Abstract versus the Concrete

Dry Dipstick has made an intriguing though ultimately unimportant comparative analysis between "failed" Y2K doomsayers and oil depletion gurus. Although he does admit to being burned out from getting sucked into the pre-2000 hysteria, any rhetorical argument placing an abstraction problem (Y2K) against a concrete problem (oil) on the same playing field has serious flaws.

But many of the now-retired Y2K people do know about Peak Oil. They do know what's going on. And they do see the dangers inherent in Peak Oil. But are they going to come out of retirement, dig their soapboxes out of the attic or garage, stand tall in the public square and once again warn that their communities are imperiled? After what happened last time? How much credibility do you think they have left as apocalyptic prophets? Could they preach to the multitudes and proclaim the Second Coming of Potential Disaster? Maybe even proclaim that the Bush Administration, while it might have made the wrong choice, at least had somewhat—God help us—noble intentions. (Or at least quasi-noble.)

Not bloody likely.

I claim that the whole Y2K situation had in its origins a lack of abstraction skill level and lots of premature optimization (due to perceived savings in storage costs) from a limited suite of 60's and 70's era software suites. The big flaw in the argument is that many industries had been using software development languages with the right abstraction level for at least a dozen years previous to the year 2000.

For example, recent military software has used Ada, which had standardized on a built-in Calendar abstraction to its base library system since its inception. Not only was there a suitable year range designed into its opaque Time type, but the whole backend implementation could be changed without an impact to the user's software. Many refer to this as the power of encapsulation.

Unlike oil, there are no shortage of abstractions in the software world. Doing an abstraction right, a software engineer can slip in a replacement behind the scenes, with no one the wiser.

That ain't going to happen after the peak. The whole concrete infrastructure will have to change.