capacity vs. capital
When Stuart Feldman created Make in Bell labs in 1977, he wasn’t thinking of owning a product he could sell. He just wanted to get the job done. (sorry, I tried, but I can’t find a good reference for this story. still I’m sure I heard it from a credible source). The job, in his case, was to get a bunch of source files to compile together, in the right order, when needed. Dr. Feldman was fed up of typing a long sequence of compile and link commands to the terminal every time he changed one of his source files. And the most annoying bit was that you usually forget some dependency between files, miss out one command or the other, and the whole project is a mess. What Dr. Feldman needed was a program that could be given a set of rules, and when invoked would use these rules to determine which sequence of commands to execute.
So he wrote one, or rather, hacked it. After all, it had been a long day. Then he thought some of his team mates might like it, so he put it on the shared drive and emailed them.
Later that night he thought there were a few flaws in the rule-file syntax, things he would have done differently if he would have designed it properly. But by the time he got to the office the next day, it was too late. The use of make had spread throughout the lab and beyond. People needed it, and it was free, and easy to install. Make has been part of every *nix system since.
One of the sources of power of the open source model is that it prioritises capacity over capital. Capacity is the ability to do things, capital is means of production – that might or might not give you power to do things.
IBM, Oracle, SUN and others fund open source not out of benevolence, but because there are places where the option of having something for free is worth more than owning it. In fact, sometimes ownership has negative value. How so? Well, if Dr. Feldman would have owned Make, no-one else would ever contribute to its improvement. No major OS provider would include it. As it is, make evolves through small contributions of many engineers, and whenever you find yourself at an unknown machine with some complex project to compile, you know you can count on Make.
Capacity is intangible, so material economies use capital as an approximation of capacity. Obviously, capitalism is based on ownership of capital. The assumption is, if you have the means of production, you can produce, and thus generate value. Of course, its not so simple: you may have boards and nails, I have a hammer, neither of us can build a shed. That’s what markets are for. You can sell me your stuff or I can rent you mine. But that’s an approximation, it doesn’t always work, and when it does, its not always be the most efficient model. As the old saying goes, I want a glass of milk, I don’t want to buy the whole cow.
Open source software production distributes capacity without bothering with ownership of capital. I use a wordprocessor to produce text, I just want it to be there when I need it, I don’t need to own it. That’s easier to do with software, because I can give it to you and still have it. But we’re seeing capacity-centric models popping up in other places. Take cloud computing as an example. I don’t care how many servers Amazon or Google own. I just pay them for the capacity I need, when I need it. Car clubs are another, more down-to-earth example.
So why aren’t we seeing more of this? I think the main obstacle is legal. Our whole system is designed to measure, monitor, and manage the flow of capital. Until we figure out good ways to regulate the trade in capacity, we’re stuck with capital.