Chapter 1. The Economic and Legal Foundations of Intellectual Property
When programmers get together to talk, the conversation is likely at some point to turn from .NET frameworks or memory usage patterns to copyrights, patents, or trade secrets. People in the computer field realize that a cluster of legal concerns known as intellectual property (IP) plays a big role in its development. Consider just a few of the headline-making legal issues in technology over the past decade, most of which will be remembered by readers of this book:
One of the most explosively popular applications in modern times, Napster, was shut down by a copyright infringement lawsuit in 2000. The founders of Napster thought they were safe from copyright infringement charges because the service itself never copied music files. But because its users shared copyrighted music without authorization from the copyright holders, the Supreme Court took down Napster a theory of “contributory copyright infringement.”
Around the same time, a promising new file-sharing service called Aimster was temporarily shut down on a different IP basis: America Online claimed infringement on its AIM trademark.
The shutdown of Napster (and Aimster) fostered a sudden interest by the public in new or previously obscure peer-to-peer file-sharing protocols. The changing technical and legal landscape has forced the music recording industry to shift its enforcement efforts to individuals, leading occasionally to lawsuits against six-year-olds and grandmothers, and sparking debates over whether colleges should collaborate in making students obey music industry restrictions on network use.
The SCO Group, a tiny computer company formerly prominent in the field of Unix, sued the most famous computer company in the world, IBM, in 2005. SCO put forward a cluster of complaints (soon taken up in lawsuits and countersuits involving other companies) covering just about every area of IP: abuse of its UNIX trademark, copyright infringement, and theft of trade secrets. (The trademark is officially on the uppercase name UNIX, but most of the computer field uses the casual spelling “Unix”.)
Although legal and technical experts scoffed at the claims, many industry analysts worried that the suit would stunt the growth of the open source operating system Linux, which was becoming increasingly important to IT departments in large corporations.
As the SCO cases proceeded, they turned up a range of bizarre claims and debates, including questions of who owns Unix, the enforceability of open source licenses, and what constitutes “copying” of programming source code.
Most of the claims in SCO’s case were rejected in August 2007, and SCO filed for bankruptcy the following month, but the case is still winding its way through the courts.
A series of court rulings in the 1980s and 1990s established that software and business methods could be patented in the United States. The rulings decided that software could be considered a “process” or “machine” (both of which are patentable) instead of an “idea” or “algorithm” (which are not patentable). This resulted in a 3,000% increase in software and business patent filings between 1995 and 2001 alone.
Software patents have been the subject of high-profile lawsuits such as NPT, Inc. v. RIM (which nearly shut down the widely used email service on Blackberry handhelds) and Eolas v. Microsoft (which claimed ownership of a key concept in web browser navigation). In 1999, Unisys decided it held a patent that entitled it to payment from any web site that used a picture in GIF format; GIF was and remains one of the most popular formats for online pictures.
There are so many opinions about IP that just starting the discussion opens the proverbial can of worms. There are IP maximalists who argue that intellectual property is at the foundation of our society, a fundamental building block of our economy. Others tie the right to control our creative expression to our rights and identities as creators. For them, intellectual property is intrinsic to who we are.
There are IP minimalists who argue that intellectual property doesn’t exist; that the very concept is a contradiction in terms because “knowledge cannot be owned.” Others argue against intellectual property because it restricts our range of creative expression. Still others oppose IP on more pragmatic grounds, pointing out that the term “intellectual property” puts many separate laws and concepts into a single indefinite box.
Intellectual property law is, in many ways, a study in contradictions. I think that it is easiest to understand, however, by thinking about intellectual property law as code.
Law and Code
Imagine you are a software developer embarking on a new project with a large existing codebase and an active group of developers. On first impression, the code is messy and contradictory. It is plagued by corner cases and inexplicable design decisions. Your first thought might be to discard all of it and start over fresh. Indeed, some of the long-time contributors agree.
With time, however, you begin to understand some of the design decisions that went into the code. Many of the pure abstractions failed, and the previous contributors patched the code in order to achieve workable results in particular circumstances. In most cases, the original design was roughly followed, but parts of the code were extended or trimmed to accommodate for bugs or adjust to new circumstances. There are some new users of the code, as well—other groups have started using the code to do things that the original developers had never foreseen. Those new users have to be accommodated. The code may be messy, but at least it is understood, and it works where it needs to.
