Science as a Gateway to Understanding: International Workshop Proceedings, Tehran, Iran (2008)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

3 The Innovation Ecology WILLIAM WULF President Emeritus, U.S. National Academy of Engineering T he National Academy of Sciences, National Academy of En- gineering, and Institute of Medicine report Rising above the Gathering Storm (Committee on Prospering in the Global Economy of the 21st Century, 2007) and other reports have made recommendations to improve the climate for innovation in the United States, and by inference, internationally. However, many components of an “ecology” of interacting laws, regulations, poli- cies, and institutions are not mentioned in the report—the intellec- tual property system, a broad tax policy that encourages invest- ment, and a culture that encourages risk taking, among others. Two problems with the components of this ecology are immediately ap- parent. First, many of the components were designed for technol- ogy that was developed in the past, but not for technology that is being developed now or will be developed in the future. Thus, they are not optimal for achieving their avowed intent. Second, society almost never completely rethinks the nurturing of technologies, but rather it attempts to adjust the ecology in relatively small ways to reflect current reality. The result is an ecology that is far less than ideal for spurring innovation, and in some cases may even be counter-productive. I will not dwell on the details of the aforementioned report, but the essential message of Rising above the Gathering Storm is 15 

16 SCIENCE AS A GATEWAY TO UNDERSTANDING that America’s children and grandchildren may not continue to en- joy the level of prosperity that has been taken for granted over the last several decades. Innovation has been the driver of our success, and if we want to continue to prosper, we need to nurture innova- tion. The report makes some recommendations about how to do that: • First, we need a strong technical work force, and science and technology education are necessary to this endeavor. • Second, we need to generate good ideas, and we should strongly support basic research. • Third, we need to attract the best minds to science and engineering education. In addition to these crucial components, a larger “ecology” that supports innovation, permits and even encourages risk-taking, and offers “patient capital” to the entrepreneur should be fostered. Laws and regulations must protect the public while simultaneously encouraging experimentation through tax laws that support and reward investment and intellectual property (IP) laws that ade- quately and appropriately protect intellectual property. The list of essential components in the innovation ecology—the collection of interacting and interdependent policies and activities that support innovation—is extensive. Perhaps most importantly, the world is no longer a set of independent economies, which means that each component must interact harmoniously with its foreign counter- part. Let me illustrate the two points above, namely that the ecology we have today was invented for the technology of yester- day, and that the changes to the components of the innovation ecology have generally been incremental. I will give examples from the U.S. experience because I know that system best. I be- lieve all these examples, however, have counterparts in other coun- tries. Let’s start with the current patent system. The system was originally designed to protect large physical machines, which were 

THE INNOVATION ECOLOGY 17 based on the design of their physical models that were, until very recently, filed with the patent application. It seems unlikely that this system designed for large machines can be applied equally well to the protection of software, pharmaceuticals, snippets of DNA, or business processes—all of which are now covered by our patent system. I spoke to a group of about 30 Chief Technology Officers from California in 2005, and each one of them said that the patent system was broken. They only continue to patent items for defen- sive reasons, so that they can trade patents with those firms that prefer to buy patents of developed products rather than to invest in competitive research. They said that from the point of view of promoting innovation, the current patent system was irrelevant and in some cases even counterproductive. Particularly in the informa- tion technology (IT) industry, the speed of the product cycle is so great that the notion of protecting technology for decades is irrele- vant because the patent system is rooted in protecting large ma- chines, not software or biotechnology. It is not tuned to technology as it exists today; and for them, the patent system simply is not working. But it’s even worse than that. Much innovation comes from small companies. However, these companies often cannot afford to build a portfolio of defensive patents, and they are disadvan- taged relative to older, larger, and perhaps less innovative compa- nies. The patent system was created to foster innovation, but it may be having exactly the opposite effect by disadvantaging small companies. Clearly, intellectual property protection is important as a component of the innovation system, so let’s address copyright. The very idea of a prohibition on “copying” as the mechanism of protecting some intellectual property does not make sense for many twenty-first century technologies. Every once in a while I encounter a web page containing the copyright symbol. This page was copied at least a half dozen times on its way to my screen. It was copied from the hard drive of the server to the primary memory of the server. It was copied from 

