Long Live Closed Source Software?

There's an interesting article on the Discover magazine website called "Long Live Closed-Source Software" written by Jaron Lanier. Jaron is a colleague of Richard Stallman though they are philosophically opposed. A quick summary of his thesis is that though the trendy thinking in synthetic biology includes incorporating the worldwide collaboration ideas (aka web 2.0 - whatever that is) and techniques used in open source, that truly innovative work cannot be done open source. Jaron cites examples such as the iPhone and Macromedia (Adobe) Flash as examples of closed-source systems that could never be created by open source processes. Furthermore, open source is a hindrance to innovation by causing brain-drain - my summary of his point that good talent is side-tracked by working on open source. He refers to Linux as a polished knock-off of closed-source unix and the general open source mentality as 70's. He concludes that applying open source ideas and ideals to synthetic biology will not result in an explosion of innovation as touted. As an adjunct he maintains that science, as currently practiced, is open enough and that the incubation period of an idea or hypothesis must occur in a closed environment so it can be protected (like a cell wall) and refined until the point it's ready to release for public review. He makes a lot of good points but also glosses over many and I think takes a stunted view of computing history and market economics in the 80's and 90's. Before Linux and Windows, for example, unix vendors, including Sun and AT&T, failed miserably to supply the market with what it needed: a solid, extensible operating system running on inexpensive hardware. The market was very fragmented, especially in the unix desktop arena and the software and hardware for unix platforms was still considered very expensive. I remember early in my career at Prime Computer, I worked on a port of System V Unix to the Prime 50 series minicomputer and I had printed out the entire source code for System V and was carrying it around in my backpack for study like I had a state secret. After all, Prime paid dearly for a license to System V and if I got mugged, someone would have the source to System V. It seems incredibly silly now. The commercial unix market prior to Linux is a counter example to Lanier's closed source arguments - it was an abysmal failure of closed-source, commercial efforts to give the market what it needed. The commercial unix market had over two decades to fill the vacuum even as Microsoft and Intel plowed the way to inexpensive hardware and software in the 80's and 90's. The irony of course is that it took the adoption of a single hardware platform and to a large degree a single software platform to reach an economy of scale where diversification and mass innovation was possible. Without WinTel, where would Linux be? To his point that integrated circuits and microprocessors are the end all in encapsulation and the pinnacle of closed-source development, I think it's another example of why it may take a mass commercial effort and the rewards that come with commercialization in order to provide a platform for an industry. Any capital intensive development does not lend itself initially to open source. However, once it's reached the mass market stage, the monopoly doesn't lend itself to innovation - stagnation will ensue. In order to build a platform upon which sustainable innovation can occur, historically speaking, it's true that major investment must come first, but that's hardly the end of the story. The mass market in silicon was well established in the late 80's and early 90's. However, I can remember a discouragingly stagnant period of several years when the world had either expensive mainframes/minis or lowly PCs with DOS and no one could think of any more innovative things to do with a Terminate and Stay Resident program or a dot-matrix printer. Then came the internet. The development of the internet is well documented. It was not a closed-source effort and was a government funded initiative at the outset. It expanded by establishing open standards, designing in diversification, providing reference implementations. This was followed by a period of huge investment, commercialization and technology transfer. That in turn enabled mass collaboration world-wide in every topic known to man which continues to this day. If the government funded IP of DARPA was instead incubated commercial/closed-source, how would it ever have sprang to life? It's like any civil infrastructure project - contrary to libertarian views, individuals aren't out to build their own personal toll roads. So, was the explosion in innovation brought on by the internet a result of open standards, government funding, entrepreneurs, close-source, or open-source? It seems to me to be the height of arrogance for any one driver named above to claim responsibility for creating the internet (remember Al Gore?) and by proxy, for enabling all the innovation to come from it. Lanier's argument that closed-source is the only way to true innovation seems patently false given the all-time best platform for innovation, the internet, was not and is not a closed-source development. I think the best argument for closed-source innovation is that it may still be the fastest technology accelerator we have found. However, as the history of Unix shows, closed-source isn't a guarantee for innovation or for giving the market what it needs. Once successful in a narrow band, closed-source methods have a hard time scaling when you consider the vast expanses of the internet or entire fields of science and discovery. For that scale, open source and collaboration between government, commercial and private interests is the only model we've ever seen work. The synthetic biology argument he makes could well be true but only within the narrow confines of jump-starting an industry the way Apple, Microsoft, and Intel did with personal computers. But in the comparison of computing history and innovation, to get to an "orgy" of synthetic biology innovation, it will take open standards, government funding, a massive commercial investment, consolidation, and technology transfer, not to mention an actual market need. That's a tall order for closed-source thinking. Pure commercial innovation and closed-source development, while necessary, is not sufficient for mass innovation.Follow @openlogic
Follow @OSCloudServices

This work is licensed under a Creative Commons Attribution 3.0 Unported License
.