Public Key Cryptography

[llasram]

Originally Posted by Taylor514ce

Keep in mind the goal of this system isn't to prevent piracy, a technological impossibility. It's a system to key a book to a person, which is analogous to a person walking into a store and picking a copy of a book off a shelf and buying it, with the additional goal of providing a reasonable degree of protection against casual copying.

So if I understand properly, the goal it to make operating within the DRM scheme convenient enough that the overwhelming majority of users would feel no need to remove the DRM even if it were easy to do so?

[Taylor514ce]

There is no DRM, in the sense of locking a book to a device. The file containing the book is encrypted, that's all. Only you can decrypt it, via your private key. Copy the file wherever you want, as many times as you want, and still only you can read it. Use any book reading program or device you want (they would all have to support public key encryption, and provide a way for you to enter your key - imagine this as a step in the software installation or device registration). Doesn't matter. The books are keyed to you.

If you're a skilled programmer with the ability to write an encryption/decryption program yourself, so that you can decrypt your books outside of a reader application, and then spread the plain text around, nothing in this system can stop you. Similarly, if you want to photocopy your Cybook screen as you turn each page... nothing will stop you. Nothing will stop dedicated pirates from pirating books.

Writing a public key encryption/decryption program isn't exactly trivial - so nothing in this system makes it "easy" to remove the DRM.

This system would even be stronger that today's typical DRM systems, because there would be no single universal "code cracker". A program to decrypt my books wouldn't work on your books.

[moz]

Originally Posted by Taylor514ce

Sorry, moz, I was just dense.

That's fine, you get it now, thanks to everyone else for explaining while I was away. And thanks for the apology (unnecessary, but appreciated).

Understand I'm talking about a theory only, and in this ideal theory, DRM would disappear, and all book publishers would agree, if not on a format, at least on an encryption infrastructure.

Search for the "open DRM" discussions to see why this can't work. It fails on both levels - in practice, as soon as one person cracks it everybody loses all DRM rather than the current plethora of systems requiring individual cracks. On a theoretical level, the more information people have about the DRM system the easier it is to crack. What keeps FOSS working is that there is limited profit to be made from breaking it, and a lot of people who want it fixed. With DRM, a lot of those people would dislike the whole idea and some would actively interfere.

Why would someone like me contribute to developing an "open DRM" system that stops me reading the books I've bought? Wouldn't I be more likely to sign up and build a tool that lets me do whatever I want with the DRMbooks I buy? Then, of course, I'd release those tools so everyone else could do the same with their DRMbooks.

[Taylor514ce]

Re-read the post right above yours. A universal "crack" to public key crypto just isn't feasible. Maybe for the NSA. They read everything anyway.

Nothing in a pk system stops you reading the books you've bought. In fact, in enables ONLY you to read the the books you've bought, and doesn't limit you to a single device.

Note that I'm not discussing an unproven system. It's already in use in secure email systems and secure web sites, and no one is complaining that they can only browse secure sites with particular programs or particular specific machines, nor has web commerce died because of a "universal crack" of HTTPS. It's probably the most attacked and proven system out there.

[llasram]

Originally Posted by Taylor514ce

Re-read the post right above yours. A universal "crack" to public key crypto just isn't feasible. Maybe for the NSA. They read everything anyway.

Well, that depends on what you mean by "crack." The system you're describing involves giving customers encrypted content and the key to decrypt it with. They have the ciphertext and the key. Ciphertext + key = plaintext. That's like the whole point of cryptography, right . But that means in this case there doesn't have to be a "break" in the algorithm.

If you look at the Mobipocket DRM tool it's actually pretty funny. The device PID is the key. Using that to produce decrypted content from DRMed files isn't a matter of finding a cryptographic weakness in the algorithms used, but simply a matter of figuring the algorithm. Then key + algorithm + DRMed content = decrypted content.

This is what moz was getting at about "open DRM." Cory Doctrow has some articles about this, but all DRM -- or "DRM-like"? "DRM-ish"? systems -- depend on giving customers both the encrypted content and the keys to decrypt it with. To prevent "unauthorized" decryption you then have to hide either the algorithm or the keys or both. Which isn't possible -- that information can only be obfuscated.

[moz]

Originally Posted by Taylor514ce

It's already in use in secure email systems and secure web sites,

Yes, but the whole point there is that anyone can see them. With DRM, you're trying to stop everyone except the one special user from seeing it, and you're also trying to prevent that user from saving the data unencrypted. Try this experiment: open an encrypted email and forward it unencrypted. If you can, you just "cracked" the open DRM system you're proposing.

[sanders]

I don't think this would work. The reason is that there's no incentive for the sharer to keep his private key secret (as there is in his WiFi and other instances). The book will just end up on the Darknet along with the required key.

