I did a little experiment with my home PC today to check out its power consumption. Nothing too precise - just checked the rate of consumption on the electricity meter with the PC switched on and with it off. It seems my electronic box of tricks uses about one unit (1 kWhr) of electricity every 11 hours or so. So my theory about individual digital information files becoming rivalrous, in the context of the second law of thermodynamics doesn't hold a lot of water.
If we start looking at it from a digital eco/infosystem perspective, however, things get a little more interesting. Suppose twenty million homes in the UK run an average of one PC for 6 hours a day. That's roughly 10GWhr or 10 million units of electricity. The second law says there is wasted energy in power generation (40% efficiency was pretty good for a power plant when I was studying mechanical engineering many moons ago), transmission and end user devices such as PCs. Guesstimate 10% of the potential energy in the material sustaining the power plant comes out as useful energy in the digital ecosystem and that means we need 100GWhr worth of coal, oil, gas, nuclear, renewables etc. to just keep it ticking over.* The individual digital files themselves don't contribute a lot to global warming but the architecture of the technological infosystem required for those files to exist is massively wasteful of our diminishing energy resources (as the Google, Sun et al energy bills already demonstrate). The second law of thermodynamics says there's no such thing as a free lunch in the energy sphere (excuse the mixed metaphor) and more efficient technologies ultimately won't solve the problem of increasing consumption. They can slow down the rate of deterioration and it's time we started paying much more attention to initiatives like Ndiyo, not only for their obvious inherent value but for their potential to reduce the thirst for energy of our wasteful eco/infosystem.
*Health warning on the numbers - they are about as reliable as those bandied about by the music industry or the government when trying to sell one of their crazy schemes, so absolutely no reliance should be place on them. I'm hoping to do some research to dig out some numbers with a more solid empirical base sometime over the summer, when I've got the first draft of the book completed.