[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Google Book Search and fair use

Sandy, my point hinges on a distinction between digital copies of a printed page and digital copies of a text. What is significant about the Google process is that they are making copies of the text, as well as copies of the page (they give the copy of the page back to the library, but they keep for themselves the much more valuable 'interpreted' digital copy of the text). There are lots of ways of producing useful digital copies of a text, but most of the efficient ones use an electronic file as the source (or something similar). This copying can be managed by the publisher or the rights holder (and publishers are rightly quite concerned over the distribution of the PDF file). Google for the library project has not gone to publishers for 'source files from the text' they have generated their own digital copies (interpretations -- and with mistakes) of original texts. The Google process requires a lot of investment and a big software system. But disseminated low-cost ways of digitizing texts are coming.

What happened to music about 30 years ago, and is now starting to happen with books, is that it is quite feasible to produce useful copies from the physical product. You dont need the source file. I predict that there will soon be lots of ways of producing (image scan->OCR->text database) usable digital copies of any printed text. The chances are someone is working on how to do this with an iPhone right now. Point the iPhone at your book, flip the pages, and *hey presto* you have your enhanced text database, accessible from and fed by your iPhone.

CDs still have the rubric (similar to that which you find on books) that "Unauthorised copying, duplication, hiring, broadcasting ... etc is prohibited." But nobody (including all the book publishers of my acquaintance) ever seeks authorisation before copying for their iPod, their home music centre, or before sundry other things that may or may not be authorised by the music publishers. Because copying the physical copy is so easy (and so useful) music companies have had to abandon the fiction of opt in permissions for copying music. They still print the rubric on the sleeve of the CD, but I bet they dont get many requests for digital copies from end-users.

If book publishers judge that maintaining the fiction of an 'opt in' permissions system will work when it is so easy and so useful to make databases of books from physical volumes, they have a rude shock coming.

Adam


On 10 Jul 2008, at 00:04, Sandy Thatcher wrote:

I'm puzzled by your last paragraph, Adam. By your reference to music, you seem to be alluding to the compulsory licensing regime that exists for music in certain of its uses. But that is not an "opt out" regime because copyright owners have no say at all in who uses their music; the only obligation the user has to the owner is to pay a mandated royalty set by a regulatory body established by law.

Also, there are already "scores of different ways of digitizing texts," yet we publishers seem to have managed just fine under an "opt in" procedure in dealing with them all. So, why does sheer volume justify switching to an "opt out" regime?

By the way, instituting a compulsory licensing regime will not forestall piracy: pirates will continue to do what they do and ignore any requirement for paying royalties, just as they now ignore seeking permission.

Sandy Thatcher
Penn State University Press


Some portions of the Koehler article make it very clear that he is
well aware that 'snippets' are not the issue:

"Yet many of us are uncomfortable with the Google plan. It looks
like an unfair taking of intellectual property. Google tells us
that it will only serve up snippets. But we need to remember that
Google is serving up snippets from the whole thing copied from
the original. That "whole thing" may well be a copy of a work
still in copyright." (from the penultimate section of his piece
"I am not a lawyer")

Perhaps the most troubling aspect of the 'opt out' solution
proposed by Google is that it has operated as though it has the
sole right to determine the 'opt out' procedure. That cannot be a
fair and reasonable solution; its a nightmarish prospect for
rights owners since there could be any number of different
methods of secondary digitisation which involve a myriad of
different 'opt out' protocols. At least the robots.txt (which was
designed for web sites not for printed materials)  is a
convention which is general and accepted by the industry.

I have no doubt that there will be scores of different ways of
digitizing texts in 20 years time and lots of digital formats.
When digital copies are easily and trivially made from physical
copies an 'opt in' permission has no chance of prevailing (see
the fate of 'opt in' for recorded music in the last 30 years). In
such circumstances an opt out procedure is going to be preferable
for publishers and authors to an 'anything goes' principle. But
the opt out procedure must be principled, fair and balanced.

Adam