Wednesday, September 3, 2008

The Cathedral Crumbles

My inspiration in coding came from cathedrals.

Not literally.

I grew up alongside the early video game consoles and their 16-bit glory. The games were elegant and brilliantly built, much like the Notre Dame. Their almost infinite depth could be realized if you spent just a little time to look closer, and there was always something new to discover.

At that point my life's goal was to build the biggest and prettiest cathedral I could and see how many people I could impress. It was an innocent wish.





I stepped into this course, DPS 909, with the hope that I would figure out what this "Open Source" business was by the time I finished. After reading Eric Raymond's "The Cathedral and the Bazaar", I think I've gotten a solid idea of what "Open Source" means before the second class of the semester; much sooner than I had expected. The paper can be found here:

http://www.firstmonday.org/issues/issue3_3/raymond/

Traditional proprietary software is likened to a cathedral, where the masses it is designed to serve are kept ignorant until the grand opening. When it does open, the audience is awed and amazed at all the work and beauty thrust upon them. The architects hope dearly that the users don't notice any of the major flaws, lest they be chased out of the country.

Open source software is likened to a bazaar where the users influence most of its existence, from available products to daily price changes. No regular customers are surprised when they stop to shop; the bazaar just works. Whatever didn't work was cut out countless versions ago by the demands of the masses.

Rather than sit and wait for your users to (hopefully) enjoy the software upon a big release, Raymond suggests that the users be actively involved in evolving the software. It's an efficient way to gather a large pool of genius without high organization overhead. It is also efficient for ensuring that your software's scope will please the target.

The most modern example of a bazaar is alluded to near the end, when Raymond mentions a revolutionary leap by the Netscape company. Netscape takes a big gamble by releasing their source code, an unprecedented action for such a large player in the industry, in hopes of continuing competition against a free Internet Explorer. Sadly, the paper (and the Revolutionary OS video) stops there and leaves us hanging on the results. It didn't take long for me to realize that this was the seed from which Firefox would emerge and dominate the market.





The fact that this Bazaar model has produced a strong competing operating system (Linux) and one of the most popular browsers on the planet (Firefox) is enough to convince me that the Bazaar model really has something to offer. It will be taken into serious consideration as I contemplate the path of my future career.

If I ever achieve that life goal of mine, I will definitely share my source code; but perhaps only a header file or two.

No comments: