One of the best empirical economists working on patents is Petra Moser of Stanford. She has an extraordinary ability to find unusual data that answers really hard questions. Her most recent paper Compulsory Licensing with Allessandra Voena does just that.

There is - a frankly rather ridiculous view, one held only by lawyers for whom the distinction between 10 ^ -6 and 10 ^ +6 is invisible - that somehow compulsory licensing in a country will reduce innovation because it will be every so much more attractive to "steal" all those great foreign inventions. In this view - which crops up in these pages occasionally in anecdotal form - the great thing about patents is that it increases innovation by forcing people to "invent around" patents.

The "Compulsory Licensing" paper attacks this question directly - looking at an episode where the US "stole" a bunch of inventions by compulsory licensing after World War I. The consequent effect on innovation in the areas covered by the licenses? It went up by 20%.

In my view debunking the rather silly view that compulsory licensing would diminish innovation is not this papers main contribution. We recognize that patents have two effects (ignoring "invent around" and "revealing secrets" both of which are of at best minuscule significance): increasing innovation by increasing incentives to innovate, and decreasing innovation by making it more costly to innovate. This paper gives us a pretty clean measure of the latter effect: the benefit of being able to access existing ideas without negotiation or licensing raises innovation by around 20%.

Pipapics: A Case Study in Innovation by Glenn Thorpe

For the past couple of months I have been working on developing the pipapic concept and the tools to generate pipapics - you can find out about it at my website www.pipapic.org.

Here is the idea: Take a color cube. Pull out an evenly spaced palette. Place it on a screen canvas just the right size. Re-arrange the pixels as desired. Into say, the Mona Lisa, or into irises. This is a pipapic. So a pipapic is pixel art with constraints: call it the haiku of color. All pipapics have a well balanced brightness and colorfulness. Any image can be converted to many many pipapics that retain some semblance of the original. Some existing images have the color balance close to being a pipapic without any editing. Examples include some images of parrots and flowers.

The concept is probably patentable. My attitude is...well I read this blog every day, so I don't believe I need say more. But what I would like is for all pipapics to be virally free. So if anyone creates a pipapic it is free, full stop. I would really like to impose a moral patent, and say: I have the legal right under the current system to gain monopoly rights over the entire concept. I do not want monopoly rights nor your stinking money. All I want is freedom for me, for my kids, for your kids, for Chinese kids and for Talibani kids to use this as they wish. If you don't actively support the no intellectual property concept leave pipapics alone. Go. Now. There is nothing for you here. And don't use the pipapic balance in your fashion, in your makeups, in your colorful foods or any other place.

I am concerned about the legality of the pipapic tools. These tools take existing images, images that will often be "owned" by other people, and manipulate them. This is in fact even against some Creative Commons licence conditions on derivative works. By releasing tools to enable people to create pipapics I am enabling them to break the covenants imposed by copyright. It can be viewed as being worse than Napster. And to be blunt, I am seditious, I consider IP laws to be garbage and I am most definitely encouraging people to treat these laws with the contempt that I believe they deserve.

Copyright law is a huge problem. There are examples of pipapics that I have derived from original images. The original images have been well and truly butchered - large portions have been discarded to provide an image with the pipapic balance, the remaining portion has been reshaped and resized to provide a pipapic canvas, and then every picture has had at least 100% of the pixels recolored. I also display the original for comparison purposes. I consider this to be fair use. In the context of my website they are demonstrative and educational.

I'm eager to get examples from the public, and want to display those examples as well. These may also be modified from existing images that are under copyright. Do I have to request the pedigree of every example? What if the examples are collated from many sources? They could be doctored in many ways before I get them, including being modified to look like another image. Is it me who must devise the questions to be asked of any submitters? If I ask the wrong questions will I later have to discard the images until they are re-certified as meeting the new compliance standards?

Look at all the legal opinion needed just to release and promote a artistic imaging scheme. All in order to comply with undemocratic laws devised in an undemocratic country and imposed undemocratically on the entire world through secretly negotiated treaties. The rule is: Before you can start to be creative you must pay for a legal opinion. This is supposed to stimulate creativity and enrich the public. Surely you jest!

Truly fantastic speech by Jeff Tucker on the problems of intellectual property.

Who Owns You? Corporations Patenting Your Genes

A debate between IP opponent David Koepsell and patent attorney Gene Quinn about gene patents. As noted in IP Debate? (see also Gene Quinn the Patent Watchdog and Is It So Crazy For A Patent Attorney To Think Patents Harm Innovation?), Quinn previously said he wanted to debate an IP opponent, before transparently banning me from his site after I called his bluff; then David Koepsell offered to debate Quinn--it was supposed to be held yesterday at Cardozo law school, but then Quinn backed out, and finally agreed to the online thing you see above. Good thing for him he was not there face to face to present his embarrassingly weak "arguments."

[SK cross-post]

One of the most interesting things about patents is that many entrepreneurs I have talked to are unenthusiastic about them. You can make money by creating a product, building a business and developing customers. Or you can spend your time in court trying to grab money from someone else. Successful entrepreneurs tend to do the former. Richard Corsale is an entrepreneur, and writes to us his views about patents. His career titles include: software developer, project manager, CIO, CTO and CEO and Sandwich Artist. He is currently launching an Open Source software company with a value added business model. The concept centers around IP that is placed in the public domain (tofui.com). He is considering producing a mini-documentary about the launch of tofui and other companies that are embracing alternative business models.

Richard says: I have been an entrepreneur for as long as I can remember. I generally believe that whats good for the ecoystem of an industry is good for everyone in that industry. I view a robust public domain as crucial for intellectual advancement and cultural identity. Politically I suppose I lean left on social issues and center/right on fiscal policy.

He submits the following remarks about software patents:

Five Reasons the US Supreme Court should explicitly deny patent protection to software.

1. Software is simply a solution to a problem. Patenting the solution is pointless since it is protected under copyright. So what we end up with are patents that cover a conceived problem. Of course the problem does not exist until it is conceived, and thus it is abstract. Abstract ideas are not patentable subject matter and neither are algorithmic solutions. Software patents have nothing to do with software, rather with the purely imaginary problems that they solve. What we are left with is technical process or method patents. I will refer to these as Software patents throughout this document since that is what they were originally billed as. Applying exclusivity to a process (regardless of whether it involves a computer or not) is tantamount to oppressing creative freedoms.

2. The arguments in favor of software patents are based on a series of assumptions regarding innovation that do not translate to modern, digitally creative arts. Innovation in technology companies that are effected by such patents obviously do not require industrial infrastructure to acquire market share. Their goods and services are essentially inexhaustible. Patent protection was intended (in spirit) to grant a "head start" to inventors so that they might establish market presence on their own inventions, before having to compete with established companies with an existing industrial infrastructure; thus ensuring the incentives of upward mobility. The alternative was thought to be losing domestic markets to foreign sources. If this reminds you of the horse and buggy then you see where I am going with this. Today's technology juggernauts started in garages without patent protection and most are less than a decade old. Patents have in fact, hindered innovation through rent-seeking. For example: MySpace and Facebook both developed without the protection of patents. Both are constantly assaulted with multiple concept patents from companies that innovated nothing. Essentially the industry has enormous incentive to innovate without patents. The situation is analogous to an arms race and oneupmanship is the rule. Software protections are absolutely unnecessary.

3. The people filing amicus briefs on concering software patents are primarily people and organizations that benefit from monopolies on ideas. The reason that Microsoft, IBM and other patent laden Goliaths expect to benefit from patent protection on software is obvious. However Bill Gates famously said: "If people had understood how patents would be granted when most of today's ideas were invented, and had taken out patents, the industry would be at a complete standstill today....The solution to this is patent exchanges with large companies and patenting as much as we can." Others that argue reap the benefit of software patents through "donations," licenses or litigation fees. If you want to know how innovation is really effected, ask software engineers and startups what they think. With assured anonymity they would provide honest unbiased feedback without fear of legal retaliation. They usually side with common sense rather than special interests.

