Despite suffering wicked withdrawals from a lack of black and whites, it’s good to be back home. Texas and I are going to have to talk about this 95 degrees in September thing, however, now that I spent time again in a place that has all four seasons instead of just two (hot and hotter). New York CIty? I love your Fall!
Back to work we go…it’s time to finish up the series from my interview with a former USPTO Patent Examiner with the most timeliest topic we discussed: software patents.
The sixth and tenth of my 10 questions were these:
6. What is your opinion on software patents, and the mantra that “computers do math and math is not patentable”?
10. Any general comments?
Software patents are all the rage now, or at least raging against them is. Your friend and mine Gene Quinn recently
had a conniption fit wrote a blog post about how “intellectually dishonest” we all are for thinking that software patents are a problem at all, which was kind of odd. It was a fairly technical post and long about the time people start inserting pictures of computer guts, my attention span tends to fade. Nevertheless, I’m certain he has valid points, I just don’t know what they are.
The majority of the patent blogosphere and Twittosphere (which is to say “Twitter”) has major issues with software patents, and my examiner was no different. He laid the blame at the feet of patent claims, where he said
Software patents are for a processor with instructions and you can have the instructions to be anything. Because the process is intangible you can write anything into the claims. It becomes a word-smithing exercise. It’s based upon elements which I don’t think should be patented.
It’s too difficult to ask an examiner, because of the limitations on them including that they only search the database and not, for example, technical manuals, to make that call. It’s a lot harder for the claims to be examined.
The second paragraph goes back to his statement that the corpus of knowledge that patent examiners search for prior art is too limited. And that’s an issue…patents are slipping through the process because people aren’t looking in the right place for prior art, or because the filing attorneys have done such a good job with the wording of the claims that it’s just different enough to exiting patents that it will not get any direct hits when the patent database is searched. That’s a problem that has solutions (hello, Article One Partners!)
As for the first part, I agree with him that it’s a word-smithing game. When you look at some of the patents that are being litigated, the “auto-complete” patents, for example, how is that not obvious? One of the points my examiner made was that something may be non-obvious based on the fact that it hasn’t been patented before, or based on the fact that the applicant’s high-dollar attorney is telling you it’s a novel idea and not obvious because that’s what he’s been hired to do, or because you just didn’t talk to the right people.
My point being that particularly in the case of software patents, one conversation with a programmer and you’ll learn that, seriously? Auto-complete has been around in one form or another since 1991 because why? It’s obviously a useful feature. So how is it all of a sudden now a patent that anyone who writes websites has to have a license for or they’ll get thugged out of existence?
That’s just dumb. Legal, but stupid.
If you don’t know who Martin Goetz is, watch this. What you’ll see is that the whole idea of software patents was born out of the fact that one company was giving away for free what another company wanted to charge for. I’d love to tie this into Mike Masnick’s thoughts on industry disruption, and how companies that write software could have found other ways to monetize what they’re doing rather than sue IBM and cause a ripple effect that years later, would bring start ups to their knees.
With regard to the tenth and final question of the interview, my patent examiner’s comments were a recap: give examiners access to a wider range of prior art search options to prevent bad patents from being issued, and find a way to draw attorneys into the job so that they can match wits and intellect with the applicants’ hired guns. He suggested that putting together a database of the text of technical manuals would be a good use of increased funding, and mentioned that Google was working on a project like that, or so he’d heard.
There you have it, straight from the horse’s mouth. The single biggest thing I learned is that Julie Andrews was right.
Let’s start at the very beginning, it’s a very good place to start. When you read you begin with A-B-C. When you sing you begin with Do-Re-Mi.
Issue good quality patents by having the right prior art search tools and the right mindset about what should be patentable and what shouldn’t (hint: not software), and you create an atmosphere where innovators can feel safe spending their years and their dollars dreaming up new products. The patent system was set up to encourage people to create something new and give them a time frame to be successful without the fear of others piggy-backing off their hard work.
Somewhere along the way, we lost sight of this basic idea, and it became all about the money and the weaponry employed by modern companies as they seek to wipe out the competition using any means available. I’m no kum-bye-ya (and hello? exactly how many ways are there to spell that??) hippie chick, I know that companies are not all going to just get along. But if we can fix some of the things that are going wrong at the USPTO, we can at least force a little bit of disarmament, no?