18 SCIENCE AS A GATEWAY TO UNDERSTANDING the server’s main memory onto the Internet. The page may have been copied anywhere from one time to several hundred times in the process of being transmitted over the network. Finally, it was copied from the Internet to my computer’s primary memory in or- der to be displayed on my screen. Of course, the person who put the copyright symbol on that web page didn’t mean to prohibit that kind of copying, which is essential in displaying their page on my screen. It is perhaps ironic that the web page with the copyright symbol would have no value if it hadn’t been copied. If it were just stuck on the hard drive of the server, it would be of no value whatsoever. The problem is that this kind of copying is absolutely indis- tinguishable from the kind of copying that the author didn’t want to happen. The difference has to do with the intent of making the copy—whether to display the page or to steal its content. At the level of the machine, there is no way to distinguish between the two. The concept of prohibiting copying as a means of protect- ing intellectual property makes sense when artistic or literary ex- pressions are on physical pieces of paper. It doesn’t make sense when it comes to digital information. “Copying” is simply the wrong concept to use in the digital world as the tool to enforce what the copyright laws were intended to accomplish. Let me address another aspect of intellectual property—the activities of the Food and Drug Administration (FDA). A member of the National Academy of Engineering who is a very successful entrepreneur has been involved primarily with medical device companies, although recently he has branched out to deal with some treatment techniques as well. He claims to have an extremely effective cancer vaccine that has been given to a person diagnosed with cancer. This vaccine destroys the cancer by creating an im- mune system response to it, and he says that it is extremely effec- tive against tumors that are otherwise very difficult to treat. What is the problem? Well, it is unlikely that this vaccine will ever be sold in the United States because the FDA test of safety and efficacy requires a randomized, double-blind clinical 

THE INNOVATION ECOLOGY 19 trail. The problem is that this vaccine was manufactured for one specific patient and one specific tumor; it is the ultimate product of personalized medicine. Since there is only one tumor in a specific person against which it is believed to be effective, it is impossible to do a randomized, double-blind test. Thus, under current FDA policy this therapy cannot be approved. Isn’t it ironic? A procedure created to ensure safety and ef- ficacy is, in fact, preventing access to what my friend claims is ef- fective therapy. Once again, a procedure created for an old type of technology doesn’t work for a new type of technology. Let me move on to the antitrust laws which are a compo- nent of the innovation ecology and which help make room for new entrants into a field. These laws were passed in the late nineteenth century, in the era of railroads and steel monopolies. They were developed in the context of the economic theories of the time that equated value with scarcity. Diamonds were more valuable than gold because diamonds were scarcer; gold was more valuable than copper because gold was scarcer, and so on. If we had to compete in the marketplace to buy diamonds because they were rare, we would pay a relatively high price for them. However, with software and other IT market items, the premise of scarcity as a determinant of value is backward. Value is not related to scarcity. Instead, it is related to ubiquity. You may have heard of Moore’s law (the number of transistors per unit area doubles every 18 months), but you probably do not know about Metcalfe’s law. Robert Metcalfe invented the Ethernet, and his law says that the value of a computer network is proportional to the square of the number of nodes connected to it. If only one person in the world owns a telephone, it’s not very valuable. Its value goes up as more and more people have telephones, and it is at maximum value when everyone has one. The same principle holds true when it comes to certain soft- ware. For example, I use Microsoft Word. I do not use it because it’s the best word processor, or because it is the most error-free, or because it is the cheapest. I use it because I am reasonably confi- dent that if I create a “.doc” file and attach it to an e-mail and send 

20 SCIENCE AS A GATEWAY TO UNDERSTANDING it to you, you’ll be able to open it, edit it, and send it back to me. It is the ubiquity of Word that makes it valuable to me, not its scar- city. A law based on the economics of scarcity is not likely to be effective in an economy of ubiquity. Moreover, the remedies in the antitrust system do not reflect the economics of ubiquity, either. Take the company Microsoft as an example. Microsoft has a mo- nopoly, but a few years ago the courts supported Microsoft when the government tried to enforce antitrust laws. I think that the judge made the right decision given the con- text of the case. That is to say, it would not have made any differ- ence whatsoever if Microsoft had been broken up into a company for Windows, a company for Word, and a company for Excel. The motivation to use Word would have remained exactly the same. Now let’s address the problem of tax credits for company expenditures on research and development (R&D) and the U.S. Congress’s annual authorization of those tax credits. The same set of thirty chief technology officers that I mentioned above agreed that the R&D tax credit had no influence whatsoever on their R&D investments. The reason is quite obvious: R&D takes many years. If they invested this year to take advantage of the R&D tax credit and then next year there was no tax credit, they may have to stop their re- search; and they would have just wasted money that they had spent. It is only with the assurance that such a credit will last for a reasonable period of time that it can have a meaningful effect. That doesn’t necessarily mean it should be made permanent. It just means that it must be valid over a reasonable number of years. Each of these aspects that I have addressed is an example of how our current innovation ecology was designed for the past and not the future. I have one final example which concerns the export of technology. The intent of export controls is completely logical; it pre- vents dangerous exports, such as military aircraft to terrorist groups, for example. On the other hand, the implementation of U.S. export controls is broken and counterproductive when we 

