Debate On Software Patents Fails To Convince Silicon Valley That Patents Increase Innovation

from the not-a-surprise dept

Yesterday, I went to the Computer History Museum in Silicon Valley for a lunchtime debate on software patents. It was one of those things where an official "motion" was put forth, and then people voted before and after to see if the debate changed anyone's mind. In this case, the motion was simple: "software patents encourage innovation." Arguing for the motion was Bob Zeidman, a consultant/writer who sells software for... analyzing intellectual property. Arguing against the motion was Edward Lee, a Berkeley CS professor. The audience, as is (unfortunately) typical at the Computer History Museum definitely skewed "older." I ended up at a table with some guys who worked at IBM many decades ago. There were also a fair number of patent lawyers in attendance.

There's this view outside of Silicon Valley that folks here love patents, but in my experience, most people don't seem to like patents at all. They get in the way of actually innovating -- as we've shown time and time again -- and the thing that Silicon Valley folks like more than just about anything else is building something cool. The idea that someone else can sue you for building something cool just seems incomprehensible.

Before the debate, the vote was already against the motion: 64 people voted for the motion (35.7%). 89 people voted against (49.7%). Another 26 didn't vote at all.

The debate itself struck me as rather tame. Niether side really made as strong an argument as they could. Obviously, given what I write about here, and the rather staggering evidence that's out there that the patent system tends to do a lot more harm than good, I was inclined to support Lee "against the motion." But even there, it felt like he could have made much stronger arguments. For the most part, he just repeated claims from James Bessen and Michael Meurer's book Patent Failure. It's an excellent book and completely worth reading, in large part because it cites dozens of studies to show how the costs of the patent system greatly outweigh its benefits. It's definitely a good starting point, but there's a lot more out there that let's you take their argument further, and I was a bit disappointed that Lee didn't do so (though the debate format is a tough one). For example, Lee repeated a few key claims from Patent Failure, such as the idea that patents don't work well as property, because the reason "property" works is because there are clear boundaries. But, with patents, the boundaries are (often intentionally) unclear, meaning more and more litigation. Good property rules should mean less litigation.

But all that presumes that property rights make sense around ideas. And there's a strong argument that they don't. I also would have liked at least some more discussion about the fact that so many patent lawsuits involve work that was independently invented. Lee mentioned it in passing, but it's a key point in this debate, and it wasn't highlighted nearly enough.

Lee did get in a few good one-liners, though, which the crowd appreciated. He noted that in Silicon Valley today, if you actually do something well, you're guaranteed to get sued for patent infringement. Separately, he argued that patents tend to encourage more patents, rather than innovation. The one point he "conceded" was that patents did seem to encourage "innovation" in one area: Zeidman's arena of writing software to analyze patents.

The one point he made that struck me as compelling and not discussed nearly enough was his response to the common claim that patents are all about "disclosure," and without patents everything would be kept as a trade secret. There are a bunch of good responses to this that we've discussed in the past, but he noted (1) that patents don't seem to disclose much that is useful to a software developer and (2) that the real purpose of "disclosure" is education, and there are better ways to educate software developers. I hadn't made that connection between disclosure and education before, and it's a good point.

As for Zeidman, not surprisingly, I found his arguments completely uncompelling. He kicked off with an attempt to play to the ego of the crowd, asking them how many had programmed stuff, and how many of those people felt that "just anyone could do that." He was a little surprised that some people said, yes, just about anyone could program what they programmed, and accused the audience of being too modest. But, really, this argument is a total non-starter. He used it to claim that creating software is a "creative skill." That's true, but it's meaningless. The patent system has nothing to do with encouraging creativity. And, more importantly, the very crux of this particular debate for many people is that software is already covered by copyright. And the whole "creative" part is the part that copyright covers. So what does that have to do with patents? I submitted that as a question, but the moderator didn't ask it. The whole argument made no sense in the framework of what was actually being debated.

The other thing that caught my attention was that Zeidman tried to brush off the argument of patent trolls suing the companies who actually innovate, by suggesting it's not really a big problem. He specifically stated that "less than 2%" of patent lawsuits involve patent trolls. That number struck me as ridiculously low, so I first asked folks on Twitter if that could be possible, and received back a few responses, including from Mark Lemley, who said the real number is between 25 and 40% depending on how you count. Separately, a few folks pointed me to a research paper by Colleen Chien, which suggests the number has been growing recently, and agrees with Lemley's basic range. For example, from 2006 to 2008, defendants sued by trolls represented 36% of all defendants sued.