This scenario, which any programmer would dread, is like the current state of intellectual property law. The law is a code, just like computer code. It is even described that way; the books that hold the laws are described as the United States Code (USC). There are definitions, reserved words, and code sections. There are the rough equivalents of subroutines, symbol tables, and linkers. Lawyers and judges act as interpreters. (Lawsuits concerning single passages of the code often take years, making other interpreted languages look like a lap of the Indianapolis 500 in comparison.)
It gets worse: every line of the legal code was written by committee, and almost every line of it has been patched by a later piece of legislation or modified by a court. Indeed, IP law is rooted in a more than 200-year-old codebase. Is it any wonder that it is a mess?
Nevertheless, there is usually logic behind the apparent messiness (or even madness) in the law. Just as with the long-time developers above, the original design of the intellectual property code has been stretched in some places and squeezed in others to make it fit new circumstances and changed priorities. Also, like the developers above, new laws have come to depend on the specific structures defined as intellectual property. We even have courts to carry out a form of test-driven development for new laws. Like the code described above, it may be messy, but at least it is understood, and it usually works where it needs to.
IP is a broad, nuanced, and difficult subject. This book is not about the debates and extremes in intellectual property. It is not meant to argue for or against any particular laws. Rather, this book is an attempt to describe and provide tools for working with the IP system as it currently exists.
The Types of Intellectual Property
There are four main branches of intellectual property, each designed to protect a different type of intellectual product. Later chapters will focus on individual types of intellectual property. For now it is enough to introduce the four primary systems that constitute IP.
Patents are time-limited statutory monopolies designed to protect inventions and technological developments. In return for full disclosure of your idea, you are granted the ability to prevent anyone else from making, using, selling, offering for sale, or importing the invention. Patents last for a maximum of about 20 years, after which the invention becomes part of the public domain.
During its life, the patent protects all implementations of a particular idea. You have the right to prevent other people from practicing (either making or using) your invention, even if they independently invent or re-implement the advancement described in your patent (in other words, even if they didn’t copy your idea).
Because patents offer such strong protection, they are designed to be hard to get. A patent must disclose an invention that is “useful,” “novel,” and “non-obvious.” Unfortunately, this doesn’t mean that all granted patents are useful, novel, and non-obvious! Further, the patent must completely describe the best way to implement the invention using highly technical language. Well-drafted patents usually cost from $10,000 to $50,000 to obtain and generally require the assistance of a registered patent lawyer.
Copyrights are limitations on the expression of an idea. They are designed to protect paintings, sculptures, writings, boat hulls, dramatic works, architectural drawings, and anything else that shows individual creative expression. According to the copyright statute, copyright protection automatically attaches to anything you create as soon as it is “fixed in a tangible medium of expression”—basically, as soon as it is written down or recorded somewhere. Copyrights can last from 90 to about 150 years, depending on the circumstances.
Generally, copyright protection is not as strong as patent protection. Copyright protection does not prohibit other expressions of the same idea. As an extreme case, identical works created completely independently do not infringe the others’ copyright. Further, copyright law has some built-in exceptions that allow other people to use copyrighted materials without the consent of the copyright owner.
Copyright law is applicable to software as a non-dramatic literary work. Although copyright law does not cover purely functional expressions, most code has enough originality to receive at least weak copyright protection.
Trademarks protect the association of a provider of goods or services with a picture, word, slogan, or tune, known generically as a mark. Trademarks were originally developed as an extension to the concepts of unfair competition and consumer protection; trademarks were used to protect consumers by preventing the counterfeiting of goods and to protect the reputations of individual artisans. In the past century, the role of trademarks has expanded to include the concepts of branding and customer loyalty.
You can gain a trademark for free simply by using it, although registration of the trademark with the United States Patent and Trademark Office (the USPTO or PTO) gives additional rights. Registration with the PTO generally requires the assistance of a lawyer and can cost from $2,000–$8,000 (or more) including all fees.
Trademarks are unusual in several respects. First, certain trademarks can last forever, as long as they are actually used. For example, it is highly unlikely that anyone will ever be able to open a steakhouse named McDonald’s. That name is too attached to the well-known fast food chain, and it is quite conceivable that it will be used and associated with that chain for at least the next thousand years.
