Software patents are…

Many software companies are anxiously waiting for the outcome of Alice Corporation Pty Ltd. v. CLS Bank International. This is the first Supreme Court case on the patentability of software since 1972’s Gottschalk v. Benson. Gottschalk v. Benson held a process claim directed to a mathematical algorithm was not patentable, that allowing it to be patentable would “in effect be a patent on the algorithm itself.”  Because software is a mathematical algorithm, the ruling was viewed as saying that software itself could not be patented. Cue the patent attorneys who got around this by claiming that the algorithm in combination with a computer programmed to carry out the algorithm was patentable. This left a gray area with little guidance to determine when the algorithm that is software is an abstract idea and not patentable and when it is a process implementing the idea and is patent eligible. Forty-two years later the Alice Corporation case has the potential to clear this up by providing clear rules to determine under what circumstances a computer implemented business method invention is abstract and not patent eligible.

The patents at issue in Alice Corporation are about mitigating risk in currency transactions by ensuring one party won’t default in performing after the other has already performed.  The process uses an intermediary, an escrow agent, to hold funds. That’s right. This is about an escrow.  They want to patent an escrow. A computer that facilitates interactions between the escrow agent and the parties implements the escrow. Because the computer implements the escrow the argument is the escrow concept is patent eligible. Simply using a computer shouldn’t be enough for an abstract thing to be patentable, just as an algorithm, a law of nature, shouldn’t be patentable because a computer implements the algorithm.

Alice Corporation is about business method patents. A business method patent usually involves combining software automation with one or more business methodologies. With 41% of the patents asserted by NPEs between 2011-2012 being business method patents and nearly 2,000 of the patents asserted in 2013 by NPEs being software patents a narrow test in Alice Corporation could dramatically reduce the litigation by NPEs by invalidating almost half of the portfolio they assert .  I do not think the ruling will go that far and at most they will invalidate the patents without providing a test, or at least a helpful test, to determine what computer implemented inventions are patent eligible. That would leave us with the same gray area we’ve had since 1972.

While the Alice Corporation case is going on there around a dozen proposed acts intended to reform the patent system by protecting end users, curbing abusive troll (oops NPE) strategies, and making it easier to get the USPTO to look again at overly broad patents that have already been granted.

All of these proposed regulations are aimed at reducing the damage done by patents, primarily software patents.  But (Please note: I am not a patent attorney, I do not even have the undergraduate degree necessary to become a patent attorney and again this blog is always my personal opinion not that of my employer) none of these proposals would ever fix the problem because software shouldn’t be patented at all just as I think Gottschalk intended. That is why I personally hope Alice Corp. comes out with a clear and narrow test that invalidates most software patents.

If I go to my magic 8 ball that is Google® and type “Software patents are…” the below auto fills.



I’ll take it a step further and say software patents are killing innovation.

The purpose of patents is to encourage the development of new inventions.

Let’s take a look at this from a software patent perspective.

The term on a patent is 20 years.  Someone please tell me a single piece of innovative software created in 1994 that hasn’t changed in 20 years.  Yeah. And so what happens is we have these patents written for inventions from 1994 with claims that the patent holders try to contort to apply to inventions today.  And then those “infringing” on the patents have to pay the patent holder (because it is almost always cheaper to settle than fight even if you aren’t infringing) for the right to use technology that wasn’t even a glint in the patent holder’s eye 20 years ago. It doesn’t seem to me that taking money out of a technology company’s budget to pay license fees on ancient technology is encouraging the technology company to innovate.

To obtain a trademark or copyright you have to submit a specimen of what you’re trying to trademark or copyright. Not so with patents. How does allowing something to be patented that you don’t even have to bring to life encourage the development of new inventions?

And the cost to obtain a patent at the very minimum is between 5-10K. I’m sure there are plenty of smart college students and very small companies out there with innovative and novel ideas they can’t patent because they don’t have the cash and no company to back them, and so the bigger companies with cash to patent (or license a patent) have a monopoly on the technology. Monopolies do not equal innovation.

Just how much profit are these patents bringing to technology companies anyway? One study analyzed data on publicly traded companies from 1976 to 1999 (the last year with enough data to do the study) and found that in 1999 patent profits were 9.3 billion but litigation costs were 16 billion 2. With 5,189 patent actions³ filed in 2012 compared to 2,318 in 1999 it is safe to assume the disparity is even greater today. So patents are causing a major loss in revenue, not bringing profitability. That doesn’t seem very encouraging of innovation either.

