right is a bit more nuanced than the definition of “invention.”

          Even upon realization of an invention (as you here more correctly capture here with the alternative viability of actual and constructive reductions), the property right that is the patent right remains an inchoate right until the “full race course,” as set by Congress, is completed.

          This subtle distinction comes into play in other aspects of patent law – and has been featured in previous discussions of such things as 101, what powers of writing/”interpreting” law belong to which branches of the government, the sui gen erous nature of Prior User Rights, and the AIA elimination of the obfuscation of trade secret time of protection (a state law item) and actual time of protection under patent law ( a federal law item).


            anon since you are so fond of emboldening and italics, might you care to use both, when typing such Latin expressions as “sui generis”? Or was it intended, when you wrote above, that prior user rights are in some way “generous”?

            You know, every little bit helps, when trying to understand what you are getting at.


              Do you mean “sui generis” or “sui generis”… ? 😉

              I am more than sure that my meaning is clear to those “with minds eager to understand”

              Yes, there are multiple layers of understanding in the “generosity” term.

              Feel free to peel that onion at whatever layer you want to engage in. As I fully indicate – the nuances here strike several different items.


                As I figured, MaxDrei does not really want to engage at all, just whinge because the post was by “anon”…

Comments are closed.