4. The length of time that software patents remain valid render them oppressive. By granting a twenty year monopoly to companies in an ecosystem where consolidation is the rule, you are ensuring that the top 1% of tech companies will hold the vast majority of the IP in the industry. This means that the state reinforces corporate monopolies with state granted monopolies furthering their power to crush opposition. How does that foster innovation? I do not think this point needs to be expounded upon. It is just common sense. The U.S. Constitution never envisioned a scenario where someone could simply imagine the future and stake a claim on it without action or liability of failure. Then as happens now upon either success or failure (through acquisition or bankruptcy or liquidation) the inventor's intellectual assets wind up in the hands of the very companies that they were intended to protect against.

5. One point that makes most proponents of software patents cringe is the international enforcement of software patents. It is impossible, impractical and unethical to deny countries access to concepts that are standards of modern computing. The patents that are filed in the US are generally considered to be ludicrous and overly obvious by the international patent community. There will be no global consensus on software patents anytime soon if ever. Quality standards in nations that have allowed software patents vary greatly and are difficult to enforce even within the EU. At the end of the day it comes down to this: one person or company on the planet Earth owns the exclusive rights to a concept. They then attempt to divert revenue from companies around the world into their coffers. This is obviously not going to benefit countries that are charged with enforcement. Developing countries with little IP have no incentive whatsoever to render verdicts favorable to foreign companies that lay claim to broad concepts. International enforcement of idea patents is pure fantasy. What we are really left with is a massive disadvantage for tech companies in the US. If you want to see this in practice print out ten random technical process patents. Now imagine that you are presiding over an infringement hearing on these patents one at a time. Most would be eliminated in US courts, but presume that you are in a country with no laws to cover patentable concepts (most expressly prohibit idea patents) and see how many you find in favor of.

In closing I would like to point out that if you affirm software/technical processes as patentable a firestorm of litigation will ensue, resulting in a massive and unjust transfer of resources. The resulting effect on innovation in the US would lead to the inevitable question: Why would we expose our company to the risk of crushing litigation in the United States when our markets are just as accessible through the Internet?

Interesting post on Wired, Oct. 7, 1806: Do You Copy? Carbon Paper Patented, about Englishman Ralph Wedgwood, who received "the first patent for carbon paper." But this case only shows that simultaneous invention is common, and that inventions usually come one way or the other. As the report notes, Wedgwood's "work seems to duplicate that of Italian inventor Pellegrino Turri":

Turri in Italy had by 1808 completed an early typewriting machine he had been working on for several years. It, too, was for use by the blind and relied on some form of carbon-impregnated paper. So, his work was more or less simultaneous with Wedgwood's.

Also,

Scottish engineer James Watt, of steam-engine fame, had invented a tissue-copying process for business correspondence in 1779. But it required special inks and fluids and was a wet process for the user, so it didn't catch on.

[SK post]

The key point in analyzing patents is the fact that years of exploring statistical evidence by economists hasn't found any evidence they work to encourage innovation. Kevin Smith of Duke - who is very interested in copyright - has been pouring over economic data and stumbled over the same fact. Facts are stubborn things.

(hattip: Ruth Lewis)

I reprint below (with permission) an email from Prashant Singh Pawar:

Hi Mr Kinsella,

I am a long time opponent of patents and Intellectual Property rights (to a great part because of your work), but something always bothered me and that was the 'innovation' argument by the patent supporters. I could always see both the sides of the arguments, yet was never sure which side is right. I can see that without patents there is no incentive to develop a technology with a large investment, on the other hand, without patents, there is probably no need of a large investment.

I can see that patents make companies develop new things which without patents they wouldn't do, but at the same time patents prevent companies from doing things which they would have without patents. I found this thing common across all the anti-IP and pro-IP work I have read that they both talk about the kinds of innovations they would promote and other side would demote, but there is no clear distinction between the kinds of innovation they are talking about.

So I finally came up with the terms 'Horizontal innovation', and 'Vertical Innovation'. Horizontal Innovation is when a parallel technology is discovered (usually to avoid patent infringement). For example if a company develops a flying car using (say) hydrolic expansion, and they get a patent of it, another company develops (or has to develop) a flying car technology by using Thermo-plazma radiator engine. Both these technologies achieve the same end, they enable a car to fly, so this is horizontal innovation. This is what patent proponents talk about being squashed when they say innovation will be reduced when patents are removed. There will not be Google G1 phone,Blackberry and iPhone if there were no IP rights.