At least the patents granted are valid right? Wrong.

In 2012 92% of all patent reexamination requests to the USPTO were granted ⁴. Of the 92% reexamined 70% of the patents granted by the USPTO were invalidated!

Ok. If they’re patents that never should have been granted at the very least what is claimed as patentable should actually exist. No made up stuff can be patented right?

Wrong again. The USPTO granted a patent written by a Korean scientist for cloning human embryos ⁵. Originally the USPTO rejected the patent because it didn’t teach how to perform this cloning technique. Because, you know, it’s made up. The Korean scientist pressed forward by providing an affidavit saying he didn’t need to teach how to clone human embryos, that he had actually done the things he claimed and that cloning humans was within the skill of an ordinary person in the art. Go ahead snicker. But guess what?  The USPTO gave its stamp of approval on the patent. When called on the carpet the response from the USPTO was the patent is “definitely not an assertion by the U.S. government that everything he is claiming is accurate.” That’s great.  The USPTO is ready and waiting to approve your teleportation device. If a patent for cloning humans was approved I can only imagine the crazy software patents that have been approved.

What leads to the high invalidation rates and patenting made up stuff?

Because of the hundreds of thousands of applications per year on average a patent examiner can only spend 16-17 hours per application, often spread out over several years, to determine whether an application should be granted ⁶.  I spend more time than that per month on some contracts! There is no way to conclude a piece of software is novel and non obvious in only 16-17 hours and so all too often I think patent applications are approved because the examiner did not have enough time to determine whether or not it was truly patentable.  And so imaginary and/or obvious and/or not novel inventions are stamped as approved.

Patent examiners of course are skilled in the field. But how skilled in the field? Money draws talent. The salary rate table for a patent examiner is available at and shows as of 1/01/2014 the salary for a patent examiner ranges from $42, 389.00 to $157,210.00.  The average salary for an engineer in the United States 2013 was $104,303.00 7. With the below market starting pay my theory is the highly skilled engineers either don’t even start work as a patent examiner or don’t stay employed as one very long.  The USPTO needs those highly skilled engineers especially because of having to make judgment calls on patents in so few hours. Between the little time they have to review the patents and the examiners possibly not being as skilled as they could be I think this is why there is such a high invalidation rate and things like cloning of humans wind up approved.

All of these factors lead to the granting of bad patents, which in turn gives NPEs their ammunition. NPEs in turn prey on companies that actually provide innovation to the world because those are the companies the NPEs know will be forced to license. NPEs’ business model is legalized extortion; technology companies’ business model is innovation.  And yet it is the NPEs whom the patent system benefits.

My guess is most of the companies that patent software either do it defensively to assert if they are sued or they are the Microsofts of the world and patent to have a monopoly. The patents aren’t held to encourage innovation. They’re held as ammo in trench warfare.

The obvious question is, how do we protect our software is we can’t patent it? I’ll throw out copyright as an option. It’s not a perfect fit but using patents as a way to protect IP in software seems to be doing far more harm than good. The most likely objection to copyright on software is probably that copyright does not prevent different expressions of the copyrighted material. That’s called encouraging innovation.


“The NPE (“Patent Trolls”) Minefield. ” IPEG Intellectual Property Expert Group. Web. 2014.

2 Fitzgerald, Michael. “A Patent Is Worth Having, Right? Well, Maybe Not.” New York Times. Web. 15 July 2007.

³ Price Waterhouse Coopers LLP, 2013 Patent Litigation Study. PWC. Web.  June 2013.

⁴ Masnick, Mike. “Why Do We Assume Patents Are Valid When Patent Office’s Own Numbers Show They Get Things Wrong All The Time?”  Techdirt. Web. 20 August 2012.

⁵ Salzberg, Steven. “US Issues Patent For a Fraudulent Human Embryonic Stem Cell Method” Forbes. Web. 24 February 2014.

⁶ Lichtman, Doug. “Patently Obvious.” The New York Times. Web.  15 April 2011.

7 Sethi, Chitra. “2013 Salary Survey: Engineers Ride the Wave.” ASME. Web.  January 2014.


Leave a Reply

Your email address will not be published. Required fields are marked *