Google Book Search is an enormous asset for consumers and may one day hold much of the collective knowledge of humanity: http://fishtrain.com/2007/11/30/google-book-search-wealth-of-giants/

Keep in mind an important distinction. Traditional copyright infringement only robs the author once, taking the royalties he might have gotten. Making his book available online for free robs him twice. 1. No one is getting any money, so there’s no money to purse in court as damages. 2. No one is likely to want to bring back into print a book of little importance that available for free online. The author not only gets nothing, he will never be able to get a penny for his labor. That’s bad, very, very bad. Google really is robbing authors to enrich themselves.

Google and others are missing another point of copyright law. The law says that to publish someone’s book you need their permission. It doesn’t say that you can make some vague announcement to the world that you’re going to publish anyone’s work you want, requiring them on their own time to find you and tell you no. Copyright is the right to determine who copies, not the right to search out copiers and tell them to stop. In the latter case, there are court decisions beyond number that infringement has already occurred.

Claims that searchable full-text databases are fair use can’t survive even the most cursory comparison with a technology that’s newer than books–movies, but nevertheless covered by copyright. A movie repackages the content in a different format, but otherwise builds on the original. Even if most of the content is removed due to time constraints and even though a lot of additional creativity is required to create a movie, it remains a derivative that must be licensed. That’s well-established law.

Searchable databases are even more closely linked to the original than a movie and one can’t argue for an exception based on nothing more than their mere newness. The content is identical and the mechanism for scanning and OCRing requires no creativity. If fact it’s almost always done by machines. If OCRing would break a copyright by rendering it different in a mere technical way, then so does using a copy machine, and the law says that isn’t so.

And yes, there is an enormous problem with locating a copyright holder of a long out-of-print book and getting their permission. But that’s where Google and Microsoft should be using their considerable political muscle to get the problems of orphan works written into the law. They can’t just assume that because a problem exists, they have the right to impose a solution of their own choosing.

–Michael W. Perry, author of Untangling Tolkien