Second, trademark protections must not overlap. Two companies cannot use the same mark in the same market for the same goods. This is to ensure that there is always a clear association between a particular mark and the associated trademark holder.
Third, trademarks must be defended. Copyrights and patents don’t have to be asserted to still have value. Trademarks, though, will die if they are not defended when they are infringed.
Trade secrets are the oldest form of intellectual property. A trade secret is just information that derives value from being kept a secret. For example, Apple keeps information about its future product plans a secret—this helps increase the excitement around each product release and Apple show. Trade secrets last as long as their secret status is actively protected.
The Intellectual Property System
Even from these brief descriptions, it should be obvious that the term “intellectual property” encompasses a number of divergent and even contradictory bodies of law. Returning to the law and code analogy above, intellectual property isn’t really analogous to just one program. Rather, it is more like four (or more) programs all possibly acting concurrently on the same source materials. The various IP “programs” all work differently and lead to different conclusions. It is more accurate, in fact, to speak of “copyright law” or “patent law” rather than a single overarching “IP law.” It is only slightly tongue in cheek to say that there is an intellectual property “office suite” running on the “operating system” of U.S. law.
With so many different moving parts in the system, simply naming and describing the different types of intellectual property doesn’t do enough to explain why the intellectual property system works the way it does. To understand the reasons behind the messiness of intellectual property, it is necessary to stand back and look at the system as a whole, as well as the problems intellectual property was designed to fix.
Intellectual Property and Market Failure
Intellectual property starts with economics. Intellectual property law is, at its most basic, an attempt to remedy a failure in the market for knowledge. We want more knowledge in society, but the nature of knowledge tends to discourage (or technically, underencourage) efforts to create and share new ideas.
Normally, economists analyze society in terms of preferences, markets, and incentives. We all have preferences—things that we want and things that we don’t want. A market is the place where we exchange goods and services with others, making decisions about how to best satisfy our preferences. There are costs (incentives) associated with getting what we want; the “price” of something is the result of balancing how much we want some good (our demand) with how much other people are willing to provide that good (the supply).
The interesting thing about markets is that they involve tradeoffs. Because we have limited resources, we have to make choices between different goods. If something costs very little, we tend to substitute the low-cost goods for high-cost goods.
Normally, the balancing of costs and preferences results in an optimal aggregate distribution of goods. Every once in a while, however, we encounter a market failure, a situation where balancing costs and preferences results in overproduction or underproduction of a certain good.
In this particular case, the good that we want is knowledge. As we will see, creating new knowledge is costly, and normal markets tend to discourage the creation of new knowledge. Intellectual property is the tool that we use to remedy this market failure. That is, intellectual property is the tool we use to change incentives to increase the amount of knowledge in society.
More specifically, intellectual property law is designed to fix the problems that arise because: 1) knowledge costs more to create than it costs to copy (or consume); and 2) secret knowledge is more valuable to individuals, but shared knowledge is more valuable to society.
The Cost of Creating Knowledge
Thinking is work. It is sometimes hard to compare thinking to other kinds of work—at the end of the day, there are no holes dug, or products made, or rooms cleaned, but anyone who has worked over a particularly hard problem all day knows that it takes time and effort to create solutions to problems. Although we embody our solutions in code or in writing, the real effort is the cost of creation. The code we write is simply an artifact that allows us to share the products of our thinking.
Once a person has paid the cost of creation, however, the economic cost of a second person using that knowledge moves down to essentially zero. The SSL libraries used to encrypt HTTP traffic are a good example. The cost of creating SSL was (and is) enormous; it includes the cost of developing the theories and algorithms governing SSL, as well as the cost of translating those algorithms into fast, efficient, and correct code.
For those who want to understand the technical details of SSL, there is still effort involved in learning and understanding the code. Nevertheless, the cost of acquiring that knowledge via OpenSSL and its documentation is vastly smaller than the cost of originally creating that information. Even Isaac Newton, generally regarded as a genius for his creative effort, acknowledged that his work built upon the work of others. When Newton stated in a letter to Robert Hooke in 1676, “If I have seen a little further it is by standing on the shoulders of Giants,” he was acknowledging the mental work expended by others to raise his base level of understanding. The difference in cost between acquiring knowledge from another person and originally creating that knowledge is substantial.
