It's been six weeks since my last AliceStorm update, and we've had plenty of action:  twelve §101 decisions, and fourteen patents invalidated in just that period.  That said, the success rate of motions on the pleadings is dropping, now down to a mere 69.6%.  At PTAB, ten new institution decisions, all of which were granted on ineligibility grounds.  And PTAB continues with its 100% kill rate, with seven (!) final decisions invalidating patents.  

Here's the data.

May 28, 2015:  

First, an update from the AIPLA Spring Meeting. Drew Hirshfeld, Deputy Commissioner of Patent Examination Policy made some interesting announcements. First, the Office is preparing to release additional examples of eligible and ineligible patent claims for both life sciences and software; he would not give a firm date, but suggested that it would by the end of June. Second, in response to a direct question from a member of the audience as to whether there was a "101 panel" that reviewed all 101 rejections, Hirshfeld specifically stated that there was no such panel, and that each primary and SPE had authority to sign off on a 101 rejection (or withdraw a §101 rejection).  There are numerous SPE and others who are considered experts on §101, but they do not dictate 101 rejections or allowances.  My read of this is that when an examiner tells you she has to talk to the 101 panel, it's a classic appeal to authority negotiating tactic.

Finally, we recently heard from a primary in the Patent Technology Center (TC) 3600 that there has been another round of examiner training in that TC, whereby examiners were instructed to not reject claims unless the claims were similar to those that have been invalidated by the Federal Circuit.  The examiner went so far as to say that some of the examiner's prior rejections would likely not stand under the new training.  Let's hope this is an accurate statement of a shift in Office practice, particularly in TC 3600, which has a current §101 rejection rate of over 50%.  

Federal Circuit:  Two decisions at the Federal Circuit.  First, a per curiam (Newman, Reyna, Taranto) affirmance of the invalidity of U.S. 6,192,347 in Fannie Mae v. Graff/Ross.  Not a surprise, as the claims were a heavy computer-window dressing on a method generating a purchase price for a "component" of a "fixed income asset."  Interesting irony: the invalidated claim was claim number 101!  Here is the claim marked up to strike out the window dressing:

101. A method for making financial analysis output having a system-determined purchase price for at least one component from property in consummating a sale, the financial analysis output being made by steps including:

converting input data, representing at least one component from property, wherein the property is a fixed-income asset, into input digital electrical signals representing the input data;

providing a digital electrical computer system controlled by a processor electrically connected to receive said input digital electrical signals and electrically connected to an output means;

controlling a digital electrical computer processor to manipulate electrical signals to compute a system-determined purchase price for the at least one component from property in consummating a sale and corresponding purchase of the component; and

generating the financial analysis output at said output means.

Yes, this was filed back in 1992, and the sentiment at that time was this would work. But even so, there really should have been more real meat to the invention, other than simply computing a price of an asset on a computer.  It is claims like this that give real software inventions a bad name.

What is really disturbing however is the decision in Allvoice Developments v. Microsoft.  Here, the Court (Prost, Dyke, O'Malley) invalidated on a very unusual ground:  the claims did not recite something in the four statutory categories of machine, process, manufacture or composition of matter. The claims recited an application programming interface for a speech recognition engine: 

 "A universal speech-recognition interface that enables operative coupling of a speech-recognition engine to at least any one of a plurality of different computer-related applications, the universal speech-recognition interface comprising...."

 "A speech-recognition interface that enables operative coupling of a speech-recognition engine to computer-related application, the interface comprising...."

According to the Court:

Software may be patent eligible, but when a claim is not directed towards a process, the subject matter must exist in tangible form. Here, the disputed claims merely claim software instructions without any hardware limitations.

And software is not tangible because it does not meet the Supreme Court's definition of either a machine or an article of manufacture. And where did these definitions come from you ask?  The definition of "machine" comes from the 152 year-old decision in Burr v. Duryee (1863): "the claimed invention must be a ‘concrete thing, consisting of parts, or of certain devices and combination of devices.’”   Upshot: the Federal Circuit has just fixed patent-eligible "machines" in the framework of the Industrial Revolution.  If your invention would not be recognized as a "machine" of parts and pieces by a person who could not even conceive of a digital computer, to whom something as mundane to us as a microwave oven would appear to be magical---"What sorcery is it that makes the food heat without fire?!"----then you are out of luck, it's not eligible subject matter.   This approach ignores the fact that the definition of "machine" as evolved over time: a machine no longer must have moving parts, for example.  Similarly, according to the Federal Circuit, an article of manufacture requires "two or more substances," and obviously software is not a substance.  

