Open Access News

News from the open access movement

Thursday, January 11, 2007

Testing Google's restrictions on Google-scanned public-domain books

Philipp Lenssen, Freeing Google Books, Google Blogoscoped, January 10, 2007. 

Google Book Search allows you to download full PDFs of books which are in the copyright-free zone of the public domain. I think Google, as well as the libraries which offered their books to be scanned by Google, deserve credit for this. However, Google wants to impose some restrictions for those books, and in their usage guidelines ask you to:

Make non-commercial use of the files ... we request that you use these files for personal, non-commercial purposes. ... Maintain attribution ... Please do not remove [the Google "watermark"].

But Wikipedia defines the public domain as (my emphasis) "...the body of knowledge and innovation (especially creative works such as writing, art, music, and inventions) in relation to which no person or other legal entity can establish or maintain proprietary interests within a particular legal jurisdiction. This body of information and creativity is considered to be part of a common cultural and intellectual heritage, which, in general, anyone may use or exploit, whether for commercial or non-commercial purposes." ...

In other words, Google imposes restrictions on these books which the public domain does not impose. Iím no lawyer, and maybe Google can print whatever guidelines they want onto those books... and being no lawyer, most people wonít know if the guidelines are a polite request, or legally enforceable terms**. But as a proof of concept Ė the concept of the public domain Ė Iíve now ďset freeĒ 100 books I downloaded from Google Book Search by republishing them on my public domain books site, Authorama. Iím not doing this out of disrespect for the Google Books program (which I think is cool, and Iíll credit Google on Authorama) but out of respect for the public domain (which I think is even cooler). Itís great to have Google mirror & distribute our culture, and itíll be even greater if we join in and help them with the mirroring & distribution.

Thanks to Charles Bailey at DigitalKoans for alerting me to this post.  Charles adds these comments:

Since Lenssen has retained Googleís usage guidelines in the e-books, itís unclear how they have been "set free," in spite of the following statement on Authoramaís Books from Google Book Search page:

The following books were downloaded from Google Book Search and are made available here as public domain. You can download, republish, mix and mash these books, for private or public, commercial or non-commercial use.

Leaving aside the above statement, Lenssenís action appears to violate the following Google usage guideline, where Google asks that users:

Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for personal, non-commercial purposes.

However, in the above guideline, Google uses the word "request," which suggests voluntary, rather than mandatory, compliance....Note the use of the word "please." ...

So, do Googleís public domain books really need to be "set free"? In its usage guidelines, Google appears to make compliance requests, not compliance requirements. Are such requests binding or not? If so, the language could be clearer....

Comment.  There's no doubt that Google puts some restrictions on its scanned public-domain books, and I've complained about these in the past.  OCA puts fewer restrictions on its scanned public-domain books, and I've always preferred its access policy to Google's.  Charles is right that Google isn't very clear on whether its requested restrictions are binding or even supposed to be binding.  I suspect that in US copyright law there's an arguable but still-murky distinction between the status of a public-domain text and the status of a newly-made digital file of a public-domain text, just as there is between a public-domain painting and a new photograph of that painting.  If so, then there might be three reasons why Google used the language of "request" rather than stronger language to describe the restrictions it wanted to impose:  (1) it didn't want to use stronger language; (2) the law clearly doesn't allow it to use stronger language; or (3) the law is unclear.  We may find out soon.

Update. Danny Sullivan has commented on Lenssen's project and added some new information from Google:

I'm checking with Google to see what they think about the project and the legality of trying to impose restrictions on public domain books, just because they've scanned them.

Postscript: I've now heard back from Google, which says:

We have gotten this question in the past. The front matter of our PDF books is not a EULA [end user license agreement]. We make some requests, but we are not trying to legally bind users to those requests. We've spent (and will continue to spend) a lot of time and money on Book Search, and we hope users will respect that effort and not use these files in ways that make it harder for us to justify that expense (for example, by setting up the ACME Public Domain PDF Download service that charges users a buck a book and includes malware in the download). Rather than using the front matter to convey legal restrictions, we are attempting to use it to convey what we hope to be the proper netiquette for the use of these files.