For those who just want to use the fruits of other people’s knowledge, the cost is essentially zero. For example, millions of people use SSL many times each day, and never think about the hundreds of thousands of hours of effort expended to make SSL work.
This is the first basic dilemma of information: high-quality information tends to have a very high cost to create, a much lower cost to acquire, and almost no cost to use. Therefore, our incentive is to use other people’s knowledge frequently and to create new knowledge rarely.
The Value of Secrets
Secrets are valuable, but they have value only for those holding them. For example, the PIN associated with my debit card has value to me only as long as it remains a secret. Both public-key and symmetric encryption rely on secrets for their value. On a grander level, wars have been fought and lives have been lost over secrets; we have multiple government agencies dedicated to keeping our secrets and uncovering the secrets of other people.
Although some secrets are of little value to society, other secrets could have great value if they were revealed. For instance, one of greatest ceramic makers in Renaissance Florence (the Delia Robbia family) found a secret way of making particularly bright and resilient colors. Many others could have used this knowledge to create beautiful ceramics as well. When the founders of the studio died, however, the secret was lost for centuries (porcelain-making, itself, was a secret known only to the Chinese for a long time). The Delia Robbia family profited because it kept its chemical formulae secret, but society also suffered because other competent ceramics makers could not use the technique.
Shared knowledge is especially valuable because knowledge is generally susceptible to network effects: the more people who possess a particular piece of knowledge, the more valuable that knowledge becomes because it gets pooled with other knowledge to lead to new applications. For example, the success of the scientific method is based upon sharing hypotheses and experimental results. Industrial advances are generally the result of incremental progress by multiple groups working on the same problem. Economic markets are driven by and depend on the sharing of information.
This is the second basic dilemma of information: information has higher individual value when it is kept secret, but higher societal value when it is shared. Those who create new knowledge often have an incentive to keep their work a secret so that they can keep the benefits of that knowledge to themselves; if society wants to reap the greater benefits of sharing, it must counter that incentive with others.
The Nature of Information
We want more knowledge (or more generally, more information) in society. As discussed above, however, normal market mechanisms do not provide incentives for individuals to create and share new knowledge. Economists classify information as a public good that is susceptible to the free rider problem. Before getting into the law, therefore, we will take a brief detour into economic theory.
It’s All Good(s), or Information in Economic Theory
Economic theory divides goods according to two axes: rivalrousness and excludability. Different rules apply for different categories of goods.
A rivalrous good is something that only one person can have at a time. For example, food, cars, and physical goods generally are examples of rivalrous goods. If I eat a banana, nobody else can eat it. Another rivalrous good familiar to anybody with teenagers is the mirror in the bathroom; if one person is getting ready for the day, nobody else can use it. (OK, technically this is not rivalrous, since more than one person can look in the mirror at one time, but the common use makes this rivalrous—see the further explanation under the upcoming “Excludable goods” section.)
Information, on the other hand, is a non-rivalrous good. If I have a piece of information that I share with you, then we both have the benefit of that information. Thomas Jefferson, most famously the third president of the United States, but also the first patent examiner in the United States, described it as follows in The Writings of Thomas Jefferson (Derby & Jackson, 1859):
[The] peculiar character [of an idea] is that no one possesses the less, because every other possesses the whole of it. He who receives an idea from me, receives instruction himself without lessening mine; as he who lights his taper at mine, receives light without darkening me.
A non-rivalrous good, therefore, is something that people can share at the same time without any single person having to even temporarily give up part of it. In the words of Thomas Jefferson, each person “possesses the whole of it.”
You can further break down the definition of rivalrous goods. Some things are rivalrous in ownership only, and others are rivalrous in use. As noted above, a banana (in its convenient prepackaged form) is rivalrous in its use—only one person may eat the banana. The banana is also rivalrous in its ownership. Many people may “own” a banana in the legal sense (they may have some legal control over or claim on the banana), but there still must be a known, countable number of banana owners.
An excludable good is something that another person can be prevented from obtaining. I can stop you from obtaining my banana by eating it myself. No matter what you do, it is impossible for you to get the banana after I have consumed it (at least, it is impossible for you to get it in the same convenient prepackaged form). Putting a lock on the bathroom door makes the bathroom mirror an excludable good. Even though the bathroom mirror is not used up like the banana, the lock keeps anyone else from using the mirror at the same time.
