“It makes no sense for China to have better rail systems than us, and Singapore having better airports than us. And we just learned that China now has the fastest supercomputer on Earth — that used to be us.” (Nov 3, 2010)
“America became an economic superpower because we knew how to build things. We built the Golden Gate Bridge, and the Hoover Dam, and the Interstate Highway System. And now, we’re settling for China having the best high-speed rail, and Singapore having better airports? When did that happen? “(Oct 25 2011)
George Savage juxtaposes the latter Obama statement with his decision, only two weeks later, to delay approval for the construction of a Canada-to-Texas oil pipeline, which was estimated to provide about 20,000 jobs, as well as having an obvious beneficial impact on America’s energy security. Indeed, it should be obvious at this point that the main inhibitors to the building of any large project whatsoever are regulatory overreach and complexity and the exploitation of the legal and regulatory environment by precisely the kind of activists that Obama the community organizer has spent much of his life encouraging. Obama’s complaints about us not building things resemble the plea of the defendant who killed both of his parents and then asked for mercy because he was an orphan. (More thoughts on large projects then versus now at my post like swimming in glue.)
But in addition to the above point, the kinds of projects about which Obama waxes enthusiastic (to the degree that any enthusiasm is contained in his rather flat emotional range) reveal much about the “progressive” economic worldview.
The president wants to talk about trains? Okay, let’s talk about trains. beginning with the fact that US freight railroads carried 1.7 trillion ton-miles in 2008. This is more than twice the ton-mileage hauled in 1980, the year that the Staggers Act was passed and the remarkable renaissance of the US railroad industry began. Efficient freight rail offers a huge economic advantage to a country, and knowledgeable people tend to agree that the US industry is the most efficient in the world. But of course freight isn’t as glamorous as high-speed passenger trains, and “progressives” generally seem much less-interested in the workaday world of gondola cars and containers than in emulating the passenger facilities offered by the European countries with which they are so enthralled. Indeed, despite his frequent use of the word “infrastructure,” I have to question whether Obama has the slightest knowledge of or even curiosity about the way goods are transported in this country and elsewhere in the world. See Coyote Blog’s related post about shifting capital from the productive to the sexy.
The president wants to talk about computers? Okay, let’s talk about computers. It is possible that the Tianhe-1A supercomputer located in Tianjin is indeed the fastest computer in the world—but its ability to use this potential performance is dependent on dividing the work for a problem in a way that keeps most of its 7,168 Nvidia Tesla M2050 GPUs and its 14,336 Intel Xeon CPUs usefully busy. This is a software problem, not a hardware one. (And as far as hardware goes, Nvidia and Intel are of course both US companies.) Without attempting to take anything away from the Chinese achievement in building the Tianhe-1A, I’d suggest that providing the GPU/CPU components on which it is based, and doing the kind of system design work that IBM did for its much-hyped Watson system, represent technological contributions at least as significant as the Tianhe machine. Moreover, US researchers with heavy computational needs will benefit immensely from the kind of on-demand processing (“cloud”) infrastructures being put in place by Amazon and others.
One example of an important problem which is computationally-intensive is the modeling of protein folding, which is important in biomedical research. And one way to solve it is by assigning the task to very large supercomputers. But another way, described in this article, is to bring in the resources of thousands of individual home computers and gaming machines and the human skills of their volunteer owners. The latter approach recently resulted in the solution in three weeks of a folding problem that had remained unsolved for more than a decade.
The obsession with giantism, with prestige projects, and with top-down direction was of course a defining characteristic of the old Soviet Union, and had much to do with its economic failure. It is an approach which is destructive not only of a society’s productivity but also and more importantly of its spirit.