Agylen

This morning I had to drive a couple of hours to go to a customer's place and come back, so I thought that I could get a taste of podcasting with my new iPod Shuffle. I downloaded this talk by Clayton Christensen, loaded it onto the iPod, connected the latter to the cheap but effective cassette adaptor I just bought and had an informative drive.

I must say this podcasting thing is great. You learn something instead of being bored by the usual chatter, dumb music and commercials that are typical of FM radio. However, since I fortunately don't drive frequently for long distances, I don't know when I'll be able to podcast again.

By the way, the talk is great. If you have 1h48m to spare, by all means listen to it! I have to take issue, however, with one specific point:

So, I’m not a software engineer or designer, but this is what I think Linux does, or MySQL or Apache, whatever it is, is that because of the open source character of it, the architecture is modular.

I don't think those products are modular just because of their being open source. You can have monolithic, proprietary and highly optimized open source products, just as you can have modular and open commercial products.

Anyway, where Christensen is right on the spot, however, is in his theory that companies doing business with open source can be successful by exploiting the “law of conservation of modularity”. By basing your architecture on open, modular components, you can optimize the tier, applications for example, where most profits can be made.

When I hear that Google, after having based its infrastructure on highly modular, commoditized and not-good-enough Linux PC's, is hiring one Mozilla developer after another, not to mention some ex-Microsoft browser developer, I cannot help thinking that by having one modular, standardized tier below and possibly one above, they are poised to reap the greatest benefits from their proprietary products.