Chapter 1. Open Source Licensing, Contract, and Copyright Law
Open source licensing and development approaches have been challenging and transforming software development for decades. Although open source licensing is often described as radical, it is built on solid, traditional legal foundations, including the rights granted by copyright under the law of the United States (and elsewhere), and the ways in which basic contract principles can alter and supersede those rights.
Basic Principles of Copyright Law
Under the laws of the United States (and of European countries, through the Berne Convention, and of members of the World Trade Organization through the WTO Agreement on Trade-Related Aspects of Intellectual Property Rights), copyright is automatically attached to every novel expression of an idea, whether through text, sounds, or imagery. For example, the words in this paragraph are protected by copyright as soon as they are written. This also applies to diary entries, letters, song lyrics, and drawings, even if they are only done “off the cuff,” in the most casual of circumstances.
For example, a drawing of a dog made on a café napkin is copyrighted simultaneously with its creation and is the sole property—barring any contractual abrogation of the copyright—of its creator. This drawing cannot be copied, displayed, or otherwise commercially exploited by any person other than the creator for the life of the copyright. Among other things, no person other than the creator has the right under copyright law to create "derivative works”—works that depend upon or develop from the original, copyrighted work. This limitation is of particular significance to open source licensing, as will be explained later. In the United States, the period protected by copyright is very long indeed: the life of the creator plus 70 years, or in the case of works made “for hire” or by creators who are not identified, 95 years from the date of publication or 120 years from the date of creation, whichever is shorter.
This does not mean, of course, that the creator of this drawing has a monopoly on the depiction of dogs. Copyright law does not protect any particular idea. Rather, copyright protects only the expression of that idea. The creator of the dog drawing has a right to the commercial exploitation of only that particular expression of “dog.” This right is no limitation on the right of others to create, and to commercially exploit, their own expressions of “dog,” whether through drawing or other media. This limitation to expressions excludes protection from copyright of creations that are not expressed in a tangible, reproducible medium. For example, a dramatic monologue read on a street corner is not protected by copyright. However, if a reading of that monologue is recorded, whether on audio or videotape or paper, it is subject to copyright protection.
This limitation to the expressions of an idea is the principal distinction between the applications of patent and copyright. Unlike copyright, a valid patent does not protect the expression of an idea but the underlying substance of it. For example, a patent applicable to a microchip protects not the expression of the chip itself, or the electrical diagram describing it, but the idea that given circuits can be organized and made to operate in a particular way. Because of their potentially vast scope, patents are construed more strictly, require a registration process, and last for shorter periods than copyrights.
A copyright does not need to be registered to be legally effective. As already noted, a copyright comes into force when the protected work is created. While registration of the work with the United States Copyright Office has some effect on the rights of the copyright holder, it is not required. Moreover, while works published previous to March 1, 1989 need to bear explicit notice of copyright protection or risk losing that protection, works published after that date do not. Nonetheless, use of a copyright notice alerts potential infringers that the work falls under the protection of copyright.
The vesting of copyright protection in the creator of a work is subject to two important limitations: the doctrines of "work for hire” and "fair use.” Works that are made “for hire” are made by an employee in the scope of his or her employment by another, including those that are specially commissioned for use in another work or as a supplement to another work, such as a translation. Works that are created “for hire” are still subject to copyright protection, under the same terms as described above, but the copyright belongs to the employer of the creator, or the person who commissioned the work, not the creator.
The doctrine of “fair use” defines certain uses of copyrighted material as non-infringing. “Fair use” allows persons other than the creator to make certain limited uses of the copyrighted material for purposes of commenting upon or criticizing the work, reporting, or teaching related to the copyrighted material. “Fair use” is a flexible standard, and whether a particular use is considered “fair” depends in substantial part on the extent to which that use impedes the copyright holder’s exclusive rights to commercially exploit the work. In addition, one additional category of work is held to be non-infringing. A “transformative derivative work” is one that, although based on a copyrighted work, so fundamentally alters it that a new work results. Such a “transformative derivative work” is considered a new work for copyright purposes, and the holder of the copyright of the work—from which such a “transformative derivative work” is derived—has no rights over it.
Finally, the protections of copyright are subject to one more important limitation: time. Copyrighted works are protected for a set period of time, measured either from the death of their creator or from the date of their creation. After the expiration of that period of time, the copyright protection on the work lapses as the work goes into the “public domain.”
