Copyright Versus Free Software
Free Software is often portrayed as anti-copyright. David Chisnall examines the close relationship between the two ideas and argues that Free Software is actually closer to the original spirit of copyright laws than proprietary software license agreements.
A lot of people have been talking recently about the relationship between Free Software and copyright. On one extreme are those attacking Free Software using the straw man argument that all of its advocates are rabid anarchists bent on destroying the system of copyright. More moderate individuals have been discussing the impact of the new version of the GPL, a license based on copyright law.
Before investigating the connection between Free Software and copyright, it’s worth taking a look at exactly what copyright is. The name makes it sound simple enough: the right to copy something. Copyright laws vary a lot around the world, so I won’t try to address any specific laws, but rather give an overview of the principle behind the concept.
The Copyright Bargain
The right to copy something begins with the creator. (Until something has been created, no one can copy it.) In the absence of laws, the creator of a product has two choices:
- Release the product for anyone to duplicate.
- Keep the product secret.
For a long time, a lot of art was created on the patronage principle: Wealthy individuals would commission works of art to show off their wealth. The obvious downside of this practice is that the majority of culture was defined by the tastes of a very few people.
In the absence of patronage or legal protections, neither releasing the creation freely nor keeping it secret looks particularly appealing. If the creator releases his work, anyone can duplicate it without paying the creator. This solution is fine for some things; for example, the Mona Lisa is worth a lot more than any copies of the painting would be. Keeping the work secret is also far from ideal, however; the creator benefits, but no one else does.
Copyright was created as a compromise—a bargain between the creator and society. The creator gets a time-limited monopoly on a given work, enforced by society. In return, society gets full rights to the work at the end of this time limit (when it falls into the public domain), and gets at least some access to it before then. Because this arrangement makes it easier to release works and profit from them, there’s more incentive to create, and more people can benefit from the creations.
This system works fairly well, since both sides benefit from it. In recent years, however, the bargain has been skewed in favor of the copyright owners (who are often no longer the same group as the creators):
- Long extensions to the copyright duration make it unlikely that anyone alive today will ever see anything recently created fall into the public domain during his or her lifetime.
- Digital rights management (DRM) technologies restrict the use of the product.
As its name suggests, copyright covers copying. When I’ve finished writing an article, I send it to InformIT and swap my copyright for some money. This is the only right I have, however. I could choose to sell InformIT some subset of the distribution rights, such as Internet distribution, but I can’t control what they do with the article beyond that point. If we agreed to produce a printed version of this article instead, there’s nothing that I (or they) could do could prevent you, the reader, from using it as toilet paper. Copyright doesn’t give me the right to restrict use—just copying. Under the guise of copy protection, DRM attempts to twist this principle, by saying that you, the purchaser, can use the final product only in certain ways (for example, only on certain media players).
Copyright Versus EULA
Although copyright only restricts copying, someone noticed that you typically have to copy software a couple of times in order to use it. You might copy it from the distribution media onto a hard disk, and then again into RAM to run it. Because you need extra rights to make these copies, the copyright owners reasoned that they could make you give up other rights in exchange for these copying rights, and the End User License Agreement (EULA) was born.
A lot of software these days comes with an End User License Agreement, which tells you exactly what you can and can’t do with the software. If you don’t accept the license agreement, you don’t have the right to copy the software into RAM or onto your disk, so you can’t use it. This restriction violates the spirit of copyright law, and in some jurisdictions the right to make these copies is explicitly granted by law, making EULAs of questionable legality.
In this issue, Free Software is more closely aligned than proprietary software with copyright, because Free Software licenses are not EULAs, and only control distribution. You can use Free Software without accepting any license conditions if you have legally received a copy of the software, and you’re required to accept the license only if you want to make derived works or distribute the software.
This idea can be confusing, since a lot of installation programs for Free Software have a habit of presenting the license as if it were an End User License Agreement and refusing to proceed unless you click the “I agree” button or equivalent. In some cases, the confusion is caused by the installer requiring the EULA; in others, the people who do the packaging fail to read the license for the code that they’re distributing.
Intellectual Property
The expression “intellectual property” (IP) is commonly used to describe a family of legal systems that address ideas, attempting to protect those ideas with something similar to property rights. In the case of copyright, it’s not the idea but rather a given expression of the idea that’s protected. For example, copyright protected VisiCalc, but not the idea of a spreadsheet.