Vertical innovation is when a technology is built top of another technology merely by adding a new element to it. For example if you develop a Car which can travel on water, and I take that car, and add a Sail to it to make it use wind then that's called a vertical innovation. With patents, only the patent holder can think of adding a sail on the boat-car and sell it, without patents, innovations will be done all over the world by every kind of boat and car enthusiast. There will be only one smart phone in this world, but it will be having numerous variants, such as a Google gPhone (synced with google services), a Microsoft mPhone (synced with microsoft services), and so on.

Patents promote horizontal innovation, but restrict vertical innovation. Without patents we will have more vertical innovation but less horizontal innovation. Even if Horizontal and Vertical Innovations are equally good in terms of their merits, one thing is clear, without patents, a lot more people will be able to use the technology, this is some place where a patent-less society will beat a pro-patent society hands down.

Just like if words were copyrighted, and you required a license to use the words, we would have had a LOT of innovation (horizontal) in terms of development of language and you required a license from John Locke's estate to use the term 'liberty', there would have been billions of words in English (a lot of them doing the same thing what others do), but a lot less number of people would be educated, and most of our brain cells would have been wasted on keeping track of 15 different terms for 'liberty', and 'passion'.

I described everything in detail in my article (its completely different from this mail). Please take a look at it, and let me know of your thoughts on it.

Patents: Horizontal vs Vertical Innovation

[Cross-posted at StephanKinsella.com]

A recent email from Brian Helfrich at Boliven LLC got me thinking about resources for patent searches. The two best known sources are the USPTO and Google. Boliven seems to offer some advantages including the ability to search foreign as well as domestic patents, some data analysis, and the ability to download patent data on to your own computer for analysis.

In What Are Words Worth?, the author writes:

I was listening to a podcast talk from Mises University 2009 the other night called "Intellectual Property and Libertarianism", in which speaker Stephan Kinsella made the usual Slashdotty-type case against IP from a libertarian perspective. This was novel for me, perhaps because libertarians tend to be very defensive of property rights, such as Ayn Rand's assertion of IP as a right to the products of a person's own mind.

Kinsella rejects Rand explicitly, saying her case offers little more than deification of the creator. His counter-argument is interesting: IP is inconsistent with property rights because it violates the rights of others to use their property. To wit, if I own a typewriter and a stack of paper, or a CD burner and some blank discs, then those should be mine to do with as I see fit. But because of copyright, I can't use the typewriter to transcribe a book, or to use the burner to copy a CD, even if I've bought original copies of the hypothetical book and CD. IP asserts a partial ownership enough to say "you can't do that" over this other property I own. That, according to Kinsella's argument, is inconsistent and therefore invalid.

Interesting, and tricky, and I don't quite know what to make of it.

It's important because, of course, my income is highly dependent on the idea of IP. If I couldn't charge for copies of iPhone SDK Development, I probably wouldn't have spent hundreds (possibly thousands?) of hours over the last year and a half co-writing it. If I couldn't charge for apps on the App Store, would I write them?

The counter-argument comes from the open-source crowd, who say to give away your content (which, by the earlier argument, you couldn't own anyways), and make your money some other way.

The author here is admirably open-minded. I think he is a bit confused when he says the "counter-argument" is to give away your content, but he also inadvertently hits on an important insight. It's not a "counterargument" of ours to say you "should" just give away content. This is rather a prediction that some business models may employ this approach. But note the author says: "If I couldn't charge for copies of iPhone SDK Development,..." Well, of course you can charge for copies. It's just that it might be hard to prevent others from offering it for free or for a lower price, if you don't find some means of exclusion. In other words, part of the entrepreneur's job is to find efficient means of exclusion. Drive-in movie theater owners employ ticket counters, pay for locks on the doors, and for the installation of hundreds of parking-lot per-car speakers to exclude free-riders. Every business has costs of exclusion. If, in the end, the entrepreneur finds that he cannot make a profit in a given endeavor, taking all costs--including costs of exclusion--into account, then that project would require an inefficient allocation of capital. Simple.

[Cross-posted at SK]