Works currently in the public domain include thousands of songs and musical works, novels, poems, stories, and histories written before the twentieth century. Anyone is free to commercially exploit such works by selling copies of those works, creating derivative works based upon them, and by distribuing or displaying the work publicly.
Contract and Copyright
In the United States, all of the rights belonging to the creator of a work become theirs at the time of the completion of that work in a fixed medium. No registration is required, nor does any signed writing need to be executed in order to preserve those rights. Rather, these rights arise entirely from the operation of the statutory law.
Creators rarely take advantage of these rights by themselves, however. The production and distribution of works on a large scale has historically been capital-intensive, so creators of works have generally relied on others to produce the physical copies of their works and distribute them. The idea of copyright developed in parallel with the development of the printing press in the fifteenth century, and it originally protected the rights of printers to exclusively exploit works that they had commissioned. Legal enforcement, and, in particular, international legal enforcement, being what it was in the fifteenth and sixteenth centuries, copyrights were frequently disregarded. Of course, given the systematic violation of copyrights in many parts of the world today, it can be argued that the situation has not changed that much.
The relationship between the creator of a work and its publisher is often an uneasy one. The creator, naturally, wishes to retain both control over the use of the work and the income stream derived from commercial exploitation of the work. The publisher, whether a book company, a record label, or a film studio—to take three common examples—similarly wishes to retain exactly those same things: control over and income from the work. Because publishing (in whatever format) is a capital-intensive business, the dynamic tends to strongly favor the publisher over the creator of the work, except in the exceptional case of creators who have both a proven track record of generating income from their work and, perhaps more importantly, the ability to negotiate without restriction. The case of musicians and their battles with record labels is particularly well-known.
The most typical trade made between creators and publishers is the licensing of the work in exchange for payments, known as royalties. In the case of books, authors are generally entitled to royalties on every copy sold by the publisher. Music royalties are more complicated because there are more venues in which music can be sold or publicly performed, but the principle is the same. Royalties are generally owed to the songwriter for every copy of an album sold (mechanical royalties), for play on jukeboxes or on the radio (performance royalties), and for use on television or in films (synchronization royalties).
Software publishing, the subject with which this book is primarily concerned, generally does not involve the payment of royalties to individuals. Because commercial software is made, as a general matter, by large teams of people and requires the substantial expenditure of capital, the resulting work is “work for hire.” As already discussed, the copyright of such works belongs to the employer, which, in the case of software, is usually also the publisher and the distributor of the software itself.
In general, under the American copyright system an effective monopoly is vested in the creator of each work, subject to relatively few limitations. However, for a number of reasons, most of them having to do with the substantial costs of developing and distributing work in a mass-market medium, rights held under copyright are rarely enforced by the work’s creator and very little, if any, of the benefit of the copyright goes to that person. Rather, because of the negotiation of contracts by publishers with the creator or through the doctrine of work for hire, the benefits of copyright flow to the corporations that distribute the work, not the people who create it.
Open Source Software Licensing
In part as a reaction to this distributor-driven model of copyright licensing, programmers developed what is now known popularly as “Open Source” licensing. The development of this manner of software development and licensing has been described well elsewhere and will not be repeated here. For more details on the history, read Free As In Freedom (Sam Williams, O’Reilly 2002), The Cathedral & The Bazaar (Eric S. Raymond, O’Reilly 2001), and Open Sources: Voices from the Open Source Revolution (DiBona et al., O’Reilly, 1999).
The fundamental purpose of open source licensing is to deny anybody the right to exclusively exploit a work. Typically, in order to permit their works to reach a broad audience, and, incidentally, to make some sort of living from making works, creators are required to surrender all, or substantially all, of the rights granted by copyright to those entities that are capable of distributing and thereby exploiting that work.
Because these entities, by their very nature, do not see work as work in the first instance, but rather as the source of an income stream flowing from its exploitation, they are jealous of their right to exclusive exploitation of the work. They are similarly reluctant to share any part of the value of the work with others. While the potential consumers of a literary or musical work will be limited by the costs of acquiring the work—costs that are set exclusively by the person or entity that controls the right to distribute it—market forces will tend to reduce prices so as to maximize returns to that person or entity. Because the marginal costs of mechanical reproduction are relatively low, selling more copies of a work (at lower prices) will generally result in a larger stream of income to the publisher.
