Julian Sanchez header image 2

photos by Lara Shipley

Patent Impending

June 12th, 2007 · 11 Comments

I’ve been remiss in failing to link to Tim Lee’s excellent New York Times op-ed on software patents, but at least my sloth means I can simultaneously links his sharp response to some early criticism. The latter post makes a perceptive Hayekian point which, alas, suggests it should be difficult to get a sound empirical resolution to this debate: Since the stifling of innovation is typically invisible (if you had expected it to be there, it wouldn’t be innovation) it’s very difficult to figure out how much there “ought” to be at any given time, or whether your policy is constraining it. You can point to a few clear-cut cases like Verizon/Vonage, ut the clearer the rules are (however inefficient), and the cannier the lawyers, the fewer such cases you’d expect to make it to litigation at all.

Tags: Tech and Tech Policy


       

 

11 responses so far ↓

  • 1 Dave W. // Jun 13, 2007 at 8:03 am

    I haven’t read the links, but being a patent lawyer, this is something I have thought a lot about. Here is one way to get empirical resolution on what is obvious and what is not:

    1. Put the burden on the patent applicant to keep its idea confidential for, say, 2 years after applying.

    2. If somebody else comes up with the same thing independently within the two years, then it was an obvious idea. If they don’t its not.

    While this may sound far fetched, it really isn’t. In a typical case, it is many years between conception of an idea and its instantiation in the market. It is even longer until the market reaches a peak level (10 years from conception to large scale market is a good rule of thumb). Which means that 2 years of secrecy at the beginning is not such a far fetched idea, especially since you can release to investors under NDA. Many inventors already do keep their ideas secret in the 18 months between filing and initial publication anyway already.

    The empirical beauty of this model should be readily apparent. If the idea was that manifest, with due consideration of research allocations, human knowledge, market prognostications and the like, at the time the inventor invented it, then somebody would do it independently within two years. This model also gives plenty of incentive for early defensive publications. If you don’t want to pay the patent attorney, then at least get your drawings and concepts out there early to defeat anyone who may have beaten you to the punch by a year.

    A few legislative tweaks need to be made to patent law to accomplish this improved model, but fewer than you might think.

    Hope to get time to read the links today. Normally patent law articles infuriate me because they all sound like they are written by cynical toddlers.

  • 2 Dave W. // Jun 13, 2007 at 7:20 pm

    =======================================================

    And sure enough, Tim Lee asserts without evidence that translating an ip number into a phone number when the Verizon inventor did it. Sure it is obvious now, in 2007, but whether it was when the Verizon inventor did is another question, and one that should not be left unaddressed in an article like Mr. Lee’s.

    I would still say that if the idea was obvious, then someone would have published it within a couple years of the Verizon inventor filing on it. If no one did, then it was one of those ideas that was not obvious at the time, but only is now that so many people have computers and stuff. If it is true that the patent inventor was first by at least a couple years, then there are a couple competing guesses about what is happening with her (impressive) contribtion to human knowledge:

    1. One possible read is that behemoth Verizon is only investing in Internet telephony at this early juncture because there was a good patent around for it to purchase (somehow I doubt the inventor worked for verizon when she did the actual inventing). In that case, the patent system is doing what it is supposed to and the original inventor staked a claim precisely so that a company with the wherewithal, like Verizon, can bring in the heavy capital and mine with a sense of security, and a diminished sense of risk about entering a new business.

    2. Another possible interpretation is that Verizon purchased the patent from the lonely pioneering genius so that it could stay out of the area of Internet telephony, and keep others, like Vonage, out, too. that way we stay with the Dinosaur technology of cell phones and land lines. In this case, the patent system is counterproductive. However, the best remedy to this type of problem is a working requirement or compulsory licensing. As Mr. Lee suggests, sometimes it is good to look to copyright — and there is compulsory licensing all over that intellectual property law domain. It is not a cause to be nihilistic about software patents in general, as Mr. Lee seems to be.

    To summarize the main points: Mr. Lee’s article is too conclusory / cavalier / uncurious about: (i) whether the inventor really was ahead of the pack; and (ii) how Verizon is using the rights afforded by the presumptive invention.

  • 3 Dave W. (correction to previous) // Jun 13, 2007 at 7:23 pm

    And sure enough, Tim Lee asserts without evidence that translating an ip number into a phone number when the Verizon inventor did it is obvious.

  • 4 anonymous patent attorney // Jun 14, 2007 at 12:50 am

    Tim Lee’s point is that software patents are a bad idea. I agree as a matter of policy. Software patents are not driving innovation in the software industry like pharmaceutical patents do for the drug industry. If pharmaceutical patents didn’t exist, drug companies would stop doing R&D. Patents are just an afterthought in the software industry.

