The End of the way we do knowledge

Joel Garreau's response to an article by Chris Anderson outlining how petabyet computing may fundamentally change the way we do knowledge contains the following important insight. "Data are an artifact of selection, which means they reflect an underlying hypothesis or they wouldn't have been collected. For example, in my work I discovered a frightening lack of timely data to 'prove' either my hypothesis that North America was behaving as if it were nine separate civilizations or economies that rarely were bounded by political jurisdictions of nation, state, or county. It was equally problematic coming up with the data to prove that places like Silicon Valley were becoming the modern iteration of 'city' even when the millions of square feet of big buildings were right before your eyes. It wasn't until those 'nine nations' or 'edge city' models began to be seen as useful by others that people started to go through the very great deal of trouble to verify them by collecting data in a fashion that ignored all previous boundaries. Life is not obligated to follow data and it frequently does not."
I've recently been musing on this as I'm reflecting on the models of theological reflection represented by the pastoral cycle and similar. And as I pointed out in a conversation with a student regarding writing a reflection, there is no uninterpreted data; the very selection of data implies a weighing of merit against stated or unstated criteria. It is this that the petabyte age may challenge and allow us to sidestep more fully. Sean Carroll summarises the potential import: "giant new petascale datasets that resist ordinary modes of analysis, but which we can use to uncover heretofore unexpected patterns lurking within torrents of information".
That said, we're still going to need to be able to hypothesise the correlations thrown up, but it's a potentially creative world opening up.
Chris Anderson's article.
Page of responses.