Inherent design features


At the moment, there is a big uproar because Amazon removed some books from users' Kindle devices (see also [1] [2] [3]).

I quote:

This means that all the reassuring talks by Amazon that e-books are just like books, but better is a load of absolute nonsense. You're not allowed to resell them, you're not allowed to give them away, and apparently, you don't even own them, as Amazon can delete them from your Kindle at any given moment. (Thom Holwerda,

But yes, of course. Excuse me for being blunt, but only highly naive technophiles would ever believe that anything other than the above is the case.

Okay, and innocent, trusting customers. But if somebody like David Pogue seems to be genuinely surprised by this development, it must be a case of naive technophilia.

An e-book is a virtual book over which the publisher retains full control at all times. When you licence an e-book, you do not buy a physical copy that remains in your possession, you are merely granted the temporary right to access data on a server that is beyond your sphere of influence. When the conditions that regulate the access to the data change, the publisher can easily disconnect you from the book.

Actually, the same principle also applies to printed books, it simply cannot be brought about as easily. Only the copies in the bookstores can be recalled by the publisher; the copies already sold to customers may now be illegal, but tracking them down and removing them is impossible. This is one of the most important differences between access to a data service and possessing a physical object.

To illustrate the point that the Amazon case is anything other than an exception, here are some things that happened at The Library in conjunctions with e-books and e-journals:

  • Due to an oversight, a bill for an e-book servive was paid one day after the due date. As a result, access to about 1000 titles was denied for the entire calendar month.
  • The Library subscribed to an e-journal for a few years, then cancelled the subscription. The publisher removed access to the entire journal; the Library could no longer access even the volumes that it had paid for.
  • An e-book publisher went out of business; the Library lost access to hundreds of titles at once.
  • Sometimes, technical/connection problems occur that make hundreds of titles (they are usually bought in packages) temporary unavailable.

Libraries have been dealing with problems like these for quite a while, and are as a result now taking extremely great care to check what the licence agreements and contracts say to avoid situations like these. After all, denied access can have dramatic repercussions for their users (10,000 journals gone over night -- that's about as dramatic as a library burning to the ground, and it only requires that somebody flicks a switch somewhere).

So what is the surprise here? The fact that Amazon did something that was perfectly legal, took advantage of its licence agreements and cut access to a few e-book titles? Or the fact that users believed that this would never happen?

Complaining that Amazon cut the access to some e-books is like complaining that your Kindle doesn't work when there is no electric power. It's a condition that is inherent in the medium, just like "real" books are not easily searchable, bulky and take up a lot of space.


So now there a bunch of "naive technophiles" who are naive no longer.

A couple of years ago I "bought" a bunch of songs from Rhapsody and burned some CDs; the songs showed as "purchased" in my Rhapsody library - but now (I don't use Rhapsody very often nor pay attention to it), it seems I no longer own these songs. I still have Rhapsody, (most of) the songs are still in my library, but Rhapsody confirmed that they changed the rules and I no longer own the songs. I don't care enough to make a fuss about this, but it just seems sort of silly to me.

You say "Amazon did something that was perfectly legal, took advantage of its licence agreements". Can you show exactly which of these agreements allows it to do that? A number of articles that I have read suggest that Amazon didn't give themselves the rights to remove works, though I suppose it could at a pinch be argued under the 'we can change these terms and conditions whenever we feel like it'. However, until and unless their action is challenged in court I'm not sure that I'd be comfortable saying that it was perfectly legal.

"An e-book is a virtual book over which the publisher retains full control at all times."

No, it is not.

An e-book is a digital file. Period.

Some e-books, such as those purchased for the Kindle from the Kindle store, are purchased on a license basis and, given sufficient technical ability in the reading device, could have those licenses revoked.

Not all e-books, formats, or devices support this notion. Public domain e-books ( are not under the "full control" of the publisher, for example.

Blame the publisher or vendor, don't blame the format.

"When you licence an e-book, you do not buy a physical copy that remains in your possession, you are merely granted the temporary right to access data on a server that is beyond your sphere of influence."

So when people purchase and download the PDF and Kindle versions of my e-books from my site...that's not happening? When I downloaded Prof. Boyle's book ( and put it on my Kindle...that didn't happen? It sure seems to be showing up in my book list. Prof. Boyle's site could vanish tomorrow, and my copy of the book will still be there, as would the PDF original I used for creating the Kindle edition.