    The granting of any patent reduces utility. The real issue is whether the spur of the patent system towards innovation makes up for the reduction of societal utility that results from the granting of patents in a given art area. In the software area it doesn’t because patents are not much of a spur to innovation in this area. The issue of obviousness is somewhat of a red herring.

  • 5 Dave W. (correction to previous) // Jun 14, 2007 at 6:34 am

    ==========================================================
    If pharmaceutical patents didn’t exist, drug companies would stop doing R&D.

    How do you know?

    It should also be noted that pharmaceutical manufacturers get gigantic premiums for their patents, which means that there is less access to pharmaceuticals than there otherwise would be.

    It also means that pharmaceutical research not subject to patent (eg, vitamin d and cancer) is woefully underdone.

    In the software area it doesn’t because patents are not much of a spur to innovation in this area.

    mp3’s? I know I use’em, and they have certainly spurred a big market in players and online distribution of music and the like. I am not sure what investments Verizon is making in Internet telephony because Tim Lee ignores this issue, as do most people.

    Software may be less susceptible to patents because it may be more akin to writing a book than inventing the lightbulb, in the main, but that doesn’t mean that there are no bona fide inventions in the software area.

    The issue of obviousness is somewhat of a red herring.

    Say party A invents a helpful drug, and one year later, independently of Party A, party B invents the exact same drug. However, party A went diligently to the patent office and has a patent issue at the same time the drug goes to market. Society gets a terrible deal because they pay monopoly prices for ~14 years on a drug that party B would have been willing to give at market prices, albeit starting one year later. 14 years of monopoly pricing for a year of lead time is a lousy deal.

  • 6 Dave W. // Jun 14, 2007 at 9:55 am

    ==========================================================

    Also, people are more accepting of software than drug patents because of transparency differences. As 2 examples:

    1. If the portion of monies dedicated patent royalty payments appeared on people’s health insurance bills as a seperate, itemized line item, then people would not be so accepting of the gigantic royalties paid in this area.

    2. If the software sector worked like the pharmaceutical sector, then Vonage would have to basically get pre-cleared before entering the market, instead of entering the market first, and then seeing whether there were any patents it needed to license. In other words, they would have either taken care of the patent situation, or else you would have never heard of them (and would probably not feel that sorry for some company you never heard of). This is not to say that the software sector should work like the pharmaceutical sector, but it means that Vonage gets to manipulate our heartstrings a lot more than some would-be generic drugmaker who looks in the Orange Book and goes, “awwww shoot.” At least Vonage has a good opportunity to buy into the ip telephony patent, especially if it can establish a decent showing of invalidity for duhness. If ip telephony were a drug, there simply would be no Vonage and no one with much incentive to attack “duuuh” patents.

    While these transparency differences make a lot of difference as far as how people (even some patent attorneys apparently) dole out their sympathies, they are probably misleading when one does hard economic analysis that goes beyond looking at whose stocks are doing well.

  • 7 anonymous patent attorney // Jun 14, 2007 at 7:47 pm

    >>If pharmaceutical patents didn’t exist, drug companies would stop doing R&D.

    >How do you know?

    because

    > pharmaceutical research not subject to patent (eg, vitamin d and cancer) is woefully underdone.

    >>The issue of obviousness is somewhat of a red herring.

    What I meant by this is that, it can be a good idea to not to allow patents in an given area even for non-obvious inventions.

    >While these transparency differences make a lot of difference as far as how people (even some patent attorneys apparently) dole out their sympathies, they are probably misleading when one does hard economic analysis that goes beyond looking at whose stocks are doing well.

    High profits in a sector because of patents could mean the optimum level of patent protection for that sector should be lower in some manner. I am just concerned that removing the patent protection for pharmaceuticals without having some other incentive system in place is asking for trouble.

    I am not worried about the removal of protection for software patents. Copyright is good enough.

  • 8 Dave W. // Jun 15, 2007 at 7:09 am

    ==================================================================

    I am just concerned that removing the patent protection for pharmaceuticals without having some other incentive system in place is asking for trouble.

    And I am suggesting that all the pharmaceutical research is actually bad for the economy, counterproductive.

    Sure it is good for pharmaceutical stocks, but there is more to the economy than that. If you look at what the US spends on its GDP, and how much its life expectancy is increasing, you might believe that the US is wasting a lot of money on (ineffectual) medical research, and that that money might be better allocated to technology in some other area (think of the energy we would save if it were possible to have a decent, cheap video conference call over the Internet).