Then the Court created a brilliant Catch-22: it rejected Allvoice's argument that "software must necessarily be in a machine readable, physical state in order to exist, and absent a tangible medium, software or data is not a manufacture, and it would not imply the existence of a tangible medium into the claims where none was recited.  Wait for it:  had the claim recited a tangible medium, the Court would have then said that's irrelevant, because the claims are really directed to nothing more than the information itself, and that is an abstract idea ala DigiTech v. Electronics for Imaging and thus ineligible.  Gotcha! 

Here's a riddle: Is it a patent eligible machine if it is made of a single atom?  How "tangible" is that?  Certainly not tangible in the sense expressed by the Supreme Court in 1863.  If a single atom is "tangible" then software, which is stored in some fashion using arrangements of atoms must also be tangible.  So if you are going to play the reductionist card of tangibility, software qualifies.

Second, the Federal Circuit's approach is simply at odds with how inventors in software think of their inventions.  An application programming interface, a database architecture, a communications protocol, a programming framework are just as much real "things" as any hammer or nail.  Like all machines, they "do work," they have specific function and/or structure that can be manipulated.  That's what matters about machines, not their specific material instantiation.   And that's what matters to inventors.   If a theory of human behavior does not actually explain how humans behave then we discard it, just like we discard other theories that don't fit the facts.  The patent law embodies a theory of human behavior of what it means to invent.  Section §101 as it stands now is radically at odds with how we go about inventing.  

District Courts

First, two entirely opposing decisions: Kenexa Brassring v. Hireability and Wireless Media Innovations v. Maher Terminals.  Both are motions on the pleadings.  In Kenexa, the court makes clear that motions to dismiss are disfavored for §101, that they should be the "exception, not the rule," because "every issued patent is presumed to have been issued properly, absent clear and convincing evidence to the contrary."  The court then finds that the defendants did not meet their burden for such evidence:  that there are factual issues as to whether the defendant's  proposed analogies "accurately reflect “routine” business practices," and further that "the Court cannot assume that the claims detailed above would pre-empt" the identified business practices in fact.  "Accordingly, defendants have not put forth “clear and convincing evidence” that the patents-in-suit contain no limitations that are “sufficient to ensure that the patent in practice amounts to significantly more than a patent upon the [ineligible concept] itself." Motion denied.

The court in Wireless Media went the other way, and granted the motion to dismiss.  First, there is no presumption of validity: "the Court adopts Judge Mayer’s approach and will not afford Plaintiff’s Patents presumption of subject matter eligibility." Not only that, but requiring "clear and convincing evidence" from the defendants would "effectively create a near impossible threshold for a defendant to clear when assessing a patent’s subject matter under the test articulated by Alice."  Apparently, the court was not aware that the Federal Circuit stated in its plurality opinion in Alice that the " presumption [of validity] applies when § 101 is raised as a basis for invalidity in district court proceedings."  For this reason alone, the court committed reversal error.

Second, the court held claim construction is not necessary, so there are no factual issues to be considered--even though the claims recite means-plus function elements and other complex terms. So that made it too easy to simply boil down the claims to alleged abstract ideas:  "We agree with Defendants that claims of the ‘789 and the ‘291 Patents are directed to the same abstract idea: monitoring locations, movement, and load status of shipping containers within a container-receiving yard, and storing, reporting and communicating this information in various forms through generic computer functions."  Physically moving 20 ft. shipping containers and tracking their locations in a physical shipping yard is abstract? In what universe, two-dimensional Flatland? Contrary to the implication of Allvoice, the court did not consider the "tangible" aspects of shipping containers, yards, and other physical aspects of the claims to have any significance:  "Although Plaintiff is correct that the claim incorporates tangible components into the monitoring system, this is not enough for the system to become patent-eligible subject matter."  This decision is not even wrong.

Two other notable decisions are Mobile-Plan-It LLC v. Facebook Inc. and Good Technology Corp. v. MobileIron, Inc.  In Mobile-Plan-It, the court denied Facebook's motion to dismiss.  The patent dealt with organizing meetings for attendees at a conference by using temporary proxy email addresses so that they can communicate with each other without revealing their actual email addresses.  I attended the motion hearing and I have to say that Facebook had the upper hand, with a powerful argument that the claims were nothing more than messaging through proxies--even citing Cyrano de Bergerac--and dominating the "time of possession" for the lectern. Nonetheless, the court was willing to give Mobil-Plan-it the win: "while the question is close, MPI has offered at least an argument that its patent offers more than a list of steps human beings were routinely carrying out long before the patent existed."  

Finally, there is Intellectual Ventures I LLC v. Symantec Corp..  I mention this case because of a personal connection:  the court invalidated U.S. 6,073,142, a patent I wrote back in 1997 for Park City Group, the software company that did all of the enterprise software for Mrs. Field's Cookies.   I had no idea that the invention was merely an abstract idea.  I guess it's back to the drawing board...