For a few years, Google Book Search has provided PDF downloads of public domain books. The books came with a page listing some guidelines that Google asked people to follow and the same guidelines are listed in their Google Book Search help center. The old guidelines were as follows:
The new guidelines are as follows – I highlighted the changes:
The main change is the first guideline – it used to say that only non-commercial use is allows. Now it has been replaced by two new guidelines: no hosting, and no reprints including helping people reprint.
An interesting wrinkle about the new hosting restriction is that the Internet Archive is currently hosting about 537,000 PDFs of public domain books from Google Book Search. Under the old rules, non-commercial hosting was ok. What is the story under the new guidelines?
The new no reprint guideline seems to be directed towards services like my own PublicDomainReprints.org with the second part “help other people buy and sell them” perhaps directed towards POD providers like Lightning Source and BookSurge, and e-commerce sites like Lulu and Amazon that assist in the process. It remains to be seen what the story is under the new rules and I am evaluating options as far as my service is concerned.
One thing to consider – are these retroactive? Can a company put out public domain material with a notice in the material itself, and then go ahead and change the terms?
Of course, the elephant in the room is that these books are in public domain and thus have no copyrights. Without significant creative change it would not be possible to re-assert copyrights over the public domain scans (sweat of the brow has been struct down back in 1992). Whatever terms apply are being pushed via contracts and not the tradition route of copyright licensing. This may or may not similar to what the OCLC has been recently doing by trying to enforce contract rules on stuff that cannot be copyrighted. Can a contract override federal copyright law, placing a public domain book under someone else’s legal power?
It appears that at least some public domain content within Google is already hidden behind protection. The recent addition of magazine to Google Book Search brought along some magazines which have been published before 1923, placing those issues into public domain. If you run a search for magazines published before 1923, quite a few come up. But none of them have PDF downloads and all of them say they are being displayed with permission of some magazine publisher. It seems that the publisher provided them under contract with restrictions, thus forcing Google to carry over the restrictions. But what if the same magazine is scanned at a library? Will Google allow the second version to appear with full PDF download?
One other side effect of this is that if the content in question is not copyrighted, than in theory the DMCA restriction on reversing DRM should not apply since it only applies to copyrighted material. It would still be illegal since hacking into someone else’s website is forbidden and no one should try it. But in theory, does reversing DRM as codified in DMCA apply for public domain content?
Another interesting point is that these seem to be voluntary guidelines. The legal terms of Google’s site and any of the Book search site never state these terms, and both inside the book and in the FAQ they are cauched in a language that ” we do ask that you follow some basic guidelines regarding their use”. Is “we do ask” a legal request? IANAL, but perhaps not – may be they are simply there to scare people off?
Lots of questions, and a few answers.
P.S. And there is the small matter of the Google / AuthorGuild settlement which is going to have hearings in front of a judge. A settlement that someone can object to.