You are painting the whole world of e-books with a Kindle Store brush. A few minutes of research would tell you this is nonsense. Project Gutenberg, Wikibooks, Baen's Webscription and the Baen Free Library, Prof. Lessig's CC-licensed books, and my own Warescription, are just a few examples of e-books that lack the provisions you imply are universal for e-books.

Don't get me wrong: I don't like the Kindle because of the controls you describe. I only use the Kindle because I publish books in that format...but I only publish those books outside of the Kindle Store, to avoid the sorts of issues you describe (not to mention Amazon's ridiculous cut of the proceeds). However, the problems you describe are endemic to the Kindle, not to e-books as a whole.

Again, blame the publisher or vendor, don't blame the format.

As far as I understand, there are two models of e-books consumption. One which is more prevalent for Journals and probably with libraries is subscription based - where you are just accessing the data on the provider's server.

But the more well known mode for consuming trade ebooks is downloading it to one's computer/e-book reader and in that case, there is actually a copy of the book on your device. Amazon did not cut back the access to some data on their server - they went ahead and deleted these files from the kindles of all those customer which is quite a different thing.

Is there something I am missing?


This discussion really only applies to ebooks that incorporate DRM (Digital Rights Management). There is no way for a non-DRM'd ebook to be "recalled" by anyone, publisher, author or distributor. It is also possible for the buyer of a non-DRM-encumbered ebook to pass it along or use it on more than one device, but of course that is the "problem" according to some copyright owners. A movement is growing, though, to eschew DRM. It's perceived as risky, but some publishers, like Baen Books (, have actually seen increases in sales of their print and "premium" e-versions when they release ebook editions for free.

Actually, the books had been posted to Kindle by someone who did not have the copyright, which is akin to receiving stolen property. The Kindle users were reimbursed, which is more than often happens when actual stolen books are recovered by police. And yet, because of the hubbub, Amazon promised next time you Kindle users would get to keep the books. There's not as much to this story as folks are making of it.

And furthermore, it occurs to me to ask, now that I read that you are not only a librarian, writer and university lecturer, but have published Mr Singh Has Disappeared, available from what would happen if someone other than you sold your book to Amazon's Kindle w.o. your permission and proceeded to take a share of the profits.

To their major credit, the library community were among the very first to poke holes in DRM, literally from its birth. As a participant in early DRM conferences (ca. 1995) I can tell you that the most vocal critics were librarians, who anticipated instantly and with great precision all of the problems that would occur when DRM hit works under their domain.

A solutions that was proposed more than a decade ago was based on an eContract model whereby an actionable license (between the distributor or publisher and the user) would be held by a third party; imagine iTunes or Audible outsourcing their license DBs to a third party who offered them as services with an API. Among its many benefits, this approach would facilitate cross-DRM compatibility; at the time there were at least three promising eBook technologies including Rocket, Adobe and Glassbook (which extended the Acrobat API) and the idea was to have license-level compatibility between these systems. The same would be true with the various multimedia DRM technologies, including and especially Intertrust and Windows Media DRM.

Librarians immediately saw the "escrow" potential for such a service; if a DRM vendor went belly-up (almost all have!) then POTENTIALLY the still-valid license could be applied to works (re-)packaged using another DRM technology. Other applications included porting content from one platform to another for convenience, etc.

Note that having such a cross-platform license service wouldn't necessary lead to interoperability between platforms or across time. Publishers and distributors (in particular) could not see the reasoning behind having one license that could open a work in different formats (think: an iPhone-friendly version vs a desktop version, using different DRM systems). Their argument typically went like this (this is almost a direct quote from the CEO of a DRM company): Buying the hardback version of a title doesn't give the customer instant access to the paperback version; why should the 'purchase' of a title using one DRM give the customer instant access to the title in a different DRM?

Why indeed! This is fundamentally not a question of technology, but rather of publishers, distributors and DRM vendors persistently not serving the interests of the end user. This is not new; we happen to be talking about this now due to Amazon's relative success with the Kindle, but these issues have been around since before DRM was even called "DRM!"

Leave a comment