[Previous entry: "YOU-CAN-SPAM"] [Main Index] [Next entry: "More on RIAA"]
12/10/2003 Archived Entry: "Free Software"
Is there such a thing as a free lunch? What if the production cost of lunch is zero?
I would love to see a libertarian economist do an analysis of the Free Software movement. I think this movement is challenging some of our basic assumptions about property, especially intellectual property, the incentives for software writers, and the business model for software.
In the days of Benjamin Tucker's Liberty there was a spirited debate about copyright and patent. One school (which prevails to this day) held that "intellectual property" obeyed the rules of other property, and needed to be protected by the State in order to foster innovation. Another school argued that works of the mind were fundamentally different, since they were not constrained by scarcity:
"A public idea was not a scarce good, as it could be used by an unlimited number of individuals to an unlimited extent and in an unlimited number of places. According to Tucker, this meant that intellectual property as a natural right ran counter to the very purpose for which the idea property evolved." 
In this view, copyright and patent are monopoly privileges awarded by the State, ostensibly "to promote the Progress of Science and useful Arts". 
With the advent of computer software and the Internet, we have a good which challenges all of our fundmental assumptions about "intellectual property."
1. Does scarcity apply? The marginal cost of software is fast approaching zero. Once it is written, software can be distributed on CD-ROM for cents per gigabyte, and distributed via Internet for even less. Unlike most property, software can be duplicated. (No, I do not subscribe to the labor theory of value, but I do recognize that cost is a strong determinant of price.)
2. Is monopoly privilege necessary for innovation? This is a question not of principle, but of pragmatism, and the advent of the Free Software and Open Source movements provides very strong evidence that monopoly protection is not required for innovation...indeed, that innovation seems to flourish more strongly in an environment of free and open distribution of ideas. (It has been argued that the open-source model resembles the open, peer-reviewed model of scientific research, which has delivered centuries of progress.)
Indeed, the incentives have changed. Thanks to the (innovative!) GNU General Public License -- under which much Free Software is distributed -- it can now be more profitable for a business to write free software than to write "proprietary" software. This is because profit is not revenue, but revenue less expenses, and the GPL attacks the expense side of the equation.
Imagine you're a tech entrepreneur wanting to create, say, a new compiler for a microprocessor. There's a lot of prior art available under the GPL, which you can adapt and use, as long as you release the modified software under the GPL. Do you:
a. Write the entire thing yourself, which requires the expense of thousands of man-hours up front, so that you can sell the resulting product for a healthy premium? Or,
b. Use the GPL'd software, saving you development cost and getting you to market faster, but forego monopoly prices on the result? (You can still charge for distribution and support.)
When there was very little GPL software, the choice was easily (a). But now that there is an abundance of high-quality "prior art" available under the GPL, many companies are rethinking the equation, and as a matter of cold-blooded business calculation, opting for (b). And this is an accelerating trend: as more companies adopt the GPL for their software, the incentive to use the GPL increases.
It's been said that this destroys the "business model" of software, and may destroy the software industry. Bunk, on both counts. I've been in computers long enough to know that the business model of commoditized software is a recent aberration. In the early days of mainframe computers, when the cost of the hardware greatly exceeded the cost of software, the software was generally provided gratis by the computer maker. And there was a thriving user community that looked just like today's open-source community: groups like DECUS (the DEC Users' Society) and the CP/M Users' Group provided lots of excellent software that helped jump-start their industries.
With the advent of the mass-produced PC, there was a brief dislocation. For a while, it was possible to sell software for a really large profit...and some fortunes were made. But the cost of software is still much less than the cost of hardware, and competition in the software industry is rapidly returning us to the "traditional" business model of computing -- where the money is made not in stamping out CDs, but in hardware sales, service, and support. Good news for IBM. Bad news for Microsoft.
No, this won't destroy the software industry. It will merely destroy monopoly profits...through the workings of the free market.
1. Wendy McElroy, Intellectual Property: Copyright and Patent in Liberty, http://www.zetetics.com/mac/libdebates/ch6intpr.html.
2. This phrase is from the U.S. Constitution. http://www.house.gov/Constitution/Constitution.html