[Taylor514ce]

The last three posts seem to be intentionally missing the point. I'll respond once more then drop the subject, as I think I've clearly explained PK cryptography and anyone interested in learning how it works can research it on their own.

@Ilasram: Publishers never give you the key. You never give them the key. There is a public encryption key and a private decryption key. That's the very heart of the system. You've misunderstood the system at it's fundamental level.

@moz: anyone can see what? No one can log into my webmail and read my emails except me, unless I give away my password. I'm not sure what would induce me to do that, but yes... if I choose to give away my password, people could read my email. But, here's the point, they couldn't read YOUR email with my password. The system isn't broken.

If you say "once you have your email you can copy it and forward it", yes, that's a "given" and I've already conceded that point at least half a dozen times already. Please re-read the thread.

@sanders: since all my books are encrypted with my key, publishers could easily track me down if I decided to post all my books in a torrent with the message "and here's my key". That would take a pretty dimwitted pirate. The path to piracy would be to first decrypt all your books and then to post the decrypted versions. And to repeat, again, I know it's not piracy-proof, no system can be. Please re-read the thread.

[DaleDe]

I am sorry but this all looks like DRM to me. Currently there is DRM on eReader files that have no need for locking to a device. The lock is to a credit card number and the purchasers name is in the file I believe but not used for unlocking (decryption). Also there are systems based on passport that have multiple device support from Adobe and Microsoft. Other systems have multiple device support as well and as such the number of devices could be changed it needed.

How is this different? I know technically how decryption is different. I am asking how it impacts less and offers and improved user experience over say eReader.

Dale

[Taylor514ce]

@DaleDe: it may not be different, as I haven't explored eReader, etc. Any system which keys the book to the user rather than to a device, and provides a unique transaction between buyer/seller (meaning, the book they sell me is demonstrably different from the "same" book they sell you), and that doesn't limit personal use of my book, would work.

I think the systems you describe are different, if they key to "multiple devices"... a fundamental difference is the system I propose publishers adopt doesn't key to a device, anymore than you are chained to a particular computer to login to your webmail.

[llasram]

Originally Posted by Taylor514ce

@Ilasram: Publishers never give you the key. You never give them the key. There is a public encryption key and a private decryption key. That's the very heart of the system. You've misunderstood the system at it's fundamental level.

Hmm. I think my use of the word "give" distracted you and/or you have some misapprehension about the nature of asymmetric cryptography. Asymmetric cryptography gives you: algorithm + pubkey + plaintext = ciphertext; and algorithm + privkey + ciphertext = plaintext. If you're only looking at the decryption side -- considering only the actions of the DRM-stripping pirate -- the fact that the decryption key is the private half of a public/private key pair is of no consequence: you still have the inevitable fact that algorithm + key + ciphertext = plaintext for even authorized programs to be able to derive the plaintext from the ciphertext. If your system allows one to read a book on any device just by providing your public key and the protected book, it would allow a DRM-stripper to produce an unecrypted copy of the book with just by providing it the public key and the protected book -- exactly as things stand today.

I do agree(?) that a less restrictive system would provide less incentive for piracy, but a system based on asymmetric cryptography would make it no more difficult to produce a systemic break for the sufficiently motivated.

[Taylor514ce]

You were ok up to this point:

Originally Posted by llasram

If your system allows one to read a book on any device just by providing your public key and the protected book.

Substitute "private" for "public" and you've got it. And of course, no pirate is going to have my private key, unless I am the pirate, and that brings us round full circle to the fact that nothing will stop a determined pirate, and that at some point any system has to produce plaintext, or else we couldn't read books.

[llasram]

Originally Posted by Taylor514ce

Substitute "public" for "private" and you've got it. And of course, no pirate is going to have my private key, unless I am the pirate, and that brings us round full circle to the fact that nothing will stop a determined pirate, and that at some point any system has to produce plaintext, or else we couldn't read books.

Oops. That is what I meant :-).

[moz]

Originally Posted by Taylor514ce

If you say "once you have your email you can copy it and forward it", yes, that's a "given" and I've already conceded that point at least half a dozen times already.

So your scheme is not designed to stop people giving away "protected" content after removing the "protection"? I was misled by your posting this in response to the piracy/DRM threads. I thought you were suggesting this as a way to reduce/prevent piracy, rather than just as a way to transmit books to users.

OK, in that case PKI will do everything you want.

[tompe]

Originally Posted by moz

So your scheme is not designed to stop people giving away "protected" content after removing the "protection"? I was misled by your posting this in response to the piracy/DRM threads. I thought you were suggesting this as a way to reduce/prevent piracy, rather than just as a way to transmit books to users.

What I do not get is why you need this for the transmission. The first thing i would do when I got the book was to decrypt it to protect me from negative consequences of loosing the private key.