Eric Raymond co-founded the Open Source Initiative in February 1998, and has been an advocate for the movement (and president of the OSI) ever since. Hailing from Malvern, Pa., he’s an Internet developer and writer, as well as a self-professed anthropologist and philosopher who loves science fiction and Tae Kwon Do. Raymond took a moment from e-mailing a sci-fi writer to speak with GT’s Meredith Salisbury about the blue screen of death, Linux, and why open source is crucial to an industry that relies on complex software.
What’s so important about having open source software?
RAYMOND: We need software that doesn’t suck. The closed source system simply does not produce an acceptable level of reliability or robustness, and it’s been getting worse and worse with large, complex pieces of software. Secrecy just doesn’t work anymore. Peer review of source code seems to be the only way of verifying the quality of code that scales up to really large code volumes. Other traditional methods such as structured walkthroughs seem to run out of steam at program sizes below those that are typical today. It’s not that open source is perfect, it’s that nothing else seems to work as well.
How can open source compete with the likes of Microsoft?
RAYMOND: By displacing it on grounds of superior performance, superior stability, and lower cost. Microsoft can compete against a product that’s superior on any one of those axes, but trying to compete on all three ways is pretty hard for them, which is why they’ve essentially been cut off at the pass in the server market. Linux is taking over there — it competes because it’s better.
Even though it doesn’t have the technical support that Microsoft does?
RAYMOND: You don’t need the constant handholding. You never see a blue screen of death. In a pragmatic sense, that’s what it comes down to: Would you like to never see a blue screen of death again?
Why does open source need a whole movement behind it? Who’s involved?
RAYMOND: There are two important organizations: Open Source Initiative and Free Software Foundation. We need people who are capable of explaining what we’re doing to people who are outside our little community. My best estimate is that there are about three-quarters of a million active open-source developers worldwide, and I’ve heard estimates of 12 million to 26 million users, but I think that’s probably low. We’ve seen really explosive growth, including large corporate adoption and especially in the Third World, where they can’t afford to pay the costs. The only statement one can make with certainty is that it’s huge and getting bigger all the time.
In genomics, many academics work extensively with open-source projects, not realizing that doing so may be violating their universities’ policies. What advice would you give them?
RAYMOND: This is an academic freedom issue. The developers of the software should have the right to choose the license that they want in accordance with their research goals and what they want to do with that software. This is probably going to be one of the significant academic freedom issues of the next 50 years.
I really don’t see how this is a fight that the universities can win, and I’ve never heard of a university making a lot of money from software patents. I would encourage academics to make whatever licensing choice their personal values or research agendas lead them to. They should insist that this is their choice, not the university’s choice, any more than the university should be able to exert control over a research paper. No one would stand for a university saying, ‘You wrote the research paper while you were here, and we own the rights to it’ — this is just as basic and unjust and unsustainable. Anytime you get new media coming along you get people who’ll try to assert control over it whether or not it’s legitimate to do so. The natural tendency of bureaucrats everywhere is to extend their empires.