My right to decipher a published work

Suppose I buy a book. I'm allowed to read the book, publish commentary on or a review of the book (and own the copyright on that commentary or review), understand ideas expounded in the book and exploit commercially any that aren't covered by patents. I can even sell my copy of the book on to someone else. The book's copyright owner may grant me greater liberties than the law guarantees, for instance by inclusion of a declaration to that effect in the text; but cannot deprive me of my rights, nor bind me to not read the text unless I agree to abide by some constraints stated in it (unless, I suppose, I was advised of the constraints before I purchased the book).

Without express permission from the book's copyright owner, I mayn't make copies of the book - or, at least, mayn't distribute (e.g. sell) such copies; jurisdictions may vary when the copies are for my own personal use. Even if we are not allowed to make such copies, my blind neighbour has every right to employ a machine (simple camera, optical character recognition software, text-to-speech synthesis software) to read books (at least when he's the only one who can hear what's read), and I am allowed to make such a machine and sell it to a blind (or illiterate) neighbour; albeit the talking camera may be required to keep no permanent record of what it reads. Even sighted purchasers of books might be willing to buy such a talking camera if it were hooked into software which translates books into their native language.

At least in some jurisdictions, a lawful purchaser has `fair use' rights to quote modest portions of the book (acknowledging its copyright), at least for the purpose of publishing a review or commentary, for which the purchaser is certainly allowed to read modest portions of the data and reproduce it in a document (possibly being prepared on a separate computer) discussing the book, whether the reading is done via a talking camera, eyes or - when the book is machine-readable data - third-party software.

When the book becomes a document published electronically, it comprises a body of data and is published along with software by which one may access it. The software comes with a license which requires me to abstain from using it unless I'm prepared to be bound by various conditions; the jurisdiction in which I purchased the software, or that in which I `accept' the conditions of the license (e.g. while installing the software) may or may not allow that those conditions are enforceable; but if I chose to abstain from using the software, the license is simply irrelevant - it can only concern itself with my right to use the software. I still have the publishing media, properly purchased within the constraints of copyright law, and the right to read it: I should also have the right to make a device which will read it to me.

The lawful purchaser of an electronic book may be horrified to discover that the book's accompanying software doesn't include a version which runs on the purchaser's computer (just as, on buying a book, I might be horrified to come home and find that I've purchased the Greek edition of a book, and my talking camera quotes Casca at me when I ask it to read the book); or the programs provided may be so unreliable or tiresome to use that the user would rather read the book with other software; or, though perfectly good in and of themselves, the programs provided don't interact well with the other tools available on my computer - e.g. if they don't let me cut-and-paste short extracts (for fair use) out to a review I'm preparing in another window. The lawful purchaser must have the right to read the data off the media carrying the book; albeit this right has turned out to be less of a blessing than they anticipated when they made the purchase. Just as I can make a talking camera to read a book to a blind purchaser, or translate it for a purchaser in whose native tongue it has yet to be published, I can write software to read the book off its disk - provided I can make sense of the data I find on the disk.

If, indeed, a book publisher were to sell me a book (in Mongolian, say) and a talking camera which translates Mongolian to English, I should have every right to buy a talking camera from my neighbour also; and if it translates Mongolian into English better, or into some other language in which I'd sooner hear it, I have every right to use a competitor's talking camera in place of the one the publisher bundled with the book. In place of Mongolian, the book publisher might use some other way of encoding the data in a book, or replace the book and camera with a CD and a CD-ROM drive; I should still be able to treat their recording medium with the same curiosity as has lead me to notice how books are bound; and to make devices by which I, or other lawful purchasers, may access the data and make such lawful use of it as we wish.

If the electronic book is sold, with the permission of its lawful publishers, in a jurisdiction which does not allow as enforcible clauses, in mass-market software licenses, which restrict the user's liberty to study and analyse the software, a purchaser in that jurisdiction may lawfully accept the license to the software and then, by studying it, discover how the data comprising the book is stored on the recording media of the electronic book.

