The Chronicle just posted an article on Big Data:
"Crowdsourcing is a natural solution to many of the problems that scientists are dealing with that involve massive amounts of data," says Haym Hirsh, director of the Division of Information and Intelligent Systems at the NSF. Findings have just grown too voluminous and complex for traditional methods, which consisted of storing numbers in spreadsheets to be read by one person, says Edward Lazowska, a computer scientist and director of the UW eScience Institute. So vast data-storage warehouses, accessible to many researchers, are going up in several scholarly fields to try to keep track of the wealth of information. Persuading scientists to fully embrace the age of big data, though, will require a change in academic reward structures to give new currency to papers with more authors than ever and to scientists who spend their careers crunching other peoples' numbers."
... and yet another:
"Data-diggers are gunning to debunk old claims based on 'anecdotal' evidence and answer once-impossible questions about the evolution of ideas, language, and culture. ... Google has changed the landscape. Pouring hundreds of millions into digitization, the company did in a few years what Mr. Unsworth believes would have taken libraries decades: It has digitized over 12M books in over 300 languages, more than 10% of all books ever printed.""[Yet] some worry that the lure of money and technology will increasingly push computation front and center. ... whether transferring the lab model to a discipline like literary studies really works. Trumpener is dubious. Twenty postdocs carrying out one person's vision? She fears an "academia on autopilot," generating lots of research "without necessarily sharp critical intelligences guiding every phase of it."