As a result, publishers fiercely defend the copyrighted work from unauthorized distribution of copies of the work itself or creation of derivative works based on the work. In the case of artistic works, the problem of unauthorized distribution of the original work is more common. While unauthorized derivative works occasionally result in lawsuits or other disputes, the value of artistic or aesthetic works relies on their original form of expression: they are “non-dynamic.” Consumers want to hear Bruce Springsteen’s Born To Run and to read Dave Eggers’ Heartbreaking Work of Staggering Genius; they most likely do not want to hear Dave Eggers’ Born To Run or read Bruce Springsteen’s Heartbreaking Work of Staggering Genius.
By contrast, software is both functional and dynamic. Each program contains code that is both functional, in the sense that it does work, and dynamic, in the sense that it can perform those functions in an entirely different context. As a result, each program that is created presents two distinct types of value. The first is its formal purpose as a database or another application. The second is a potential source of code for use in performing other functions.
When a consumer purchases a piece of software, say, Microsoft Excel, she acquires, along with the physical copy of the software and the manual (if there are such physical copies), the right to use the software for its intended purpose—in this case, as a spreadsheet program. By opening the plastic wrap on the box, the consumer becomes bound by the so-called "shrinkwrap license” under which she is bound not to copy the work (beyond the single copy made for her own use), not to make derivative works based on the work, and not to authorize anyone else to do either of these two things. The elimination of these three restrictions is the foundation of open source licensing.
A comparable consumer of open source licensed software is in an entirely different position. She can freely distribute (in exchange for payment or not) copies of the work because of the “open distribution” principle. She can freely modify the work and distribute those derivative works (again, whether in exchange for payment or not), because of the “open modification” principle. The only substantial limitation upon her exercise of these rights that an open source license is likely to impose is that the copies of the work that she distributes, whether the original work or her own derivative work, be themselves licensed in a manner consistent with the original license.
For example, an open source license may require that derivative works be distributed on the same terms under which the licensee was permitted access to the work under the original license. This means that those people who receive copies of these works must themselves be able to redistribute the original and to make derivative works from the original, subject only to the limitation that they allow others to do the same. This principle is called “generational limitation.” This limitation may, depending on the terms of the original license, prevent open source code from “going closed” and require that users and contributors to the code abide by the communitarian values of open source.
While open source differs from the operation of traditional copyright licensing by permitting both open distribution and open modification, the removal of the second type of limitation is probably the more important one. By requiring that copyright holders both make available a user-modifiable source code for programs that they distribute and by requiring that they permit the development and distribution of derivative works, open source licensing makes possible three substantial improvements over traditional proprietary commercial software licensing models.
The first, and perhaps the greatest, of these benefits is innovation. It is now well-demonstrated that programmers are willing to contribute to open source projects for no reward other than that of making a program more useful. Open source works. The more programmers that can contribute to a given work, the more value that work is likely to have.
The second benefit is reliability. Many programmers means many people who are available to debug a given program. Moreover, the benefit is not simply one of numbers. A knowledgeable user, who has witnessed firsthand the limitations of a particular application or the effects of a bug on a program’s operation, is generally in a better position to address that limitation or to fix a given bug than an employee of the creator of the original software. Such a user almost certainly has a greater incentive to correct such a shortcoming in a given piece of code than a software publisher, where suggestions to make such corrections must compete not only with other perhaps more pressing corrections, but also with the publisher’s own financial or organizational limitations.
The third benefit is longevity. When commercially licensed software goes “out of print” and is no longer supported by its publisher, there is generally no way that software can be updated or adapted to new uses. Such software comes to an evolutionary dead end. By contrast, open source licensed software can fall into disuse for some period but still be revived, adapted, or rewritten by a subsequent user who finds a use for it—a use that may be completely different from the use originally intended.
Issues with Copyrights and Patents
All of the licenses described in this book can be broken up into two parts. The first part asserts that the person granting the license, the licensor, has the right to license the work to which the license applies. This representation may be implicit or explicit, and may be limited to specific types of rights. A licensor may, for example, assert that he has only applicable rights under copyright to the licensed work and makes no representation about patent rights that may apply to it. The second part of every license is a grant (again, however limited) by the licensor to the licensee of rights to that licensed work.
