David Robinson, managing editor of The American, has a great article arguing that soaring spending on higher education is something to celebrate:
Modern academics often liken their work to drinking from a fire hose. Historians, philosophers, and physicists all find it impossible to keep up with every potentially relevant paper or study. It’s not just a matter of catching up to the state of the art–one couldn’t even read the research materials in an academic field as fast as they are being produced. Inevitably, this leads scholars to retreat further and further into sub-specialization, narrowing the horizon of what counts as “relevant,” of what their fields consist in. But the side effect of this constant, fractal division of the range of human knowledge is that more and more scholars are needed to cover the same range of topics. A hundred years ago, a biologist could plausibly aspire to know all the important theories and facts contained within the field of biology. But today, there are people working on genetics, proteomics, virology, ecology, and a host of other fields, each of which is a full-time, fully mind-absorbing pursuit in its own right.
This all makes sense once one recognizes that professors are the conduits carrying our accumulated knowledge into the present. Having access to something that is written in a book is not the same thing as knowing it. In order for knowledge to be available and useful here and now, someone must be practically familiar with it. And the more knowledge there is to “cover,” as it were, with practical familiarity, the greater the number of scholars needed to complete a university. This means both more professors now and a greater number of those honors undergrads, training for the professoriate. A greater throughput of accumulated knowledge among successive generations requires an ever-increasing number of conduits.
I think this observation applies equally well to the software world. As software simultaneously gets more complex and cheaper, getting access to a piece of software will be a less and less important part of the overall cost of using it. That was certainly true when I worked as a webmaster in college–keeping up with all the changes in web technology was a full-time job.
And that, of course, is a big part of the reason a company can give their software away and still have a valuable product to sell. Most of the value in free software isn’t in the code, but in the heads of the people who know it intimately. And no one knows code better than the guy who wrote it.
Another thing that academia and the software industry have in common is that non-specialists often have trouble even finding the good stuff. If you want to learn about a new academic topic, it’s often not even clear which are the most important papers to read without some guidance from someone who’s already following the literature. Somebody looking to hire a computer geek faces the same problem. If you’re not technically-minded yourself, there’s no easy way to figure out who’s an expert and who isn’t.
One good proxy, though, is to find the guy that all the others in the community respect. I decent way of identifying the best academics in a field is to find the ones who are cited the most. Likewise, a pretty good way to find highly qualified programmers is to find the programmer whose software all the other programmers use for a given purpose. It’s a pretty good bet that a member of the Apache core team knows something about web servers.
Free software, then, has promotional value for its programmers in precisely the same way that papers have promotional value for the academic. If you write a widely-used piece of software, that raises your status (and potential to earn consulting fees) in precisely the same way that a widely-cited paper raises an economist’s status and ability to charge hefty consulting fees. In a sense, the software (like the paper) isn’t an end in itself, but a means to the end of raising the profile of its creator.