Monday, April 27, 2009

More on Industrial Policy   [Jonah Goldberg]

From a reader:

Jonah,

It's even worse than that.

I'll preface this by saying that I love science.  In fact I was a scientist once myself (research chemist).  What you are saying is the unfortunate confluence of political, journalistic and scientific inanities.

On the political and journalistic ends, you have people that know nothing about the scientific method wanting to write big checks for things they don’t understand.  And on the scientific end, you have scientists playing in their small arcane sandboxes claiming that the mostly irrelevant work they do has social or economic value.

First of all, the scientific method.  I think I stated this before to you guys.  Discovery is mostly a serial process.  I.e. future experimentation is conditioned on the results of experiments that came before it.  If an experimental design takes two years to work through, it takes two years. E.g., You can’t compress a 2 month incubation process by doubling the temperature of the reactor.  Throwing money at a scientific initiative that is already adequately funded will only buy additional lab junk that won't be used and subsidize the conference junkets investigators love to take to hobnob with their peers.

Secondly, let me be redundant some more.  The guys in science may be as self-interested as anybody else, but they ain't stupid.  Funds for R&D are distributed based on recommendations from expert panels.  That happens in both government and industry.  Of course politics comes into play, but much less so than in nonscientific domains.  What that means is that the best ideas are probably already being funded.  So throwing more money at R&D will mostly buy lowehr value projects.  I.e., ones that have high probabilities of scientific and/or economic failure.

Let me say one more thing about energy research.  Unlike biological sciences in which we still have much to understand, the fundamental science associated with energy research is well understood.  We have a clear understanding of materials science and the other important limiting factor, simple thermodynamics.  E.g., there is an upper bound on engine efficiency.  And ambient temperature super-conductors ain't gonna happen.

So unlike computer science or biology, there is no energy equivalent of Moore's Law.  There are no transformational discoveries (except for fusion perhaps) to be made related to energy.  By that I mean no orders of magnitude or even significant multiples of efficiency or energy density are going to be found because nature won't let it happen..

Mix the naïve democratic politicians with dopey journalists like Thomas Friedman and hold on to your wallet.

Me: I know this will come as a crushing disappointment to some none, but I am not an expert in energy sciences. Still, I am always skeptical of "there's nothing left to discover" talk. Still, I thought the email was interesting and I generally agree with its thrust.

04/27 01:11 PMShare