Jennifer Howard from the Chronicle of Higher Education tells the story about the recent 'negotiations' between Elsevier and the University of British Columbia regarding text-mining rights:
http://chronicle.com/article/Elsevier-Experiments-With/131789/
As many of you are aware, large-scale text-mining for most of the STM literature is blocked as much by a stew of legal obfuscation and misinformation as by technical barriers. This story is significant for bringing the historically secretive terms of agreement between publishers and libraries out into the open, illustrating just how unscalable it is to negotiate text-mining access on a case-by-case basis, and - most sadly - exposing how little concern publishers and research libraries have historically had for the rights of readers when those readers are machines.
Todd Vision
University of North Carolina at Chapel Hill