Obviously, both parts of the license need to be there in order for the license to be effective. When the first part of the license is there and the licensor has all of the rights necessary to grant them to the licensee, the only question is the relationship between the licensor and the licensee under the terms of the license. However, significant complications arise when a third party has legitimate legal claims to the work purporting to be licensed.
In the case of copyrights, a creator of an original work (defined in the legal, not the artistic sense), can confidently license that work, at least to the extent to which it may be governed by copyright law. The creator (hopefully) knows that he or she has not plagiarized the work from another and therefore has the right to license it.
Patents, however, present more complicated issues. It is more difficult to obtain and retain a patent in the first place, and there is always a risk of possible, and possibly unknowing, infringement of a patented process by the licensor, and, accordingly, by his or her licensees.
Unlike copyright protection, which does not even require filing or a formal notice on the copyrighted work, obtaining a patent from the Office of Patent and Trademark requires filing of relatively complex and laborious paperwork, including, most importantly, some explanation of the novelty of the patent in question and how it differs from processes or mechanisms already known. This generally requires the participation of an experienced patent lawyer. But obtaining the patent is not even half the struggle. Because of the profitability of patent royalties, patent holders tend to be very jealous of their rights and patrol the boundaries of their patents vigorously, attempting through the courts to extend the boundaries of their patents as much as possible and at the same time to narrow the scope of patents held by others. This can be, as you may imagine, an extremely expensive and time-consuming ordeal.
Even if a patent holder has licensed that patent for use in open source software, they may not have the inclination or the resources to defend that patent. This may have substantially negative consequences for the licensees of that patent. Although the licensee may have, in good faith, undertaken the use of the licensor’s patent in full compliance with the terms of the license, at some point in the future, that patent may be narrowed or eliminated through litigation by a rival patent holder. Because of the continuing use of that narrowed or eliminated patent, the licensee of the original patent may be liable to a competing patent holder for a claim of infringement. It is possible that such a licensee would want to take action to protect the licensor’s patent, by initiating or participating in patent litigation in situations in which the licensor is unwilling or unable to defend the patent. This can, of course, get expensive.
A larger problem is that there may be patent claims that apply to the licensed software but are known to neither the licensor nor the licensee. Because licensors can only license works that belong to them, the existence of a particular software license is no protection for the licensee against claims of infringement that are not brought by the licensor but by a third-party patent holder. There are no easy solutions to this problem. Software patents are frequently granted and often maddeningly vague.
The Open Source Definition
Now that we have examined the basic principles of copyright and contract and contrasted the operation of those principles with those of open source licensing, it is worth discussing in some detail the definition of open source licensing.
The Open Source Definition is the definition propounded by the Open Source Initiative, used to describe which licenses qualify as “Open Source” licenses. The Open Source Initiative also certifies licenses as OSI Certified to indicate that they fall within the Open Source Definition. We have already seen the basic principles of open source licensing: open source licenses must permit non-exclusive commercial exploitation of the licensed work, must make available the work’s source code, and must permit the creation of derivative works from the work itself. Each of these principles is expressed in the Open Source Definition, and, as we will see later, in the open source licenses discussed later in the book.
The Open Source Definition begins as follows:
Open source doesn’t just mean access to the source code. The distribution terms of open-source software must comply with the following criteria:
1. Free Redistribution
The license shall not restrict any party from selling or giving away the software as a component of an aggregate software distribution containing programs from several different sources. The license shall not require a royalty or other fee for such sale.
This requirement embodies the open distribution principle discussed a moment ago, with the variation that free distribution is required only as part of an “aggregate software distribution.” This relatively minor modification of the open distribution principle was made to include the Perl Artistic License described in Chapter 4, under the umbrella of open source. This modification may well be removed in future versions of the Open Source Definition.
2. Source Code
The program must include source code, and must allow distribution in source code as well as compiled form. Where some form of a product is not distributed with source code, there must be a well-publicized means of obtaining the source code for no more than a reasonable reproduction cost-preferably, downloading via the Internet without charge. The source code must be the preferred form in which a programmer would modify the program. Deliberately obfuscated source code is not allowed. Intermediate forms such as the output of a preprocessor or translator are not allowed.
In order to make the open modification principle effective in software, users must have access to source code. The preferred method of distribution is for source code to come with the compiled code. As a general matter, however, distributors prefer to make source code available separately from the compiled code to limit file sizes and ease distribution.