As might be expected, however, information is not like a banana. Quoting again from The Writings of Thomas Jefferson:
If nature has made any one thing less susceptible than all others of exclusive property, it is the action of the thinking power called an idea, which an individual may exclusively possess as long as he keeps it to himself; but the moment it is divulged, it forces itself into the possession of every one, and the receiver cannot dispossess himself of it.
A non-excludable good, therefore, is a good that I cannot prevent you from consuming. Shared knowledge possesses this quality of non-excludability; there is no general method of teaching something to someone with the intended result that he or she does not learn it.
The four types of goods
Goods can therefore be categorized into four different types based upon whether they are rivalrous and exclusive. The four types of goods are private goods, common-pool goods, club goods, and public goods.
Private goods (cars, houses, computers, MP3 players, bananas)
Common-pool goods (fishing stocks, the rainforest, the air)
Club goods (cable television, golf courses, group discounts, Wi-Fi access)
Public goods (the environment, national defense, lighthouses, information)
Private goods are both rivalrous and excludable. Most things that you would think of as “property” fall into the category of private goods. It doesn’t matter if it is portable, like an MP3 player, or fixed, like your house. If only one person (or a finite, countable number of people) can possess the good at one time, and the owner of the good has the power to keep others away, then it is a private good.
One way to think about private goods is to analogize them to locks or mutexes in a multithreaded program. A number of different threads may want to use a protected resource, but control of the lock around the resource is rivalrous; only one thread can hold it at a time. You can pass around control of the lock to different threads so that each holds the lock in turn, but no two threads can hold the lock at precisely the same moment. In fact, the point of using locks in a multithreaded program is to make certain resources excludable; locks are put around critical sections precisely to keep other threads out.
For example, I have at times been a member of a CD club, one where you can get 7 or 10 CDs for the price of one. The CD promotion could theoretically be applied to an infinite number of people, since there is nothing in the nature of the CD promotion that requires that only one person or group possess it at a time (although the record-keeping and billing systems of the CD club would have a hard time dealing with an infinite number of subscribers).
To better illustrate the unlimited nature of club goods, consider the “locals” discount given by Disneyland to the residents of Southern California. Any person that presents proof of local residence can receive a specially discounted annual pass to the park. When considered over time, the movement of people into and out of Southern California makes the total number of people eligible for this discount both uncountable and infinite.
The park discount example also illustrates another quality of club goods—they are excludable. To receive the discount, you must provide proof that you live in Southern California. No proof, no discount, even though the total number of people that may eventually provide proof is infinite.
When considering information, club goods are interesting because they show how excludability can solve the problem of non-rivalrousness in some circumstances. By putting a gate around Disneyland, the Disney corporation is able to ensure that it is compensated for all visits to the park.
Common-pool goods are rivalrous and non-excludable. The name “common-pool” is used because the classic examples usually revolve around commonly held resources. For example, schools of fish migrate all over the ocean, so it is very difficult to prevent people from catching them in different places. However, the use of the fish is rivalrous—a fish can only be caught and eaten once.
The most famous example of a common-pool good is a town commons used for grazing animals. In the agrarian societies of the 17th and 18th centuries, town dwellers usually retained land for small personal farms. The individual farmers, however, did not generally own enough land to provide grazing area for all their animals.
The solution adopted by some towns was to provide a large town commons, or grazing area, available to all residents. Each day the individual farmers would bring their animals to the commons for a period of time to graze. Each night they would bring the animals home to sleep in their personal barns.
The result of this arrangement has become known as the tragedy of the commons. Individual landowners used the common area to support more animals than they would otherwise keep. Grazing area maintained on the personal farm was relatively expensive—it was space that couldn’t be used to grow more crops or house more animals. In contrast, the grazing area was relatively free—there was no additional cost associated with using the commons. Faced with an expensive choice (use personal land) and a cheap choice (use the commons), people used the commons as much as possible. As a result, the commons was degraded from overuse.
That is the essence of the tragedy of the commons. Because no individual person or group “owns” the common-pool goods, people perceive that anything taken from the common pool is free, or at least lower in cost. This creates an incentive to overuse the common-pool goods at the expense of other goods. As common-pool goods are rivalrous, some potential users of the good are crowded out and receive much less than their share.
