oreilly.comSafari Books Online.Conferences.


There Is No Open Source Community
Pages: 1, 2, 3, 4

Open source supporters have rallied around the cause, looking to prevent frivolous lawsuits against open source developers that might have a chilling effect on open source's momentum. Another concern is the habit of some media outlets to lump together open source with warez and piracy groups. This could conceivably besmirch the credibility of open source philosophy and those associated with open source software. Altogether, while open source and free software have made tremendous strides over the years, strident supporters see the continuing need for vigilance in order to protect what they have built thus far.

What's Wrong with That?

What if you discovered that everything you ever learned about open source growth was wrong? What if the narrative that pitches open source in terms of battling evil software giants wasn't actually correct? What if you learned that the recognized leaders of the open source movement were simply figureheads of a process already well under way? What if you learned that open source was neither good nor bad, but simply the manifestation of decades-old economic trends? What if companies mining the open source vein aren't taking the high road but rather ruthlessly applying a competitive advantage?

Forget what anyone has ever told you about open source, and learn this term: economies of scale. This is what has fueled the open source phenomenon. Most people recognize the role that the internet has played in terms of creating an environment capable of supporting massively parallel distributed collaboration. Too few have recognized the crucial role played by the internet in creating the economies of scale necessary for the proliferation of open source software. Furthermore, there has been little recognition in open source circles of the role the internet has played in driving down software production costs and thus software prices. It is this drastic reduction in price that is necessary for an open source-friendly environment to emerge.

The New Open Source Growth Perspective

It's the internet, stupid.

What is it about the internet that drives down production cost? There are multiple avenues to the result of reduced software cost. Foremost is the advent of the internet as a software delivery mechanism. With the internet, the possibility of practically zero-cost distribution was viable for the first time. Furthermore, digital delivery obviates physical packaging, transportation costs, and reliance on resellers that buy at wholesale, as opposed to direct customers who pay full retail price. Zero-cost distribution gave early adopters a competitive advantage over vendors that relied solely on packaged distribution. It also allowed vendors a much faster way to bring products to market.

No longer did software vendors have to allow for a certain amount of time before packages hit store shelves. When the internet became viable for software distribution, vendors could sell a product almost as soon as a release was certified for public consumption. So far, however, this establishes the internet only as a driver of reduced software costs. That alone does not link the internet to open source proliferation.

This brings up the second ramification of the global internet: instant collaboration across national boundaries. When Linus released version 0.01 of the Linux kernel to the masses from FTP servers in Helsinki, it wasn't long until developers downloaded copies in cities outside of Helsinki and in lands far away from Finland. Users of software were able to take a copy, use it, and then report back to Linus much more quickly than ever before. Gone were the early days of the GNU project, when those that wished to use the software would order tapes and wait for them to arrive in the mail. Reporting bugs and usage issues and distributing patches back to the GNU project also was no small matter. Thus, the internet sped up development time by facilitating almost instant feedback from users regardless of location, as long as they had reliable internet access.

The distributed aspect of the internet also allowed for a much larger audience than before. If Linus had released his kernel just five years earlier, practically no one outside of Helsinki, much less Finland, would have even known of its existence. This is not to say that Linux would not have survived, but its growth would certainly have been slower. It is this massive number of users that makes the numbers work out for large open source projects.

The numbers work out because enough users report back to the Linux kernel developers on bugs or usability issues. A percentage of users contribute patches. Other users take a chance with bleeding-edge kernels not yet ready for release. It is a given that with any software project, most users will be just that—users of the software. However, a certain percentage will become beta testers, bug reporters, documentation writers, or full-blown members of the development team. This number of contributors will grow only if the overall user base grows. Without the internet, this growth of users simply doesn't happen.

Imagine the amount of money that people would need to spend if the internet were not available in order to win over the same number of users and contributors. First of all, using the Linux kernel as an example, Linus would need to print thousands of copies of floppies (remember, this was the early 1990s) and distribute them. The obvious choice would be to distribute them locally on campus in Helsinki. Then, to replicate the experience of the internet, he would need to devise some way to get them into the hands of users around the world—not just any users, but interested users. Using this example, it becomes clear why large-scale open source development did not happen until the internet came along. It also becomes clear why software engineering was largely the domain of companies that could afford the marketing and outreach required to build any type of community around technology.

So far, I have demonstrated only how the internet added cost-effectiveness to software production, but that's still not the whole story. To take the argument one step further, I will demonstrate why economic trends kicked off by the internet have made open source software not only possible, but also actually necessary for survival for some developers and software vendors.

Pages: 1, 2, 3, 4

Next Pagearrow

Sponsored by: