Just in case the subtitle isn’t clear, I’ll invite you to review the twitter conversation where Gene said I was not worth arguing with, then proceeded to engage in the conversation with five more tweets. I’m not sure if he can’t count, or if he just can’t help himself? The point was, he asked me if I wanted to debate whether IBM’s Watson should be patent eligible and I told him that I was far too wordy to do that on Twitter but that I’d be happy to respond on the blog.
And here we are.
I’m coming at this issue of patentability of Watson and of software in general not as an attorney, not as an engineer, not as a patent agent, nor as a government official. In fact, other than making good on a Twitter promise, I don’t have a dog in this fight. I’m just a girl, standing in front of a boy, asking him to love her. Wait, what?
I see this issue from a business standpoint, and as someone who has a background in computer programming/software development. I’ve been doing it for over 20 years now so I think I have more than a passing understanding of what computer software is and does, and how it’s created. I spent the better part of my 20’s in little hidey holes banging out line after line of code. It isn’t as if I’m coming completely out of left field here, is my point.
The question of should Watson be patent eligible is moo…IBM does in fact own several patents for the technology behind the service. So the question is more rhetorical in nature. Scholars at every level have been tackling this question for in excess of two decades so the idea that I can work this out in a blog post or twitter conversation is hugely unlikely.
Nevertheless, you asked (and so nicely, too!) and I will proffer my answer.
Lines of code or algorithms are not, in and of themselves, useful. From the US Constitution, Patents are issued:
to promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries.
To promote the “useful arts”, not to promote “really cool code” or “computers that can win Jeopardy”. No, the intent as I read it was to protect those who spend their time and money developing something new, non-obvious, and useful. Yet how useful is Microsoft Word if no one ever creates a document with it? See where I’m going with this?
Using patented technology, IBM is promoting the useful arts. They’re giving away some of the information that Watson’s powerful computing has generated from an article here :
In a demonstration of its power, Watson has 4.7 million patents and 11 million scientific journals published between 1976 and 2000. In each article Watson carefully identified any previously unseen chemical compounds (in pictures or text), grabbed any related diagrams, grabbed relevant keywords, and lastly scooped up the author and company names.
The result was a database of 2.5 million compounds, which are thought by Watson to be unique. For each of these compounds Watson discovered the earliest patentee. IBM donated its superbot’s work — the “open chemistry” database — to the U.S. National Institute of Health (NIH), allowing scientists all over the world to dig into it.
But again, it isn’t the lines of code or the circuitry or the machine itself that makes Watson useful. It’s what comes out of it. This is why I don’t like patent trolls, incidentally. They don’t make anything, they just buy patents and extort money from people by threatening to sue for infringement even, and especially, when there’s no true evidence that any such infringement took place. But I digress.
I’ve spent the better part of a week reading all about Alice Corp vs. CLS Bank and the numerous briefs (brilliantly curated here by Dennis Crouch) surrounding it. I’ve read countless articles about Watson itself (himself?) and this one is the best. And in all my reading and Googling do you know what I discovered? This is a multi-layered and complicated question with no clear answer. Hell, if this statement via Wikipedia is any indication:
The fractured panel of ten judges issued seven different opinions, with no opinion supported by a majority. Seven of the ten judges upheld the district court’s decision that Alice’s method claims and computer-readable-medium claims were not patent-eligible, but they did so for conflicting and incompatible reasons. Five of the ten judges upheld the district court’s decision that Alice’s computer-systems claims were not patent-eligible. The panel did not agree on a standard to determine whether a computer-implemented invention is a patent-ineligible, abstract idea.
then there is not much hope that I can add thoughtfully to the discourse except to say this: I think figuring out what is patentable and what is not has become so unbelievably complicated because complications mean lawyers who are necessary to decipher what’s going on and we all know that in any dispute, patent or otherwise, the real and clear winner is always the attorney. They didn’t do it on purpose. I think complication is a natural offshoot of people who are, by nature of their vocation, extremely risk averse and also? Happen to bill by the hour. They spend all their time looking for any loophole to any given statement (either on the loophole creation side or the loophole avoidance side) so they dream up crazy tests and rules and talk their clients into suing each other over them.
But just as an aside, regarding Alice Corp v. CLS Bank, what kind of crazy-making is this, anyway? Alice Corp creates “a computer-implemented, electronic escrow service for facilitating financial transactions”. From their own website:
An Alice Market enables end-users and investors to create flexible contracts that meet their hedging and investment needs in a real-time, secure and anonymous electronic market. It also ensures that participants are not exposed to the possible failure of the parties with whom they contract.
Is it possible to “ensure that participants are not exposed to the possible failure of the parties with whom they contract” without an Alice Market? If what CLS Bank was doing uses the exact same methodology and code that your patented software does, then I suppose you’ve got them on infringement. It doesn’t seem that hard to me to figure out…CLS is either using your software methodologies and code and algorithms or they aren’t. As a coder, it’s pretty easy to figure that out so honestly, I don’t get what all the fuss is about in terms of whether or not they’re infringing. Or maybe I’m stupid and am missing something really obvious, in which case I have zero doubt that Gene someone will let me know.
Either way, this isn’t really what the case is about anymore, it’s about can you patent software? Nee’, is IBM’s Watson patent-elibible?
My answer? Like I said on Twitter…code that simply replicates business rules should not be patentable. But for the broader question of all software in general, I don’t honestly know.
You can bet, though, that I’ll be waiting on the Supreme Court’s answer to the question like some people wait for a shoe sale at Nordstrom’s.