Revisions are ongoing and should be finished by 19 November. Your comments or suggestions are welcome!

Patentability of Software after Bilski, KSR, and the America Invents Act: Looking Back and Looking Forward

-- By ThomasHou - 19 Oct 2011

Section I: Bilski's Indirect Effect on Computer Software as Eligible Subject Matter

For patent law aficionados who were expecting the Supreme Court to make a definite pronouncement on the patentability of process patents such as computer software, the Court's decision in Bilski v. Kappos was a letdown. The Court declined to "comment[] on the patentability of any particular invention" and left to the Federal Circuit ("FedCir") to develop doctrine on the patentability of process patents according to the Supreme Court's precedents (including unpatentable subject matter, such as abstract ideas). For computer software, the Supreme Court and the FedCir? both failed to address whether a general purpose computer could satisfy the machine prong of the FedCir? 's machine or transformation test, or a machine specifically adapted to the claimed process was necessary.

Since Bilski, the PTO and BPAI have in fact begun to deny a large number of software-only patent applicationS, albeit without establishing clear doctrine. A July 2011 study of post-Bilski BPAI and district court cases on patentability found that a majority of software patent applications were invalidated, mostly because they were deemed abstract ideas. It is yet unclear the PTO and the BPAI's position whether a general purpose or a specially adapted machine is required. Most notably, the FedCir? has not had an opportunity to rule on the limits of patentability for pure software, which explains why the doctrine remains unsettled in this area. Nonetheless, a clear trend is observable that the Supreme Court's Bilski decision had trickle down effects on how lower courts assess patentability of computer software claims.

Or, to put it another way, that would be following the position of the Federal Circuit itself, in the portion of its decision in Bilski to which the Supreme Court's subsequent decision is not incongenial.

However, the PTO has endorsed the general purpose computer option and with few contrary indications from the Federal Circuit, meager disagreement exists about the subject matter eligibility of computer software, at least in the courts.

I don't think that summarizes the position of the PTO or the BPAI correctly at all. I see lots of rejections that suggest a different reading of the "machine or transformation test" and doubt about whether "the Internet" is a machine either. So if your position here results from your own assessment of evidence, it would be good to read more about what you looked at and how you reached your conclusion.

Section II: The Continued Viability of Nonobviousness to Bar Trivial Computer Software Patents

Before software patentees get too excited, they should remember the Supreme Court's counsel that eligible subject matter is a mere floor and claimed inventions must also be "novel, nonobvious, and fully and particularly described. These limitations serve a critical role . . . ." The nonobviousness requirement under Section 103 of the Patent Act has long been the gateway for patentability and the hardest for patentees to satisfy. The Supreme Court laid out the standard test in its Graham v. John Deere Co. decision. In the early case of Dann v. Johnston, the Court applied the Graham factors and held a computer software program to help bank customers invalid for obviousness. The Court's most recent case on nonobviousness is illustrative and useful for testing this requirement against claimed computer software.

In KSR Int'l Co. v. Teleflex Inc., the Supreme Court considered the nonobviousness of a patent for an adjustable electronic pedal with a fixed pivot point used for vehicle control. The Court emphasized that its precedent called for a flexible and functional approach and rejected the Federal Circuit's exclusive reliance on its teaching, suggestion or motivation test. For patents claiming combinations of prior art elements, courts should be cautious and look at "interrelated teachings of multiple patents; the effects of demands . . . in the marketplace; and the background knowledge possessed by a person having ordinary skill in the art." The Court recognized that market demand will often drive design trends: "When there is a design need or market pressure to solve a problem . . . a person of ordinary skill has good reason to pursue known options within his or her technical grasp." The Court held the patent at issue obvious and cautioned courts not to stifle the progress of the useful arts by applying rigid and narrow tests for obviousness.

Through KSR, the Supreme Court maintained the traditional threshold Graham factors test for nonobviousness for the Federal Circuit to follow. Having a high threshold is important for assessing the patentability of software, for nonobviousness limits the granting of exclusive rights to software whose inventiveness is trivial and which would have been obvious to the ordinary-skilled software inventor (whose skill would be quite high today). The field of software is broad and its prior art reaches beyond computer programs to that of the electrical arts and other fields. In Muniauction, Inc. v. Thomson Corp., the Federal Circuit considered the nonobviousness of a software system for auctioning and bidding for municipal bonds using a web browser. Finding little difference between the systems in the prior art and the use of a web browser well-known, the Federal Circuit invalidated the patent. To encourage free software, the Federal Circuit and PTO should not relax the high threshold for nonobviousness. Furthermore, the Federal Circuit has traditionally relied on secondary considerations such as commercial success, long-felt but unsolved need, and failure of others when nonobviousness is at a balance. With software creation increasingly having a marginal cost of zero, the importance of those secondary considerations should be reconsidered. After all, their use is optional and depends on the context. The Federal Circuit should be wary of using secondary considerations to uphold the nonobviousness of software when the main Graham analysis is close.

You begin by talking about scope, or section 101 if you want to be statutory about it, and then move without any transition to obviousness under section 103. That these involve entirely different procedural environments for challengers passes without discussion, which it shouldn't.

Nothing here discusses the real importance of Bilski, which is not its effect on new patent applications, but the light it sheds on the validity vel non of tens of thousands of high-abstractness patents created in a more lenient period, when the state-granted monopoly was available to anyone corrupt enough to apply for it. Even after the Supreme Court dithered its way to the conclusion that there may be patents on software and business method claims that are valid even though there is no special-purpose machine or transformation of matter, the litigation prospects of patents that evidently don't contain any such graces, like the ones that are purely patents on mathematical operations, are lousy. Yet patents of that type are currently being brandished by all sorts of fools trying to extort all sorts of parties (including even some free software developers), and are being sold as weapons at pretty high prices. If you're really trying to figure out the significance of Bilski, that's where you'd want to look.

But I still don't know, after two full readings of the essay, what it is about. Case summaries don't teach me anything, and what little here isn't a glorified West headnote is passing editorialization, with no discipline or data specified. What we most need here is a thesis, after which it should be possible to figure out how to use the remainder of the essay to explicate it, respond to its critics, and indicate its implications.

Section III: How the America Invents Act Ducks the Questions Regarding Computer Software Patents

---

You are entitled to restrict access to your paper if you want to. But we all derive immense benefit from reading one another's work, and I hope you won't feel the need unless the subject matter is personal and its disclosure would be harmful or undesirable. To restrict access to your paper simply delete the "#" on the next line:

# * Set ALLOWTOPICVIEW = TWikiAdminGroup, ThomasHou

Note: TWiki has strict formatting rules. Make sure you preserve the three spaces, asterisk, and extra space at the beginning of that line. If you wish to give access to any other users simply add them to the comma separated list