Public goods are non-rivalrous and non-excludable. These are goods that we want more of in society, like a clean environment or defense from national enemies, but they are goods that by their nature cannot be owned, controlled, or provided by any single person. Private goods can be managed naturally because each owner has an incentive to preserve the resource or consume it in a useful way. Public goods have no such controlling entity; they are diffuse.
Although public goods are diffuse, they can still be costly. Taking a clean environment as an example, it may be easy to remove 90% of the pollutants from a particular area. It is harder to remove 99% of the pollutants. It may be extremely difficult to remove 99.99% of the pollutants, and may be essentially impossible to remove 99.9999%. Dealing with these problems is like dealing with computer problems that have exponential complexity. The simple cases may be doable, even trivial, but the harder cases become very hard very quickly. The only difference is that computer problems use up time and space, and public goods problems use up time and money.
Like common-pool goods, public goods are not excludable, so they suffer from a form of the tragedy of the commons. Public goods are free to use, so people feel the incentive to use as much of them as possible. Additionally, since public goods are not rivalrous, they don’t suffer from crowding out, but they can be degradable. The overuse of degradable goods tends to destroy some of their value.
It gets worse—public goods are also susceptible to a particular type of market failure called the collective action or free rider problem. The free rider problem exists because public goods are not only non-excludable, but are also costly to produce.
The free rider problem gets its name from the most common example used to illustrate this problem. Imagine that each day there are many city buses available to take people where they need to go. There is a catch, however: the first person to get on the bus pays $10,000, covering the fares for everybody else. The second, third, and all others getting on the bus get to ride for free.
In this situation, there is a high cost for being the first person to get on the bus, but a low cost for everybody else. The incentive is for everybody to hang back; nobody wants to be the person who pays the fare. Once somebody has paid the fare, however, everybody crowds on. After all, the ride is free—it has been paid for by someone else.
One way of looking at this situation is that the free rider problem is the incentive not to pay the cost for public goods (nobody wants to be the first person on the bus). Once somebody has paid the cost for the public good, however, the situation changes into a variant of the tragedy of the commons (everybody tries to ride the “free” bus, resulting in a crowded and smelly ride).
Information As a Public Good
Information is a public good that is susceptible to both the free rider problem and the tragedy of the commons. Specifically, these problems are seen in the cost of creating knowledge and the value of secrets.
The cost of creating knowledge is high, but the cost of consuming it is low. This is the free rider problem. There is a cost associated with creating new and useful knowledge that people don’t want to pay, especially if there is a way to gain the knowledge from someone else. Therefore, there is a societal incentive to not create as much knowledge as we would ideally like to have.
Secrets are more valuable to you personally, but shared knowledge is more valuable to society. This is the tragedy of the commons. As long as you are the only one who knows a secret, it is as if you are the only person who has a key to the gate around part of the town commons. The resource is valuable to you because you have the key, but it is worthless to everybody else. The only way for others to get any value from that part of the pasture is to get the key from you.
Changing the Nature of Information
Algorithms for creating knowledge
Faced with the public-good problems of information, there are a number of alternative ways in which we can, as a society, increase the production of knowledge. The first and most obvious answer is to handle information like we handle other public goods; that is, by working collectively to create knowledge by directly paying people to create new knowledge. Just as we use collective action (government) to provide environmental protection and national defense, we can also use the government to fund scientists, engineers, and inventors.
We do this already. Government grants are a major source of funding for scientists, and we have government agencies like the National Institute of Science and Technology that directly employ scientists. Tax-supported public universities are another means by which we publicly fund the creation of knowledge.
The bad news is that direct funding of knowledge creation doesn’t always work. The reason is that government-directed funding is the real-life equivalent of a procedural algorithm in code. Procedural decisionmaking, like other procedural algorithms, works best when there is a relatively constrained problem space and the procedure for moving forward is widely known. Thus, government funding works best for broad scientific knowledge and general engineering problems. The search strategy (the scientific method) is known; all that remains is the application of the strategy to specific problem domains.
Other types of information are not so straightforward; they suffer from a massive problem space with many local maxima. For example, what is the formula for creating a great novel, a winning slogan, or a piece of sculpture? While many great works of art and culture have been created using public funding, it is not possible in general to direct the creation of the next great American novel. Our best efforts aren’t much better than random chance. There is a similar problem with groundbreaking inventions. By definition, they are new—they are not piecewise refinements of existing solutions. They frequently arise out of new and unusual ways of addressing problems.