Property rights are an interesting choice as the basis for building protections for intellectual concepts. Ideas have some properties in common with physical objects. For example, ideas can be shared or given away. Unlike physical items, however, an idea can’t be stolen. When Mozart wrote out the score of Gregorio Allegri’s Miserere from memory, the Vatican was not deprived of the piece—simply its monopoly over the piece.
Much advertising money is spent trying to equate copyright infringement with theft. In a sense, this is a valid analogy, but there’s a fundamental difference. If you steal a car, you’ve deprived the owner of the car. If you “steal” an idea, you’ve deprived the originator of something less tangible: the exclusive ability to distribute the idea.
Copyright is one of four sets of laws that fall under the umbrella of “intellectual property.” The other three are patents, trademarks, and trade secrets. Since the 1960s, people have tried intermittently to use patents, rather than copyright, to protect software. Unlike copyright, patents protect an idea. Originally intended to protect machine designs, a patent, like a copyright, is a form of bargain. The creator of a machine provides the patent office with a detailed design, in exchange for exclusivity for a short period (and contingent on the payment of a fee that increases over time). The duration of a patent is typically much shorter than that of a copyright—around 20 years.
Early software patents included a complete source code listing of the algorithm. In recent years, software patents have documented the effects rather than the processes, making such information much less valuable to the public. (The idea behind the patent was that, once it expired, anyone could build the patented machine from reading the patent.)
The form of IP that most closely mirrors the Free Software way of thinking is the trademark. A trademark is a method of protecting a reputation, rather than an idea. This concept is embodied in several Free Software licenses, including the BSD, Mozilla, and Apache licenses, which grant rights to use the code but not the names of the original authors (for marketing purposes, they still require attribution). In many situations, Free Software and trademarks fulfill the same goals, providing a way of attaching a value to a reputation. Red Hat, for example, uses both. They employ Linux developers to provide credibility to their support offerings (after all, who better to support software than the person who wrote it), and use trademark law to prevent other people from using their name to gain this benefit by proxy.
A Clash of Economic Models
Proprietary software fits well with the idea of copyright. A proprietary software company sells copies of software as if they were physical property. This is a simple and obvious economic model to adopt for people familiar with physical property.
A proprietary software company creates a product, much as a car company would, and sells large numbers of more or less identical copies. This business model was popularized by Henry Ford around a hundred years ago, and hasn’t changed much in the intervening period.
Free Software is slightly different. While proprietary software claims “software has value,” Free Software claims “creating software has value.” Much of the software industry adopts the second belief, creating custom software. A customer doesn’t really want software—she wants to solve a problem. These producers don’t employ developers to write software; they employ them to solve problems. The solutions to these problems happen to be in the form of software, but it’s the talent capable of solving the problems that’s worth the money.
In a some cases, there’s very little difference between Free and non-Free software. If the software doesn’t already exist, the customer needs to pay someone to have it written. If it’s sufficiently specialized, distribution rights don’t matter, because no one other than the original customer would actually want it. In this case, it’s just a question of vendor lock-in; can the customer go somewhere else for bug fixes or feature enhancements? At this stage, the original distinction emerges. The Free Software company is selling its skills, rather than the code. The customer can go elsewhere later, but the Free Software company is likely to be more familiar with its own code than an outsider would, and so will be more cost-effective later on. A proprietary software company is selling a product. If you want modifications to that product, you need to go to the originator. In both cases, the customer is likely to go to the original supplier for changes, but for quite different reasons.
This difference leads some people to think that Free Software is somehow anti-copyright. Because copyright is based on the concept that ideas (or, rather, the expression of ideas) should be treated akin to physical property, it is somewhat antithetical to Free Software, which proposes a very different way of measuring value for intangible goods.
Copyleft
Licenses like the GPL are often described as “copyleft,” implying almost the opposite of copyright. Free Software often seeks to re-grant the rights that copyright law takes away, but it’s still based firmly in this legal framework.
Copyright law, irrespective of its original aims, is about providing creators with choices in how they distribute their work. Free Software is one of these possible choices. It’s no more anti-copyright than a proprietary EULA is, and in some ways is closer to the original spirit of the laws.
Copyleft is often detracted as taking away choice, but this isn’t really the case. A developer has the choice of using a copyleft piece of code and abiding by the condition, or writing it again from scratch. Under copyright law without the copyleft license, the only choice for the developer would be to rewrite the code from scratch.
The hacker mentality takes great pride in using tools in a way other than the creator had in mind, and Free Software is the outcome when this principle is applied to copyright law. It is no more anti-copyright than a complicated awk script is anti-UNIX.