Monday, May 22, 2006

Do patents tend to harm inventors? Part 2.

Mike at Techdirt is still writing me about my post on IPBiz about his claiming: patents tend to harm inventors more than help them. After I showed, contrary to his statements, that the Bessen article was not new, peer-reviewed, or ignored in the patent community, he replied:

I don't care about whether it's new or whether it's peer reviewed. I care about what's in the paper -- and in this case, it's a detailed explanation of how innovation can often be much greater in a world without patents.

Well, Mike, the Bessen paper is based only on comments about technology in software, semiconductors, computers, and the paper is based on a completely FALSE PREMISE that patent protection was extended to software in the 1980's. Patent protection has never been extended to software. [point elaborated below] Needless to say, the Bessen paper had no citation whatsoever for the proposition that patent protection was extended to software in the 1980's. The issue of business method patents, which is a different topic, did not come up until the State Street case in the late 1990's, but Bessen's empirical data showing no increase in research funding is for a time period before State Street. Thus, "what's in the paper" is a false statement of the law, and an observation about research spending, changes in which could occur for any number of reasons. Thus, there is no detailed explanation of how innovation can often be much greater in a world without patents.

Bessen also said: Pure software companies have not applied for many patents. Bessen mentions that Microsoft had 39 software patents in 1995. Mike, check out IPBiz for what Microsoft is patenting these days. The Bessen paper is out-of-date as to this century.

Bessen concludes: Imitation invariably inhibits innovation in a static world. The assertion is not supported by the data he has. In the static world, the existence of a patent precludes direct imitation but invites designing around the patent. The concept of "design-around" patents is well-known, wherein the existence of an initial patent fosters innovative improvements which are based on the initial patent.

***from the previous IPBiz post-->

The text --patents tend to harm inventors more than help them-- appears in techdirt. Working through the maze of links, I'm still not sure "where" (or "if") the original peer-reviewed article is published. IPBiz assumes no responsibility for the gibberish in the statements below.

from techdirt:

However, there's another, related, point where the conversation often breaks down: the idea that innovation is an ongoing process, rather than a single event. Supporters of the current patent system seem to assume that innovation is a single event -- and that event needs to be protected with the patent. The reality, however, is that innovation is an ongoing process. That's why we've said in the past that the idea of a sustainable competitive advantage is a myth. Companies compete by a series of fleeting competitive advantages. That's what innovation is. Even if your competitors can copy you, you keep on innovating and keep the lead in your market. The claims from strong patent system supporters always comes back to the idea that, without patent protection, others would just "steal" their idea. But, that's only true if the original innovator stops innovating and tries to rest on its laurels. That's not good for anyone -- but it's exactly what patents allow. In fact, there's increasing evidence of this exact problem with the patent system. David Levine points us to a new peer reviewed paper on "sequential innovation", which looks into this very issue and finds that, since innovation is such an ongoing process, patents tend to harm inventors more than help them. The report shows that if (as patent system supporters believe) innovation is a static event, then the patent system does make sense. However, since it's an ongoing process, patents tend to hurt more than help. In fact, they find that inventors tend to profit more without patent protection, but by focusing on the increased competition and the ongoing innovation. While they may not have been able to profit as much from the single event of invention, the increased innovation as an ongoing process means they end up profiting more in the long run by supplying increasingly better products to a growing market. Meanwhile, in cases where patent protection was increased, the research shows that investment in innovation actually decreased. This makes sense to us, but it's good to see it supported with research (which, if history is any guide, patent system supporters will ignore). However, it does help highlight how much of the disagreement may simply be due to a lack of understanding concerning the difference between the event of invention and the ongoing process of innovation.

It is false that --Supporters of the current patent system seem to assume that innovation is a single event--

It is false that --Even if your competitors can copy you, you keep on innovating and keep the lead in your market.-- Ask the Wright Brothers about that one. Without patent protection, Glenn Curtiss and friends would have eaten them alive. [Even with patent protection, the Wright Brothers did not fare well.]


Stephen G. Kunin and Bradley D. Lytle wrote in 87 JPTOS 991 (Dec. 2005):

With that the Benson decision (in 1972) brought advancement of software patents to
a grinding halt. Following that, software programs were routinely dismissed as
mathematical algorithms that the United States Patent and Trademark Office (PTO)
refused to patent. Most software inventors either quit seeking patent
protection or disguised them as computer hardware


The Court in Diamond v. Diehr held that the only question was whether the patent claim as a whole was directed to an otherwise statutory process regardless of whether a mathematical algorithm was employed for performing a function (curing rubber) that the patent laws were designed to protect. If that's the case, as was the case in Diehr, then the claims satisfy 35 U.S.C. ยง 101 and are patentable subject matter.

[The mathematical algorithm at issue in Diamond v. Diehr was the use of the Arrhenius equation of chemistry directed to a method of curing rubber, and not what be considered a "software" patent or a business method patent. IPBiz doesn't think Microsoft, Oracle, and Sun are concerned with the use of chemical principles directed to chemical methods. Moreover, the use of a hundred year old piece of chemical knowledge hardly qualifies as a "software invention." ]


The next important case was decided 13 years later (1993), where, in In re
Alappat, n13 the Federal Circuit court ruled that a new and useful computer
software is patentable subject matter if it has a practical application that
produces a concrete, useful and tangible result.

***The State Street case was decided in 1998. Although I don't necessarily think that there weren't "business method" patents before State Street, I do think the State Street case highlighted to the public the potential patentability of such things. However, as to Bessen's contention, State Street was decided in 1998, not in the 1980's, and there wasn't any big push for "software patents" in the 1980's.


Post a Comment

<< Home