Thursday, 14 May 2009

Deindustrial Revolution?

Thought provoking post over at thearchdruidreport about the implications for declining fossil fuel energy reserves on the viability of today's IT infrastructure. Below are a few samples, but well worth reading the whole thing. See also the climate group report I covered last year, confirming the collossal growth in the energy footprint and emissions of this sector: "The ICT sector’s own emissions are expected to increase, in a business as usual (BAU) scenario, from 0.53 billion tonnes (Gt) carbon dioxide equivalent (CO2e) in 2002 to 1.43 GtCO2e in 2020." Greer's book the long descent looks very well worth checking too. I've always found peak oil one of the most liberating of our looming problems (how 'looming' is a debate, but it's still an inbuilt mortality to our oil fuel technocratic society) because it forces you to think far beyond the status quo.

Very few people realize just how extravagant the intake of resources to maintain the information economy actually is. The energy cost to run a home computer is modest enough that it’s easy to forget, for example, that the two big server farms that keep Yahoo’s family of web services online use more electricity between them than all the televisions on Earth put together. Multiply that out by the tens of thousands of server farms that keep today’s online economy going, and the hundreds of other energy-intensive activities that go into the internet, and it may start to become clear how much energy goes into putting these words onto the screen where you’re reading them. It’s not an accident that the internet came into existence during the last hurrah of the age of cheap energy, the quarter century between 1980 and 2005 when the price of energy dropped to the lowest levels in human history. Only in a period where energy was quite literally too cheap to bother conserving could so energy-intensive an information network be constructed. The problem here, of course, is that the conditions that made the cheap abundant energy of that quarter century have already come to an end, and the economics of the internet take on a very different shape as energy becomes scarce and expensive again. Like the railroads of the future mentioned earlier in this post, the internet is subject to the laws of supply and demand. Once the cost of maintaining it in its current form outstrips the income that can be generated by it, it becomes a losing proposition, and cheaper modes of information storage and delivery will begin to replace it in its more marginal uses.

Governments will have very good reasons to maintain some form of internet as long as they can, even when it becomes an economic sink – it’s worth remembering that the internet we now have evolved out of a US government network meant to provide communication capacity in the event of nuclear war – but this does not mean that everyone in the industrial world will have the same access they do today. Instead, as energy costs move unsteadily upward and resource needs increasingly get met, or not, on the basis of urgency, expect access costs to rise, government regulation to increase, internet commerce to be subject to increasing taxation, and rural areas and poor neighborhoods to lose internet service altogether. There may well still be an internet a quarter century from now, but it will likely cost much more, reach far fewer people, and have only a limited resemblance to the free-for-all that exists today. Newspapers, radio, and television all moved from a growth phase of wild diversity and limited regulation to a mature phase of vast monopolies with tightly controlled content; even in the absence of energy limits, the internet would be likely to follow the same trajectory, and the rising costs imposed by the end of cheap energy bid fair to shift that process into overdrive.

1 comment:

danburgess said...

John, that yahoo server stat is frightening. wow.