We received a deluge of comments about my response to the false proposition that it takes more energy to make and ride a bike than to make and drive a car. Most heartily agreed that any idiot can see it takes less energy to move a 35-pound bike than a 1.5-ton car. Some pointed out the health benefits of cycling; others brought up shining examples of commuters who have logged thousands of miles biking to work. Some said it was idiotic to expect people to commute by bike, and some just said everybody involved in the discussion was an idiot.
The smartest readers caught a very embarrassing error. I said, wrongly, that it would take only the amount of energy in a half-pint of vodka to provide enough calories for a cyclist to pedal 48 miles. The fact is, it would take about a quart of vodka. I simply misread the USDA's nutrition data chart. The only and quite feeble excuse is that the misreading may have been caused by rigid enforcement of my paperless office policy--reading on-screen, no matter how big you blow up the text to combat presbyopia, does seem to propagate errors. (By the way, the USDA’s National Nutrient Database is a wonderful and sometimes surprise-filled source of information on food.)
Despite that mistake, the main point--that you're better off drinking ethanol and burning it by pedaling than by running an engine--is still valid. As noted, it would take about two gallons of pure ethanol to propel the average car the stated distance. This means the biker is 16 times more efficient than the driver (since a quart of vodka only contains a pint of pure ethanol).
Finally, while it's obviously crazy to expect people to commute 30 miles or more by bike, it's not impossible to get into a carpool. Carpooling may not provide the exercise celebrated by many readers, but it does go a long way to reducing commute costs and energy use. You can search for ride sharers at eRideShare.com or one of the many sites that offer carpool info. Let me know if it worked out for you.