The Association of American Publishers and Google’s
agreement last week to settle the publishers’ long-running
litigation over Google’s library scanning program put an
end to the lengthy and expensive suit, but without
resolving any of the underlying copyright and fair use
issues. The main component of the deal: copyright owners
with books scanned by Google under its library program can
choose to “opt out” of the program and have their books
removed. Of course, that settlement condition is something
Google has offered copyright owners with books in the
program all along.
In other words, Google scanned and posted to the web first, and authors could opt-out if they so chose. So they didn't get consent first. Their rationale was good - there was no machine readable contact information and many of the authors were dead or had moved - but in this respect the situations are reasonably analogous.
Had Aaron scraped say 10-100k rather than 4M articles and done some innovative NLP/data mining on the text (like Google N-grams), then showed that to JSTOR, despite the lack of up-front consent he might well have been able to get more or convince them to open up, especially if this 10-100k experiment showed monetization potential and could allow JSTOR to move to an open access model. Sadly, we'll never know.
http://www.publishersweekly.com/pw/by-topic/digital/copyrigh...
In other words, Google scanned and posted to the web first, and authors could opt-out if they so chose. So they didn't get consent first. Their rationale was good - there was no machine readable contact information and many of the authors were dead or had moved - but in this respect the situations are reasonably analogous.Had Aaron scraped say 10-100k rather than 4M articles and done some innovative NLP/data mining on the text (like Google N-grams), then showed that to JSTOR, despite the lack of up-front consent he might well have been able to get more or convince them to open up, especially if this 10-100k experiment showed monetization potential and could allow JSTOR to move to an open access model. Sadly, we'll never know.