Markets, on the other hand, are like parallel or evolutionary algorithms. They are designed to find solutions in massive problem spaces by attempting many different solutions and testing each of them for success or failure in the market. Therefore, they could provide an alternative method of funding the creation of knowledge—if not for the free rider problem and the tragedy of the commons.
Making public goods private
If you recall, I used bathroom mirrors and shared resources in programs as examples of private goods. As I hinted above, this is not exactly true. More than one person can look in the bathroom mirror at one time, and more than one thread can use the shared resource at one time. In both cases, though, the result of multiple simultaneous use is a mess—somebody or something blows up.
The solution, for both the bathroom door and the shared resource, is a lock to make these sharable goods private. This illustrates an important point: legal or technological controls (that is, code) can change the nature of goods.
This is also a solution for information. Legal controls are added to the information to convert it from a public to a private good. By making some aspects of the information private, normal market mechanisms can be used to make investment choices and to allocate resources; the parallel algorithms of the market can be used to find and fund the creation of more knowledge for society.
If we could share our secret knowledge and still prevent it from being used or understood by others, there would likely be no secrets; no particular advantage would be gained by keeping the knowledge away from other people. The function of intellectual property law is just that—it allows us to share our secrets while still controlling how they are used.
The price that inventors and authors pay for receiving exclusive control over their knowledge is that control is only granted for a limited period. Article I, Section 8 of the United States Constitution makes this bargain explicit: “To promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries.”
The result of this bargain is that knowledge is temporarily mixed with law to create a hybrid good called “intellectual property.” After the limited times decreed by the Constitution and by Congress, the knowledge reverts to its natural state as a public good; the freed knowledge is then said to be in the “public domain.” The intended consequence of this process is an ever-increasing store of knowledge that is freely accessible to the public.
The purpose of intellectual property in economic terms
Modern intellectual property law has two objectives, first, to allow individuals to be compensated for the costs and risks inherent in knowledge creation, and second, to balance and align the interests of individuals and society by providing individual incentives to develop and share knowledge and societal protection for certain types of secrets.
Intellectual property law accomplishes these objectives by creating IP out of information; it takes intangible products of the mind and imbues their creators with special legal rights over their creations. These legal rights make certain aspects of the underlying information excludable, allowing private markets to allocate the appropriate resources to knowledge creation.
The rights granted under IP law are sometimes called negative rights, which are actually the rights to prevent other people from taking a particular action. In other words, the most consistent effect of the IP laws is to make some piece of knowledge excludable.
There are two basic misconceptions to address before continuing. First, thinking of “property” as an inherent part of an object rather than artifact of the legal system, and second, thinking of property as a singular right rather than a collection of various rights.
Property as a legal concept
A “good” is an item. It has an independent existence regardless of the law. Land, trees, books, bananas, clean air, and information all are goods. They may not all be tangible, but they exist in some sense outside of any system of laws. They would still exist on Mars, where there are no governments or laws (yet).
“Property,” on the other hand, is a legal concept only. Property is the name that we give to something that we have legal control over. The land on Mars is not anybody’s property. It won’t be anybody’s property until there is a system of laws that allows some entity to claim it, control it, and use force to evict people from it.
Among the IP minimalists mentioned at the start of this chapter, some go so far as to denounce the very use of the term “intellectual property.” The Free Software Foundation (FSF) places it among “Phrases that are Worth Avoiding” (http://www.gnu.org/philosophy/words-to-avoid.html#IntellectualProperty). Others scoff at the idea that using an idea, song, or algorithm could ever be “stealing.”
There are some IP maximalists, on the other hand, who find it ridiculous that an idea could ever be “free.” They would argue that using someone else’s idea is always stealing, that you just can’t always enforce your rights.
Both of these extremes represent opposite sides of the same fundamental misunderstanding. Intellectual property is a hybrid good made up of equal parts information and law. The IP minimalists look at the hybrid and see only the underlying information; the IP maximalists look at the hybrid and see only the legal controls. In this hybrid sense, at least, the term “intellectual property” is appropriate: it acknowledges both the informational and legal aspects of the combined good.
As a corollary, it is interesting to note similar confusion around the term “stealing.” Stealing is the violation of a legal property right, so unauthorized use of intellectual property is stealing. However, once that property right expires, it is impossible to steal the remaining information. No matter how attached the creator may be to a particular piece of knowledge, everyone in the world “possesses the whole of it.”