Even in jurisdictions in which such clauses are enforcable, a legitimate purchaser may study the data on the media (though they won't get the program's ways of accessing the data as clues to what the bytes mean) without using (or accepting the license to) the software on it; they may, if determined enough or provided with enough clues from elsewhere (e.g. friends explaining what they've learned, sharing the fruits of their own lawful investigations), discover (possibly ad hoc) ways of accessing the published data. Either way, once someone has lawfully discovered how to read the data off the recording media, they may lawfully publish (effectively as commentary on the published work) an account of the data format - how to read the book.

Now, the data format these days may be encrypted - which some jurisdictions might regard as an unlawful restraint on competition, since it attempts to prevent competitors from offering, to legitimate purchasers of the electronic book, a free market in reader software. Publishers have pressured at least some jurisdictions into forbidding folk from publishing commentary on the technical measures they try to put into electronic publications, e.g. such encryption. They claim this is so as to prevent copying of the data: even though bulk byte-for-byte copying of the entire medium on which the book is stored always bypasses it. The actual effect of such legislation is to prevent software authors from writing programs which provide lawful purchasers of such publications with a free market in software to use in accessing such publications. The users of such software would be bound, like the owners of books, to abstain from distributing copies of the work: albeit the software system into which their reader for it is integrated may include programs which would make copying easy.

Consider the case of the book, containing encoded data, and a talking camera which can read it to me in English. Doing so may void any guarantees, but if I sell the package to a francophone, he has every right to open up the box and insert an English to French translator unit into the machinery within, if he can. However, for the parallel with software under a license accompanying the data that software reads, we need to suppose that I only get to hire the talking camera which comes with the book, though I buy books from those who hire it to me. I then don't own the talking camera, so I can't sell it to the francophone and he can't take it apart and modify it.

All the same, I own the book. A book's lawful purchaser has every right to study the book: and, if I can break the code, I'll be able to use my favourite talking camera to read the book - albeit I may need to write some software for the talking camera to use with this book, or use software someone else has written (subject to the permission of that software's author, naturally) - even if I hired that camera from someone else (provided they let me re-program it in the ways indicated). If the author of software to decypher such books choses to distribute that software, for use on various talking cameras, the only basis on which the book's publishers can object to them doing so is in so far as that software includes portions copied from the publisher's talking camera, or from the book itself. They might, thus, prevent me from disseminating the cryptographic keys used to encode the data; but would be unable to prevent me from publishing a program which knows how to find the keys on the lawful purchaser's copy of the book and, without keeping a copy, use these to access the book.

Such software makes it possible to access the decrypted data; which it is then easy (though illegal) to disseminate, of course: but only as easy as it is to disseminate copies of the encrypted data accompanied by the original publisher's software, which will enable those obtaining illegal copies to read them even if they are encoded. Decyphering software doesn't make it any easier to disseminate copies. The illicit distributor who has decyphered the data before copying will, all the same, need to

directly parallelling the un-decyphering distributor's need to distribute the publisher's reader (distribute the reader), unless the publisher has made this reader available separate from the book (plenty of users have a reader for the format). Thus the only change brought about by decyphering is that someone else's software can be used to read the book: the illicit distributor's customers then benefit from a free market in reader software, just as the lawful purchaser of the book gains this benefit from decyphering software.

So-called `copy-protection' systems do not in any way prevent copying, they are merely proprietary file formats devised with an overt intent to prevent competing software from providing lawful users with convenient access to the data. They, in fact, restrict access without preventing copying; they inhibit the free market in reader software. The law should rather outlaw this than protect it.

Printing a book in mirror writing doesn't make it harder to copy: it just makes it harder to read. Allowing the publisher of such a book to control the mirror trade, to `protect' the copyright of their book, would be absurd.

Written by Eddy.
$Id: decode.html,v 1.2 2001/08/28 11:36:33 eddy Exp $