The Economic Engine Behind Development of Open Source Software

The interesting stories of where Linux comes from helps illustrate the strong economic model that is driving the development of this OS.

The Open Source community has had to overcome the stereotype of the hobbyist hacker. According to this stereotype, Linux, for example, is built by fourteen-year-old hackers in their bedrooms. We see here an example of the Fear, Uncertainty, and Doubt (FUD) foisted on the software industry by vendors of proprietary systems. After all, who wants to trust their mission-critical enterprise applications to software written by a fourteen-year-old in his spare time?

The reality, of course, is very different from this stereotype. While the ``lone hacker'' is a valuable and important part of the development process, such programmers account for a minority of the code that make up the Linux OS. Linux's creator, Linus Torvalds, began work on Linux while he was a student, and much of the code in the Linux OS is built by professional software developers at major software, engineering, and research organizations.

A few examples include the GNU C and C++ compilers that come from Cygnus Solutions Inc. of Sunnyvale, California. The X Window System originally came from the X Consortium (made up of support from IBM, HP, Digital, and Sun). A number of ethernet drivers are now largely the responsibility of engineers at NASA. Device drivers are now coming frequently from the device manufacturers themselves. In short, building open-source software is often not so different from building conventional software, and the talent behind Open Source is by and large the same talent that is behind conventional software.

Grant Guenther, at the time a member of Empress Software's database development team, wanted to enable his co-workers to work on projects from home. They needed a secure method of moving large files from their office to home and back. They were using Linux on PCs and using Zip drives. The only problem was that at the time (1996), good Zip drive support was not available in Linux.

So Grant had a choice: throw out the Linux solution and purchase a much more expensive proprietary solution, or stop what he was doing and spend a couple of days writing a decent Zip drive driver. He wrote one, and worked with other Zip drive users across the Internet to test and refine the driver.

Consider the cost to Red Hat, or any other software company, of having to pay Empress and Grant to develop that driver. Safe to say the cost would have been in the tens of thousands of dollars, and yet Grant chose to ``give away'' his work. In return, instead of money he received the use of a great solution for his problem of enabling Empress programmers to work from home, at a fraction of the cost of the alternatives. This is the kind of win-win proposition offered by cooperative models like the Open Source development model.