October 23, 2003
Unfair Use? Amazon's Free Book Giveaway
by Brian DearAmazon.com launched a new service today, called Search Inside the Book. In a word: fantastic. For someone like me who's spent years doing research for a nonfiction history book, it's an incredible tool. This is the best thing on the web since Google unleashed a fully searchable Usenet archive dating back to 1982.
However, I am doubtful the service -- as it exists today -- will last long. It is too good..
Why is it too good? Because if you're determined, you can copy entire chapters out of books --- or, if you are really determined, entire books.
Here's how.
Case Study: Copying A Book From Amazon -- For Free.
I happen to be writing a book on the history and significance of the PLATO computer system. It has required a huge, time-consuming amount of research --- interviews, oral histories, emails, Google searching, archival documents, video and audio recordings, screen grabs, photos, travel around the country, you name it.
Along comes Amazon's new service. So I gave it a try. I typed in "plato computer" in the search field of Amazon's main Books page. This resulted in a long list of search results. Most of them contained individual results of their own --- full text results from inside the respective books.
So I selected a book, Case Studies in Computer Aided Learning, edited by C. Dianne Martin. Turns out Chapter Two, entitled PLATO Mathematics: The Teacher and Fourth Grade Students Respond, is, shall I say, a target-rich environment. So rich, that for all intents and purposes, the word "PLATO" appears on every page of the chapter.
Awesome. Page 53. Page 54. Page 55. Page 56. Hmm. I sensed a pattern.
Oooh. Then I noticed this at the bottom of the results:
Seven pages of links. Why, it's a huge chunk of the book.
So I clicked on a link, and saw this:
First thing I noticed was, of course, the green boxes. While it is impressive that Amazon was able to highlight -- no doubt in realtime? -- the actual search keyword hits on the actual pages of the book, I immediately wondered why they didn't use a half-oblique green shading instead of a solid box. That way you could see the text "behind" the box. Kind of like how Google highlights its search terms in the results pages (when you view Google's cache of a particular search result). But I'm not going to complain. This was awesome --- an actual page right out of the book!
But wait! There's more! Lots, lots more. Notice the "Previous Page" and "Next Page" links. Amazon lets you see two previous and two subsequent pages relative to the page you're on. Suddenly you're able to see five full-page scans out of a book you don't own. Didn't take long for the lightbulb to go off.
Think about it: copy all these scanned pages, return to the search results, grab more images, and after a while you have dozens, maybe scores, of pages out of a copyrighted book you don't own.
With my Safari browser on my Mac, I can just position the cursor of the mouse somewhere over the scanned image of the book's page, click the right mouse button (I use a 2-button mouse on my Mac), select the "Download Image to Disk" menu item, and bingo, the copyrighted page is saved to disk. Go to next page, repeat. Next page, repeat. Run out of pages, go back to search results. Select next unseen page, repeat the process and keep moving forward, slurping down as much of the book as you want.
Once you exhaust the search results for a book, just go to the last page you were able to grab, look for some interesting phrase or keywords, and search for that using the "Search inside this book" option. Amazon freshens your results and you're able to scoop up more of the book.
Unsettling Questions.
How in the world is this beneficial to publishers, let alone authors? What incents me to buy this book? Since I have essentially the entire text of the book at my fingertips for free, I don't see any reason to buy the book. Certainly not a $37.95 book that's on special order (read: prolly long out of print). But even more certainly the latest, hottest novel.
I don't know. All I know is that in this one case study, I am able to access huge chunks, if not all, of a $37.95 special-order book that until today would require my a) buying, or at least, b) going to the library and xeroxing. Both involve money. Not anymore. Now, thanks to Amazon's "Search Inside the Book," people have a way of getting portions or entire books saved to their personal computer's hard drive for later use -- either personal use or . . . gasp . . . republishing a la a sinister version of Brewster's Bookmobile. . .
How could this kind of feature possibly qualify as fair use of a copyrighted work? Amazon does not seem to place any burderns or restrictions in the way that would prevent such unfair use --- for that is surely what this is.
At the bottom of the scanned pages, note Amazon's disclaimer: "This copyrighted material is viewable by permission of the publisher." Now, did the publisher grant permission after collecting a fee from Amazon? Or is Amazon going to pay the publisher a fee based on actual viewings of these pages? Who pays Amazon?
How can it be a sustainable, profitable business for Amazon? If indeed there are any fees behind-the-scenes right now, the money would be coming from Amazon itself. The customer -- at least currently -- does not pay a fee for the privilege of viewing (or worse: saving to disk) these full-scan search result images. I paid no fee to use this service. (Or did I?)
I did, however, have to sign in, I noticed. So Amazon knows it's me doing these searches. Theoretically Amazon is collecting every page I view. Will I get a bill some day? Can they do that? It sure seems like this is not a completely-thought-out product. Released prematurely due to its insanely-greatness? Don't know.
I'm going to continue to explore this new service from Amazon and I'll follow up with more observations and questions. In the meantime, I welcome your constructive comments.