3. Derived Works
The license must allow modifications and derived works, and must allow them to be distributed under the same terms as the license of the original software.
This paragraph concisely describes the open modification principle that is fundamental to open source licensing. This paragraph also permits, but does not require, the imposition of a generational limitation (such as copyleft) by the license. As will be made clear in the next chapter, such a generational limitation, even if present in a particular license, may not necessarily bar software from “going closed”—being incorporated into proprietary code—depending, of course, on the terms of the particular license.
4. Integrity of The Author’s Source Code
The license may restrict source-code from being distributed in modified form only if the license allows the distribution of “patch files” with the source code for the purpose of modifying the program at build time. The license must explicitly permit distribution of software built from modified source code. The license may require derived works to carry a different name or version number from the original software.
This is a permissive, not a mandatory, part of the definition. Licenses may limit the open modification principle by requiring distributions of modified source code as original source code plus patches, as described, and still fall within the definition. This license provision allows creators to protect the integrity of their work (and presumably of their reputations) by requiring that modifications be provided and identified as separate from the original work. Such a limitation, however, can apply only to the source code. In order to fall within the definition, the license must permit the free distribution of compiled code as modified, although the license may require a distinct name or number for the modified program.
Because of the logistical complications created by the distribution of source code with patch files, licenses that require such distribution are not recommended.
5. No Discrimination Against Persons or Groups
The license must not discriminate against any person or group of persons.
6. No Discrimination Against Fields of Endeavor
The license must not restrict anyone from making use of the program in a specific field of endeavor. For example, it may not restrict the program from being used in a business, or from being used for genetic research.
These anti-discrimination provisions ban restrictions on the use or modification of code by selected persons or for particular uses. The motivations behind such restrictions tend to be moral or political: abortion rights activists might oppose the use of their code by those opposed to abortion; oil companies might object to environmental activists using their work, or vice versa. However well-intentioned such restrictions may be, they are antithetical to the notion of open source and, in practice, are damaging to its objectives. Every limitation on the use of a given piece of code restricts the number of potential contributors, and thereby limits the flexibility, reliability, and longevity of that code.
7. Distribution of License
The rights attached to the program must apply to all to whom the program is redistributed without the need for execution of an additional license by those parties.
This requires that licenses have legally effective provisions that give the identical rights to and enforce the generational limitations, if any, on second and subsequent generations of users.
8. License Must Not Be Specific to a Product
The rights attached to the program must not depend on the program’s being part of a particular software distribution. If the program is extracted from that distribution and used or distributed within the terms of the program’s license, all parties to whom the program is redistributed should have the same rights as those that are granted in conjunction with the original software distribution.
This provision is included to close a loophole under which individual parts of an aggregation of software would be distributed under a different license than the aggregate package, which would be licensed under open source. This loophole allows a fairly obvious end-run around open source principles and is therefore inconsistent with the purposes of open source licensing.
9. The License Must Not Restrict Other Software
The license must not place restrictions on other software that is distributed along with the licensed software. For example, the license must not insist that all other programs distributed on the same medium must be open-source software.
This is not really an open source licensing question at all, but a question of the manner in which software may be distributed. It is included not to directly further the goals of open source but to ensure the freedom of software distributors and to maximize the availability of products licensed under open source licensing.
10. The License must be technology-neutral
No provision of the license may be predicated on any individual technology or style of interface.
This is a housekeeping provision. Some licenses required, as a precaution, that a user take an affirmative action to assent to the license, such as mouse-clicking on a particular box. Because such provisions effectively prohibit the distribution of the program in media (like paper) that are not capable of interpreting acceptance by the user, these licenses effectively limit the free transmission of the code.
To understand the effect of the warranty disclaimer, it helps to have some understanding of what the terms used in it are and what it means to have a warranty associated with or implied by the acquisition of a particular work. The most obvious form of warranty is an express warranty . If upon the sale of a particular item, the seller explicitly states to the buyer that the item being sold, say, an answering machine, will perform a particular function, say, automatically answer incoming calls, the warranty is part of the sale. In the event the product does not perform as stated, the buyer has a remedy against the seller, generally either to have the price of purchase returned or to receive an equivalent but functioning item in exchange for the defective one. Express warranties are very common in sales of consumer goods. My stereo speakers, for example, were warrantied against defects for 10 years from the date of sale.