Property as a bundle of rights
The second major misconception about property rights is that they are singular. The truth is that any kind of property is a collection of separate and independent rights. Each one of those rights may be individually sold, licensed, given away, or destroyed.
One common analogy is to compare property rights to a bundle of sticks. Each stick represents a different legal claim that you have on the underlying good. For example, land has a number of rights associated with it: there are separate rights associated with being on the land, drilling or mining on the land, building on the land, living on the land, and walking or driving across a particular portion of the land. In some cases there may be other rights, such as the right to build a tall building on the land (separate from short building rights), the right to use the water on the land, and the right to fish or hunt on the land. Each one of these rights (and almost any other rights you might be able to come up with that concern the use of the land) is an individual property right. Many times people own the whole bundle of rights, but just as many times they don’t. For example, most people in Texas don’t own the mineral rights to the land underneath their houses. That is, they don’t own the right to anything more than 500 feet down from the bottom of their houses.
Reproduce the copyrighted work in copies or phonorecords.
Prepare derivative works based upon the copyrighted work.
Distribute copies or phonorecords of the copyrighted work to the public by sale or other transfer of ownership, or by rental, lease, or lending.
Perform the copyrighted work publicly.
Display the copyrighted work publicly.
In the case of sound recordings, perform the copyrighted work publicly by means of a digital audio transmission.
In addition, each one of these rights can be broken down further into even smaller “bundles.” Take the right “to reproduce the copyrighted work in copies or phonorecords.” That right can be broken down temporally (you may reproduce the work for one year), geographically (you may reproduce the work in your own state), by format (you may reproduce the work on compact disc only), or by almost any other restriction you might care to dream up.
There is no obligation to be consistent about how you slice and dice your bundles of rights. You may sell the same right to different people, you may license different rights for different amounts, or you may keep all of the rights to yourself.
Property rights and enforcement
Despite the negative associations some people have with the idea of intellectual output being termed “property,” the status of intellectual property as property has an important function in our legal system. Specifically, property rights can be enforced with an injunction. That means that the courts will enforce the property owner’s rights to exclude others from using the intellectual property; they will order any infringing users to stop. This is in contrast to equity rights, which will be enforced by the courts only by requiring an infringing user to pay damages (usually money).
Evaluating the System
So, does intellectual property work? Yes...and no. In one sense, our intellectual property system has been phenomenally successful in encouraging people to create intellectual property. For the past 50 years—and especially the past 30—there has been a tide of stronger intellectual property protections across industries. This growth in IP has encouraged people to invest heavily in the development of new intellectual property, and has moved IP to the core of many business strategies. For most businesses in the United States, in fact, the intellectual property part of the business is the most valuable aspect of the business.
Nevertheless, people’s attitudes about intellectual property are changing. We are starting to see a swing away from stronger intellectual property protections, and toward more openness and collaboration. As things change, it is important to understand not only the current intellectual property laws, but also the structure and purpose of the underlying system. Part of this swing toward openness is reflected in the growing acceptance and importance of open source software.
Whether or not people agree about the desirability of intellectual property, it still has to be acknowledged as an independent discipline and a major force in the computing industry. For example, there are intellectual property divisions in law schools, intellectual property departments in corporations, and intellectual property lawyers in the telephone book.
Furthermore, different concepts under the intellectual property umbrella work together and it takes a lawyer to help you understand how they are coordinated and apply to your specific situation. For instance, should a particular inventor rely on a trade secret or a patent for protection? Is copyright enough to protect a cartoon character, or should it be registered as a trademark as well? These concepts become entwined through use.
The next chapters take a deeper dive into the specifics of each branch of intellectual property. Except where necessary, I will not return again to the broader foundations of intellectual property law. As you read, however, it would be valuable to consider the philosophical foundations as they relate to each branch of the law. In some cases, the original intent has been frustrated by later developments in the law. In other cases, the utilitarian bargain is more or less working as expected.
Either way, intellectual property is in a state of flux. The development of IP law and the ability to enforce IP rights usually lags behind the technologies that both enhance and threaten those who define their lives and their businesses in terms of intellectual property. Keeping an eye on the fundamentals is one way to predict the ways in which IP law and technology in general will move in the future.