UPDATE: 6:30pm PT, 23 Oct 2003:
I noticed Amazon has provided a FAQ page for more information about the "Search Inside the Book" service. The one that is most interesting to me is this:
What exactly would be a negative experience for the "copyright holder"? Exactly what is going to happen to users who view too many pages out of a single book, through this service? And what would constitute too many pages? Amazon goes on in the FAQ to say: "We require credit card information for security purposes only. We will not charge your credit card account any fees for using the Search Inside the Book feature." Hmmmmmmm.
UPDATE: 3 Nov 2003:
See my next Nettle blog entry for an update on the Amazon search-inside-the-book service.
Posted by brian at October 23, 2003 10:08 PM
Hi Brian
Don't you think the fact that the text is only provided as an image (ie, not exportable as a text file to your favorite e-book reader) would tend to limit wholesale piracy? I can see your point about expensive, hard-to-come-by books for research, but I don't think that this would apply to the latest bestseller. The idea of reading an entire novel via slowly-loading images on Amazon doesn't appeal to me.
You can see my entry about it here
That would be here:
http://www.thelimeybrit.com/blog/archives/000456.php
According to the Wired magazine story (http://www.wired.com/news/business/0,1367,60948,00.html) you "won't be able to view more than a few thousand pages per month, or more than 20 percent of any single book."
Posted by: ML at October 24, 2003 02:06 PMInteresting comments --- and the WIRED article is very interesting as well. I am intrigued that Andrew notes that the green highlighting boxes are translucent in his browser. I am using Mac OS X, and I see opaque green boxes as shown int the screen grab above in the article. I see these boxes in Safari, Mozilla, and MSIE on MacOS X. Must be a MacOS X thing. I haven't tried looking at the Amazon search results on a Windows browser yet.
Posted by: Brian Dear at October 24, 2003 03:31 PM"Must be a MacOS X thing."
Apparently. I'm running WinXP, checked with Firebird 0.6 and IE6, and it's translucent on both of those browsers. You would have thought that Amazon would test stuff like this on a Mac platform before making it live.
Posted by: Andrew Duncalfe at October 24, 2003 11:47 PMI did a search on a knitting procedure discovered that, at least for the particular book I was looking at, images were not included in the page display, only text.
Also, it seems that this feature will give Amazon a lot of data about your personal interests, which it can then use to try to sell you things.
Posted by: Leslie Turek at October 25, 2003 04:21 AMBrian,
I think you're over-reacting, for two reasons.
First, while Amazon doesn't make a big deal of telling you up front, there are in fact limits to how much of a book you can see. The Wired article's 20% is consistent with what I've heard. Now a determined user could get around this by creating multiple identities, using multiple credit cards, but at that point, it's probably easier just to go out and buy the darn book. And, as someone else pointed out, the fact that it's an image rather than a file you can easily export the data from is also significant. The images also make it fairly tedious to do more than the intended purpose of the service, to wit, to evaluate a book for purchase.
Second, Amazon's experience is that if you give people more information, they buy more books. The original Look Inside let people get a sense of the book, but this is even closer to the experience of browsing in the bookstore. And think about it - in a bookstore, if someone wants, they can sit there and read the whole book, or do their research. And this is in fact a problem for bookstores that make themselves too comfortable. But on the whole, a better browsing environment is also a better selling environment. Amazon likes to point out that their interests are closely aligned with those of both authors and publishers -- if Search Inside means that they sell fewer books, you can be sure that Amazon will modify the program!
Where I do think there is a potential problem is that certain types of books are used for occasional reference, and for those types of book, the ability to search and look up specific types of information without buying the book can be useful in and of itself. As a result, I think you'll see far fewer reference publishers participating at this stage. After all, we know that online access to our material is valuable enough that people will pay for it.
Right now, I think it's important to wait and see how this pans out. It's a great innovation if it works out. I'd hate to see it get caught in the kind of overhyped copyright backlash that we've seen around online music.
Posted by: Tim O'Reilly at October 25, 2003 10:08 PMWhen I said "this is a bit of a problem for bookstores", I should have added "because books get shopworn, and then are more difficult to sell." But even with that problem, improved browsing in superstores has led to a great increase in book sales. Online, you don't have the shopworn issue.
In short, while there may be some issues for certain types of content, overall, this is an exciting step forward.
Posted by: Tim O'Reilly at October 25, 2003 10:14 PMThis is great, I was able to read a book on Amazon that I was originally going to purchase. Thanks amazon.
Posted by: John at October 29, 2003 03:48 AMJeff Bezo said in an interview (sorry, can't find it right now) that they have done several things to make it very hard and inconvenient to copy whole books. That's what the requiring a valid credit card thing is about. If I recall correctly it's something about only being able to browse X pages a day, X percent of any book and X pages a month.
- ask
Posted by: Ask Bjørn Hansen at November 7, 2003 01:02 AMhello,
send your book imidietly
thank you
I reached my 20% book limit, and on 1st March I could still not continue, so it's not 20% "PER MONTH"
Posted by: David at March 2, 2004 02:38 AM