Virtual Consumer Mar-Apr 2009

Storm in a teacup

Column for Ethical Consumer magazine Mar-Apr 2009

Aside from the arguments about accuracy, Wikipedia has one very big advantage over a conventional encyclopaedia: the ability to search the entire text and find almost instantly what you are looking for. For anyone over a certain age that is remarkable. But even more remarkable are the search engines like Yahoo and Google that pre-date Wikipedia, capable of searching billions of websites for words and phrases in a fraction of a second. With the advent of the “semantic web”, when search engines will be able to look for meanings rather than words in text, they will come into their own, eliminating the need to organise on-line knowledge in metaphors for old-world concepts like encyclopaedias. The combination of the net and search is democratising access to knowledge.

So for anyone who shares my belief in the progressive power of the net, the story claiming “Google and you'll damage the planet” (Sunday Times, London, Jan 11) may have come as a shock. The article, which was picked up by the BBC and news outlets and bloggers all over the net, went on to assert that according to research by Harvard physicist Alex Wissner-Gross, performing two Google searches used the same energy as “boiling a kettle for a cup of tea” and so was responsible for 14g of CO2.

The implications of this are truly startling: it takes a lot of energy to boil a kettle. Using government figures I calculated that, at 7g per search, just clicking the Google search button once a second would generate the same amount of CO2 as driving a medium sized car at 60mph. We have become used to “googling” freely – many people instead of typing a web address into the address bar of their browser will type it into Google instead, presumably wasting another 7g in the process.

This column has already criticised some of the “facts” concerning the environmental impact of IT that you find circulating in the media and on the net. They are often the result of dubious science, or a very poor level of understanding on the part of journalists. One assumes though that this “Harvard physicist” knows what he’s talking about.

Well, indeed he does, but it seems to me that the Sunday Times article was very misleading.

The first reaction to the story came from Alex Wissner-Gross himself, who protested that he hadn’t mentioned Google and had said nothing about boiling kettles. Then, predictably, Google waded in, pointing out that they take data centre efficiency very seriously (they do) and that their own calculations showed that a single search required just 1kJ of energy – equivalent to 0.2g of CO2. This is actually rather a lot, but it is 35 times smaller than suggested by the Sunday Times.

The Sunday Times has since produced a “clarification” where they accept Google’s figures and claim they were “referring to a Google search that may involve several attempts to find the object being sought and that may last for several minutes”. This they say results in a carbon consumption of between 1g and 10g according to “experts”. Not 7g then.

I wonder how many people heard “Google search” and thought “oh, that means a process involving several minutes of Google searching”. And I wonder how many times we’ll be hearing “7g for a search” repeated as established fact.