A warranty of merchantability is not an express warranty, but rather a variety of implied warranty , a warranty created by the operation of law, not by the seller’s decision to make a particular representation. This type of warranty is generally applicable only to merchants, persons who make a business in the sale of particular goods. This warranty operates as a general guarantee that goods sold by a merchant are suitable for use as generally intended. A purchaser who buys rope from a hardware store, even if there is no express warranty, is nonetheless guaranteed that the rope will function as rope generally does. By contrast, if you buy a car from your cousin, who is not a car dealer, you have no guarantee that the car will run in a particular way, or even that it will run at all.
A warranty of fitness for a particular purpose lies somewhere between a warranty of merchantability and an express warranty. Like a warranty of merchantability, it is implied by law, and not by express guarantee; but like an express warranty, it applies to a particular function. Its name describes its function. For example, if you buy rope in a hardware store, and prior to the purchase you say to the person selling the rope, “oh and by the way I am using this rope to pull the car I just bought from my cousin out of a ditch,” and the person selling it says, “oh yeah, it’s strong enough for that,” a warranty of fitness for a particular purpose is implied. If the rope does not work, the buyer, again, has a remedy against the seller.
A warranty against infringement is a type of warranty unique to intellectual property. Such a warranty is a guarantee by the seller, say, a writer or a musician, that the work that she is selling is in fact a work that she has copyright to, generally because she is the creator of the work.
This is probably a good moment to address consequential damages . As described above, the remedies for a breach of one of the warranties just described include the familiar ones of the return of the price of purchase or the exchange of the defective item. However, under at least some circumstances, a seller of a defective product may be liable for more than just the sale price of the item. If the defect in the item causes damages of a type that were reasonably foreseeable at the time of the sale, the seller of the item could be liable for damages that flowed from the defect. These damages are often far greater than the sale price of the item and are known as consequential damages. Suppose, for example, the manufacturer of a brand of coffeemakers makes a particular model of coffeemaker that, contrary to its warranty, will start a fire if left on for more than four hours. If one of those coffeemakers starts a fire that burns down the house of the unfortunate purchaser of that coffeemaker, the manufacturer may be responsible not only for reimbursing the price of the coffeemaker, the so-called direct damages , but also for the value of the house and contents, the reasonably foreseeable consequential damages flowing from the defect.
As described in Chapter 7, warranty disclaimers can also produce business opportunities for developers willing to sign contracts to provide support for products that come without a warranty otherwise. However, these contracts are usually in addition to the open source license, not a part of it.
In light of the potential liability, disclaimers of warranties like that in the MIT License, described in Chapter 2, are commonly found in open source licenses. The use of such disclaimers is not necessarily foolproof, however. A contrary representation or agreement, particularly one made as part of a sale, may end up nullifying the disclaimer and result in liability attaching at least to the person making the relevant representation or entering into the particular agreement. In addition, state or federal law may limit the enforcement or the effectiveness of such disclaimers. Accordingly, licensors should consult with an experienced lawyer before relying on such disclaimers.
 The value of work that simply inspires pleasure in the observer is self-evident. However, the fact that software essentially operates like a tool—it is more like a handsaw than a sunset—makes it fundamentally different than a purely aesthetic creation.
 Such “shrinkwrap licenses” are provided with virtually every copy of commercial software sold today. Although such licenses do not present the formalities that people usually associate with contracts, they are generally enforced as binding contracts. Specht v. Netscape Comm. Corp., 00 Civ. 4871 (AKS), 2001 WL 755396 (S.D.N.Y. July 5, 2001). The enforceability of shrinkwrap licenses is discussed in Chapter 6.
 The term “copyleft” has been used to describe this type of restriction of redistributions of such a work and derivative works. Copyleft is described in more detail in Chapter 3. Because licensors can (and do) impose other types of limitations on second and succeeding generations of derivative works, copyleft is not the equivalent to a generational limitation but is rather one example of such a limitation.
 This may be another meaningful distinction between software and aesthetic works. Aesthetic works may benefit less from contributions from many participants.
 For a more thorough discussion on the effects of patent laws and licensing on open source and free source software, and a compelling argument for limiting the application of patent law to software, see “The Danger Of Software Patents” in Free Software Free Society: Selected Essays of Richard M. Stallman (Free Software Foundation, 2002).