    In other words, people pour money into the pharmaceutical sector precisely because patents are so strong. But simply pouring money into a sector is not a good thing. If it were, then we could simply have the government fund research.

    What is a good thing is to incentivize the special lady or man who is years ahead of the pack, ahead of the “demand kurv” in what’s hot in research today. Maybe the “Verizon inventors” were such people. Maybe they weren’t. It is hard to tell without a lot more background than Tim B. Lee gave us. First off, I doubt they were working for Verizon when they invented. Chances are, they were woring in the hope that someday they would have something of value to sell to Verizon. If that is the case, then the only reason they were working early (as opposed to now) is because of the patent system.

    Tim B. Lee used to invent helpful intenet related things, but had a thing against patents. He was ahead of the pack, at least as a software inventor if not as a physicist. Now he doesn’t make Internet related inventions anymore.

    HOW OBVIOUNESS HELPS

    Tim B. Lee is one problem the obviousness requirement patent system was designed to redress: the talented individual who could not capture a nice income stream from his intellectual work to keep him working in the field where he is talented. In a world where not everybody is non-obvious, the non-obvious inventor makes more money for his employer.

    The economy choking problem of pharmaceutical patents is another problem that the obvious requirement was designed to redress. The idea is that if money is being pumped into a sector such that a large pack is coming up with the same things at the same time, then there is no need to provide further incentive to do research in that sector at the margin. It is only when there is not a large pack devoted to a given research area that, systematically teasing out everything that can be teased, that patent incentives are helpful in giving the economy (consumers, that is, not corporations) something it would not otherwise have.

    I think obviousness law has stopped accomplishing the above two objectives, and the lack of decent obviousness law is causing a lot of people to lose confidence in the whole patent system. the patent system was never intended as a means to help the Dow Jones. It is meant to give consumers things they wouldn’t otherwise have, regardless of whether that helps short run stock prices. In fact, the patent system (with a good obviousness requirement) is intended to make up for the fact that true innovation, fast innovation, some d00d way ahead of the pack type innovation will generally hurt short term stock prices, especially when the exchange is dominated by large, stodgy firms.

  • 9 Tim Lee // Jun 17, 2007 at 2:21 pm

    Just for the record, I’m not the inventor of the Internet. As for Dave W.’s other points, when you have 600 words, you necessarily have to leave out a lot of details. You can read my analysis of the Verizon patents here.

    As for your patent reform proposal. It sounds like a perfectly reasonable proposal, but it’s not really a response to my argument, which is that patents aren’t needed in the software industry. There are lots of ways we could tinker with the patent system to make it less harmful to the software industry, but given the overwhelming evidence that it’s doing more harm than good, wouldn’t it be simpler and more efficient to just get rid of it entirely?

  • 10 Dave W. // Jun 18, 2007 at 10:37 am

    =======================================================
    given the overwhelming evidence that it’s doing more harm than good, wouldn’t it be simpler and more efficient to just get rid of it entirely?

    I disagree with this overwhelming proof thing. As just one important counter-example, I don’t think Apple would exist without patents, and if Apple didn’t exist, then there would not be any competition in personal computers at all. I think the salient fact of the software world is that Microsoft runs it (for reasons having little or nothing to do with patent), and they discourage innovation. It is simple and efficient to have one company run things. It also means that computer software progress is pitifully slow these days. With Vista, it may actually be going backwards.

    Likewise, the problem with Verizon v. Vonage isn’t the patent. The problem is that Verizon has a lot more market power than it should, and has the anti-competitive luxury of being able to slow technology. I think you are blaming things on the patent system that are not ultimately the fault of the patent system (even with its traditionally lousy obviousness requirement).

    Final note: thanks for taking the time to respond, Mr. Lee. I have always thought of you as a famous guy, and also an influence on my own thinking, even though I got co-opted by Big Patent somewhere along the way.

  • 11 Tim // Jun 20, 2007 at 3:45 pm

    I don’t think I was clear enough about this: I’m a different Timothy B. Lee–not the guy who invented the Web. Anyway, on your substantive point, I would be very interested in an elaboration of this argument:

    As just one important counter-example, I don’t think Apple would exist without patents, and if Apple didn’t exist, then there would not be any competition in personal computers at all.

    A quick patent search reveals that until the late 1980s, almost all of Apple’s patents were hardware patents, which I find much less problematic than software patents. Apple had only a handful of software patents by the end of the 1980s, and none of them appear to me to be particularly crucial to Apple’s business. Which specific software patents do you feel were critical to Apple’s success? And why do you think Apple couldn’t have succeeded without them?

Leave a Comment