The America Invents Act includes many provisions that will be analyzed and litigated for many years. Much of the media attention is focused on the transition from "first to invent" to "first to file." This article explains what that transition really means and how the implications are significantly more complicated and damaging than they may appear.
The Act was passed into law in September 2011. It represents the most fundamental change to patent law in more than 50 years. Supposedly, the Act is an attempt to fix a broken patent system, yield more certainty regarding inventorship, and harmonize the United States patent rules with those of the rest of the world. In reality, almost all of the provisions are bad for the patent system and for promoting innovation in general.
The current law relating to inventorship ("first to invent")
In order to fully appreciate the change in the law, it is helpful to understand the current law. Patent protection is available for inventions that are new to the world. To acquire a patent, you must file a patent application with the Patent Office. A patent examiner will review the application and grant a patent if the invention is new to the world.
The current law awards a patent to the first "inventor." The inventor is the person who can prove that he first conceived of the invention and reduced the invention to practice. Conception means that the inventor can explain the invention to someone else. Reduction to practice means either: 1) making a prototype or 2) filing a patent application with the Patent Office. If there is a dispute over who "invented" first, such as when two inventors file patent applications claiming the same invention, an interference proceeding is declared in the Patent Office. In this proceeding, two parties present evidence regarding conception, reduction to practice, and diligence between conception and reduction to practice. Interference proceedings are fairly rare in the electrical and computer science arts. However, because these proceedings have the potential to occur, inventors sometimes keep an inventor journal to prove conception or diligence if necessary. Keeping these journals was cumbersome. There were also complaints that even if a patent application was filed, there was uncertainty regarding patent rights because another inventor could come along and prove prior conception.
Here's an example. Joe Inventor files an application in the United States Patent Office in March. The patent examiner reviews the application and says that the invention is not patentable because another patent application disclosing the same invention was filed in February. Joe inventor may be able to "swear behind" the reference by averring that Joe reduced the invention to practice in January or conceived of the invention in January and was diligent in filing his application in March. Such a swearing behind may lead to the Joe being named the appropriate "inventor" for this invention and awarded patent rights.
Also relevant in the current law is the concept of a grace period. When an invention is shared with the public, such as in a sale or publication, you have a year to file an application with the United States Patent Office and you may still obtain patent protection. In contrast, most relevant countries outside the United States have an absolute novelty requirement where any non-confidential disclosure, generally, would be an immediate bar to patentability.
The new law ("first to file")
Under the new law, the "effective filing date" of an application is defined as the filing date of the application with the Patent Office. Anything prior to that date which discloses the invention may bar patent protection. Specifically, according to the Act, a patent may NOT be obtained if the invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
This change appears to completely remove the grace period. However, there is an important exception in the statute: a "disclosure" made one year or less prior to the effective filing date is NOT prior art if the disclosure was made by the inventor or by another who obtained the subject matter either directly or indirectly from the inventor. This latter concept refers to "derivation." The idea is that if the inventor "discloses" the invention or someone derives from the inventors disclosure, those events should still yield the one year grace period.
In an example, Joe inventor conceives of the invention in January and files an application in June. Another person conceives of the invention in February and files in May. Under the old law, Joe could submit facts indicating that he conceived of the invention before the other person's filing and be awarded patent rights. Under the new law, the other person would be awarded the patent rights unless Joe could prove that the other person derived the invention from Joe.
Here come the problems
The term "disclosure" in the exception is not defined. There was some testimony that this term means — something that makes the invention available to the public, that is, an interested person skilled in the field could find the disclosure through reasonable due diligence and understand the invention. This definition is not in the statute and thus not necessarily the law. It also does not appear to help us understand the implications of sales, offers for sale, or disclosures to small groups such as at a conference.
One of the supposed goals of the Act was to remove interference proceedings. Those proceedings include archaic rules, create potential uncertainty regarding inventorship, and require extra work because the inventor has to keep records. All true. So those interference proceedings, with known rules and decades of law are removed. And replaced with derivation proceedings — with new rules, no case law providing guidance, and decades of uncertainty while we figure it out. Here's what we know about derivation proceedings.
An applicant for a patent may start a derivation proceeding in the Patent Office if he files a petition within one year of the publication of a claim that is substantially the same as the inventor's claim. Let's break that down. These proceedings relate to a patent application published by the Patent Office only – that excludes, for example, an IEEE publication, or a blog post. Further complicating such a petition is that it must relate to a publication that "claims" the invention. Patents include: 1) a specification describing the invention and 2) claims which define the scope of the invention (typically starting with "What is claimed is:"). Here's the anomaly. If a prior filed patent specification discloses the invention but does not recite claims directed to the same invention, the prior filed specification could be used to invalidate the later filed application. However, a derivation proceeding cannot be declared because the claims are directed to different subject matter.
More troubling, how would you know if someone derived an invention from your publication? If, prior to applying for a patent, you publish about your invention, and another person reads that publication and subsequently files a patent application, the wording of his application may look very different from the wording in the publication. Suppose the chain is two or more entities removed. For example, you publish about your invention, a second person blogs about the publication, and a third person reads the blog and files an application based on the blog? What if the blog itself is used as prior art against your application? For example, you publish in January, someone blogs about the publication in February and you file a patent application in March. The blog would be a publication of the invention prior to your filing date and derivation proceedings do not appear to be available for non-patent literature.
What do you do?
As mentioned above, one of the supposed goals of a first to file system was to remove interference proceedings and so there should be less record keeping. However, you still must retain some records in case you want to prove that someone "derived" from your publication. Further complicating derivations, and perhaps requiring even more record keeping, is that in a derivation proceeding, you may need significantly more facts from your adversary to prove derivation.
If an inventor calls his own publication a "disclosure" and therefore should be awarded the one year grace period, there is an admission that the inventor's publication discloses the invention. Therefore, the publication may be used to affect the scope of a patent. Statements that inventors may put in a publication, a document that is generally not reviewed by an attorney, may affect the rights of the patent. A publication may yield all the detriments of a provisional application without the potential benefits. A provisional application is an informal application that is filed with the Patent Office. It will provide a filing date for that which it discloses. However, as it tends to be informal, it is typically not reviewed as carefully as a full non-provisional application and therefore may include language that may be harmful to the scope of protection.
I used to be very much opposed to provisional applications because of their tendency to be informal and for them to be filed with little review from a patent attorney. However, under the new law, such applications may become necessary to get an application on file quickly before a publication with at least some attorney review. Of course, if you file quickly, you may not put the requisite time into making sure each term in the claims defines the invention in the way you want it to.
So, the best advice may be, do NOT publish that article.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