After the event I went over to ask Zeidman about the 2% number, and he claimed that he actually thinks that estimate is too high. He mentioned that he cited the source in the Wall Street Journal, but the only citation I can find that seems to match is a claim that "roughly 2%" of patents end up in court. Perhaps he means that he used the 2% number from the Wall Street Journal somewhere else, but the fact that only 2% of patents end up in court (which seems ridiculously high, actually) has nothing to do with how frequently trolls are suing companies.

Anyway, after an hour-long debate, another vote was taken, and since a bunch of folks showed up in the middle, the numbers were higher across the board: 70 people were for the motion that software patents are good for innovation, 112 were against and 42 didn't vote. On a percentage basis, those supporting software patents dropped to 31.3% and those against rose marginally to 50%. Given the fact that there were a fair number of patent lawyers in attendance (one sitting at my table mocked the whole debate because no one defined software patents, and then argued that Professor Lee was probably a hypocrite, because while he contributed to open source software, he probably wanted to get paid for his book), and they seemed to be the most vocal in supporting software patents, it seems likely that of the actual developers in the audience, well more than 50% were against patents for software.

Filed Under: bob zeidman, debate, edward lee, innovation, patents, software

Reader Comments

Subscribe: RSS

View by: Time | Thread

  1. icon
    jilocasin (profile), 25 Aug 2011 @ 8:14am

    Patents served a purpose.... once (Re: ))

    Patents may no longer work, but I think once upon a time they did.

    If you came up with a new drug, a new bullet transport mechanism for a hand gun, a way to better separate the cotton seeds from the cotton fibers it was worth it to have your specify _exactly_ how you did it. In that way others could repeat/improve upon it. In exchange the government gave you the sole legal ability to control that technology for a couple of years. Sure beat having to constantly 'reinvent the wheel'.

    This is why you couldn't patent; nature, math, ideas.

    It had to be truly novel to someone _skilled_in_the_field_. No sense letting someone patent something that people would come up with anyway.

    It had to be described it in sufficient detail for someone skilled in the art to build it, using _just_the_patent_application_. No sense giving patent protection to something that wasn't detailed or complete enough for say another gun manufacturer to build.

    Combine that with the fact that technology progressed _very_slowly_, it wasn't that bad of a trade.

    Unfortunately many things have combined to make them more trouble then what they're worth.

    Technology progresses _much_ faster than it used to.

    Patents are allowed for 'ideas' rather than specific implementations. (i.e. 'using a computer...' to do anything we already do without one.)

    Patents are allowed for; nature (i.e. gene/existing plants), math (i.e using a particular sorting algorithm).

    Patents are allowed for things that are obvious or a natural progression to someone skilled in the field (pretty much anything that suffers from multiple simultaneous invention).

    And most damming, patents are awarded for things that purposely written so ambiguously that it's completely useless for the express purpose of allowing someone skilled in the field to replicate and build off of.

    Of course that makes it perfect for suing other over.

    As an example here's a portion of Oracle's '507 patent that it's using to sue Google over android.

    Claim 20:
    A method for acquiring and reviewing a body of information, wherein the body of information includes a plurality of segments, each segment representing a defined set of information in the body of information, the method comprising the steps of: acquiring data representing the body of information; generating a display of a first segment of the body of information from data that is part of the stored data; comparing data representing a segment of the body of information to data representing a different segment of the body of information to determine whether, according to one or more predetermined criteria, the compared segments are related; and generating a display of a portion of, or a representation of, a second segment of the body of information from data that is part of the stored data, wherein the display of the portion or representation of the second segment is generated in response to the display of a first segment to which the second segment is related.

    Tell me that's clear enough so that someone skilled in the art can replicate the _exact_ invention that the patent was granted for. This isn't a claim to advance the state of progress, it's one to retard the state of progress, while unjustly enriching lawyers.

Add Your Comment

Have a Techdirt Account? Sign in now. Want one? Register here

Subscribe to the Techdirt Daily newsletter

Comment Options:

  • Use markdown. Use plain text.
  • Remember name/email/url (set a cookie)

Follow Techdirt
Techdirt Gear
Shop Now: I Invented Email
Report this ad  |  Hide Techdirt ads
Essential Reading
Techdirt Deals
Report this ad  |  Hide Techdirt ads
Techdirt Insider Chat
Report this ad  |  Hide Techdirt ads
Recent Stories
Report this ad  |  Hide Techdirt ads


Email This

This feature is only available to registered users. Register or sign in to use it.