"Copyfraud" and Google Books

The Register and Slashdot have picked up a theme from a 2006 law review arti­cle by Jason Mazzone on “copy­fraud,” extend­ing the idea to explain a new incar­na­tion of it emerg­ing in rela­tion to Google Books. Mazzone wrote:

Copyfraud is every­where. False copy­right notices appear on mod­ern reprints of Shakespeare’s plays, Beethoven’s piano scores, greet­ing card ver­sions of Monet’s Water Lilies, and even the U.S. Constitution. Archives claim blan­ket copy­right in every­thing in their col­lec­tions. Vendors of micro­filmed ver­sions of his­tor­i­cal news­pa­pers assert copy­right own­er­ship. These false copy­right claims, which are often accom­pa­nied by threat­ened lit­i­ga­tion for repro­duc­ing a work with­out the owner’s per­mis­sion, result in users seek­ing licenses and pay­ing fees to repro­duce works that are free for every­one to use.

via Copyfraud by Jason Mazzone on SSRN.

The arti­cle on the Register describes a newer player in this scheme: Google Books.

Kessinger [a pub­lisher] made the [pub­lic domain] doc­u­ment use­less to schol­ars, to force them to pur­chase the full hard­copy edi­tion for $25. Links on the Google Books page directed pur­chasers to the Kessinger edi­tion on Amazon​.com and other online book­sellers. Scholars were out­raged. These works are clearly in the pub­lic domain, dat­ing back to the 1890s and beyond.

When ques­tioned, Google said it “must err on the side of cau­tion… until we have deter­mined that the book has entered the pub­lic domain.” But with the sheer vol­ume of ebooks being sub­mit­ted by out­side pub­lish­ers, there are obvi­ous delays in clear­ing rights. Some pub­lish­ers have exploited this gap, pro­vid­ing copy­fraud edi­tions where no free edi­tion was available.

via Copyfraud: Poisining the pub­lic domain from The Register.

Legally, though frus­trat­ing, the sit­u­a­tion is not always quite as clear cut a case of “steal­ing” from the pub­lic domain and “defraud­ing” the gen­eral pub­lic. Certainly, the process out­lined in the Register arti­cle is essen­tialy that:

Committing copy­fraud is aston­ish­ingly easy and costs noth­ing. I can bor­row a pub­lic domain book from any library and scan it, or I could down­load the text from Project Gutenberg. I refor­mat it as a PDF, mark it with a copy­right date, reg­is­ter it as a new book with an ISBN, then sub­mit it to Amazon​.com for sale. I may not even need to print and bind any books, I can offer it through Amazon’s Booksurge print-​​on-​​demand ser­vice, or as an ebook on Kindle. Once the book is listed for sale, I can sub­mit it to Google Books for inclu­sion in its index. I could eas­ily pub­lish thou­sands of books; most would never sell, but with zero up-​​front cost, any sale is pure profit.

The key part of the above scheme is the scan­ning or copy­ing of text directly, adding noth­ing new (or very lit­tle new, any­way — not enough to estab­lish a copy­right any­way). With no new value added (in loose terms — copy­right is more com­pli­cated than this), there is no “new” work at all. Thus, there is no valid copy­right. I can still sell these works, of course, but Google should not deny full access to them.

Re-​​typesetting a book (and pos­si­bly edit­ing it, fix­ing errors, and gen­er­ally adding value) is a dif­fer­ent story. If a pub­lisher does enough new to the book, it’s very pos­si­ble that doing so would gen­er­ate a copy­right on the newly type­set book — but still not on the text itself. That would remain in the pub­lic domain, free to be type­set by some­one else, quoted from at any length, or oth­er­wise shared. But copy­right — at least in a “thin” form — would, I believe, likely pro­tect against straight-​​up pho­to­copy­ing or scan­ning of the newly pub­lished book itself, and that would then likely include that par­tic­u­lar edi­tion in Google Books as well.

“Thin copy­right” cov­ers works that, for exam­ple, add only lim­ited addi­tional value to a pub­lic domain work. The Electronic Frontier Foundation explains:

[W]here copy­right own­ers have a thin copy­right — for instance, where they choose to release a com­pi­la­tion DVD with a pub­lic domain work bun­dled with works in which they do hold the copy­right. In either case, the copy­right owner would obtain, at best, a thin copy­right in the non– pub­lic domain ele­ments, but does not thereby obtain copy­right in an uncopy­rightable pub­lic domain work. As rec­og­nized by numer­ous cases, includ­ing the Supreme Court’s deci­sions in Harper & Row v. Nation Enterprises [471 US 539 (1981)] and Feist Publications, Inc. v. Rural Telephone Service Co., 499 U.S. 345 (1991), and the Ninth Circuit’s deci­sion in Sega v. Accolade, the pub­lic con­tin­ues to retain the right to access the uncopy­rightable parts of the com­pi­la­tion. An exemp­tion is required to allow con­sumers to exer­cise their right of access and to pre­vent copy­right own­ers from using tech­no­log­i­cal pro­tec­tion mea­sures as a boot­strap to extend their thin copy­rights over pub­lic domain works.

Remember: the law is pretty much always more com­pli­cated than it appears at first glance!

Related arti­cles by Zemanta

• Does Copyright Foster or Hinder Innovation? (inpro​pri​a​per​sona​.com)
• Random House dis­abling Kindle speech (inpro​pri​a​per​sona​.com)