THE INNOVATION ECOLOGY 21 control items that are widely available the world over from non- U.S. suppliers. Thus, in many cases the controls only have the ef- fect of damaging U.S. business. The pace of technological change is accelerating. Thus, even if we managed to fix every one of the components of this in- novation ecology to be just right for today and tomorrow, they probably wouldn’t be right for the day after tomorrow. Accord- ingly, our solution must involve a process by which we are able to periodically stand back and evaluate the intent and methods of im- plementation for intellectual property protection, import/export control, and antitrust laws. In other words, what we need is not just a set of changes to patents, copyrights, and so on, but an insti- tutionalized process for renewal that is also relevant to the forma- tion of laws and regulations. DISCUSSION Yousef Sobouti: My impression has been that scientific discoveries and technologies have developed almost in parallel to each other. Do you maintain that in the past few decades, scientific discoveries have taken a leap forward and left the technologies be- hind? William Wulf: No, I wouldn’t say that. There is a wonder- ful book entitled Pasteur’s Quadrant (Stokes, 1997). It character- izes scientific research along two axes. One axis represents the pure search for understanding and knowledge, yes or no. The other axis represents the desire to solve a practical problem, yes or no. The quadrant which expresses neither interest in understanding nor interest in practical application is the quadrant that we don’t need to worry about. The usual description of pure basic science is the quadrant that defines the search for knowledge without respect to application. The usual description of applied research is the quad- rant that defines the desire for application but not necessarily for basic understanding. 

22 SCIENCE AS A GATEWAY TO UNDERSTANDING But there is an important fourth quadrant that seeks basic knowledge and has an application in mind as well. That is “Pas- teur’s Quadrant,” and it is not pursued as vigorously as it should be. I am not particularly worried about the purely applied and not particularly concerned about the purely basic science. Of course, both could use more money. I believe that this is true in Iran, and I am sure that this is true in the United States. But perhaps we ha- ven’t given enough emphasis, prestige, or value to that quadrant of both of our concerns. Hydari Khajehpour: Most of the activities which you have described are more or less related to the private aspect of in- novation, including the property rights of innovators and the legal aspects of those rights. I would mention a point that has been al- most overlooked in the United States, in other western countries, and nowadays in developing countries such as India and Iran; namely, public compensation for invention and discovery. Many innovations are not the product of a single person’s activity or a few persons’ activities in a laboratory; instead, they have been de- veloped based on the cultural knowledge that has no legal protec- tion. Many drugs or methods of treatment are examples. Such drugs or therapies are picked up by pharmaceutical companies and patented after further development. How are the rights of the cul- ture protected in these cases? Cultural property rights all over the world are almost totally ignored. Many cultures’ knowledge has been developed for profit, but their societies have not been compensated in any way. Many local drugs have enjoyed centuries of prior experimentation re- corded in the public record of history, but now are patented as and protected as private property. Related to these public property rights is the investment spent by governments and by public insti- tutions. Legal aspects of public property rights should also be taken into account. Is this also a part of your suggestions? Wulf: That is a very interesting and useful insight. I think it is correct. There was recognition in the United States beginning in the late 1980s that there was a backlog of scientific discoveries, principally within universities, that were not being used to benefit 

THE INNOVATION ECOLOGY 23 the public. This knowledge had been paid for by the public. Well- intentioned legislation was passed by the U.S. Congress to give such property rights to universities so that they would have an in- centive to develop discoveries for the greater good. However, in my estimation, the approach didn’t work and we need another idea. But you are right. There is a huge amount of knowledge that was paid for by the public that only slowly filters back into benefits to the public. Etienne Guyon: I would like to come back to your com- ments on Pasteur’s Quadrant. When Louis Pasteur introduced his vaccine against rabies, he gave shots to a child named Joseph Meister without going through the usual preparatory steps before administering shots. In other words, Pasteur did something com- pletely inappropriate that he should never have done—even in his time it was recognized as an incorrect technique. In France, as in other countries, we promote a principle of precaution which pre- cludes the practice of an activity if there is a slight risk that the ac- tivity will be harmful. How can we reconcile this principle of pre- caution and its impediments with the approach of a man like Pasteur, whose disregard for correct aseptic technique resulted in the first effective treatment for rabies? Is there a solution, a half- way solution? Wulf: The precautionary principle is that we should not do something unless we can be absolutely certain that there will be no negative effects. Fairly recently, our Environmental Protection Agency together with the Food and Drug Administration enacted a ban on silver nanoparticles for use in antibiotics until it is demon- strated through experimentation that they have no negative effects on the environment or on people. Of course, anyone who knows mathematics understands that it is impossible to prove the nega- tive. The general public does not understand this risk or even the notion of risk. I think one thing that we need to do is to educate our public about the notion of risk. The concept that something can be totally without risk is just fallacious. Understanding the need to 

24 SCIENCE AS A GATEWAY TO UNDERSTANDING be very careful, we cannot guarantee a complete lack of negative effects in any circumstances. It is just impossible. REFERENCES National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 2007. Rising Above the Gather- ing Storm: Energizing and Employing America for a Brighter Economic Future. Washington, D.C.: The National Academies Press. Stokes, D. E. 1997. Pasteur’s Quadrant: Basic Science and Tech- nological Innovation. Washington, D.C.: Brookings Institu- tion Press.