6 ms·

I'm an aspiring cryptographer and this book is hands-down one of the best references for applied cryptography out there today. Don't let the date fool you; cryp

by ReidZB 13y ago

I'm an aspiring cryptographer and this book is hands-down one of the best references for applied cryptography out there today. Don't let the date fool you; cryptography is an area where maturity is well appreciated. It's also fantastic that the whole thing is available online, but believe me when I say that the text is well worth the price if you ever find yourself needing a good cryptography reference. Some other good references are Introduction to Modern Cryptography by Katz and Lindell, Foundations of Cryptography by Goldreich (both volumes), and The Codebreakers by Kahn (for history of cryptography). There are other good references, of course, but I find myself referring to the ones above the most. However, I am naturally biased towards texts that take a theory-based approach.

tptacek 13y ago
I guess you could argue that the advent of quantum computing is making everything old, like McEliece, new again. But I disagree with your premise. This book was written at a time when elliptic curves were more useful as a factoring method than as a cryptographic design tool. Modern concepts, algorithms, and constructions that a designer might not have at their disposal if they rely on Menezes: * Elliptic curve cryptography, which is supplanting RSA (and Rabin) and conventional DH, both of which are looking increasingly sketchy in 2013. * Secure RSA padding for encryption (OAEP) and signatures (PSS), or, really, any discussions of the problems that motivate those formats --- Menezes does point out that you don't want to use low-public-exponent RSA with repeated messages, but a modern take on RSA would say that you don't ever want to send repeated messages at all. * Counter mode, which is presented briefly as a small tweak to OFB mode with no description of its pitfalls; CTR is the second-most popular mode used today and increasingly the first choice of new systems. * Modern bulk encryption --- XEX, XTS, &c. * Encrypt-then-MAC (Menezes presents the opposite); in fact, modern crypto is at pains to ensure that encryption and integrity are co-specified and interoperate safely; a designer could be forgiven for reading Menezes and not including explicit integrity checks at all. * The authenticated encryption modes CCM, GCM, EAX, and OCB. * Modern CSPRNGs, including entropy collection and management and handling the cold-start problem. * Any discussion of side channel attacks (in fact, Menezes predates Vaudenay and Bleichenbachers block cipher and RSA [respectively] error side channels so doesn't even discuss padding oracles). It's not true that robust cryptography appreciates maturity. The ~twenty years since Menezes wrote HOAC haven't just been spent inventing whizz-bang new toys, but also in finding new ways to break the old toys. You're not always OK if you eschew the new stuff for the "mature" stuff; some of the mature tools are perilous to work with now. That's an annoying thing about working with crypto (and a fun thing about breaking it) --- designers have to know what parts of the literature they need to adopt, like safe RSA padding schemes, and what parts they need to hold off on, like homomorphic encryption. For what it's worth: I like Menezes a lot as a resource for practicing attacks. Menezes also does a much better job on theory than even _Practical Cryptography_ (which is a great book, and probably the only crypto book most people should own). It's a good book. Just be careful working from it. (I'm not trying to call you out; I just think this is a fun thing to talk about.)
- pbsd 13y ago
  > RSA (and Rabin) and conventional DH, both of which are looking increasingly sketchy in 2013. May I ask why you say this? (Not agreeing nor disagreeing, just curious.)
  - sillysaurus 13y ago
    Elliptic curve cryptography isn't vulnerable to quantum computing. EDIT: This is mistaken. I was thinking of lattice-based cryptography, which isn't currently vulnerable to quantum computing. In the video game Final Fantasy, there's a spell called "Doom" which places a countdown over your head. When it reaches 0, you die. RSA is Doomed in exactly that sense: it's just a matter of time before RSA offers zero security, whereas elliptic curve cryptography remains (for now) unbroken.
    - pbsd 13y ago
      It is [1]. Quantum computers, using Shor's algorithm, polynomially break any specialization of the abelian hidden subgroup problem; see [2] for a fairly complete list. Whatever reason to prefer elliptic curves over integer factorization or discrete log-based schemes must be classical. [1] http://arxiv.org/abs/quantph/0301141 http://arxiv.org/abs/quantph/0301141 [2] http://pqcrypto.org/quantum.html http://pqcrypto.org/quantum.html
      - sillysaurus 13y ago
        Oh. Then which (if any) algorithm is currently believed to be safe from quantum computing? I think I was thinking of lattice-based cryptography.
        tptacek 13y ago
        There's a bunch of them. There's no known quantum algorithm for quickly decoding binary linear codes, so McEliece is one. The Clostest Vector Problem in linear algebra is another trapdoor that may be QC-resistent. You didn't ask, but it's worth saying: block ciphers, stream ciphers and hash functions aren't thought to be fundamentally threatened by QC the way IFP and DLP number theoretic cryptosystems are.

by ReidZB 13y ago

tptacek 13y ago

I guess you could argue that the advent of quantum computing is making everything old, like McEliece, new again. But I disagree with your premise. This book was written at a time when elliptic curves were more useful as a factoring method than as a cryptographic design tool. Modern concepts, algorithms, and constructions that a designer might not have at their disposal if they rely on Menezes: * Elliptic curve cryptography, which is supplanting RSA (and Rabin) and conventional DH, both of which are looking increasingly sketchy in 2013. * Secure RSA padding for encryption (OAEP) and signatures (PSS), or, really, any discussions of the problems that motivate those formats --- Menezes does point out that you don't want to use low-public-exponent RSA with repeated messages, but a modern take on RSA would say that you don't ever want to send repeated messages at all. * Counter mode, which is presented briefly as a small tweak to OFB mode with no description of its pitfalls; CTR is the second-most popular mode used today and increasingly the first choice of new systems. * Modern bulk encryption --- XEX, XTS, &c. * Encrypt-then-MAC (Menezes presents the opposite); in fact, modern crypto is at pains to ensure that encryption and integrity are co-specified and interoperate safely; a designer could be forgiven for reading Menezes and not including explicit integrity checks at all. * The authenticated encryption modes CCM, GCM, EAX, and OCB. * Modern CSPRNGs, including entropy collection and management and handling the cold-start problem. * Any discussion of side channel attacks (in fact, Menezes predates Vaudenay and Bleichenbachers block cipher and RSA [respectively] error side channels so doesn't even discuss padding oracles). It's not true that robust cryptography appreciates maturity. The ~twenty years since Menezes wrote HOAC haven't just been spent inventing whizz-bang new toys, but also in finding new ways to break the old toys. You're not always OK if you eschew the new stuff for the "mature" stuff; some of the mature tools are perilous to work with now. That's an annoying thing about working with crypto (and a fun thing about breaking it) --- designers have to know what parts of the literature they need to adopt, like safe RSA padding schemes, and what parts they need to hold off on, like homomorphic encryption. For what it's worth: I like Menezes a lot as a resource for practicing attacks. Menezes also does a much better job on theory than even _Practical Cryptography_ (which is a great book, and probably the only crypto book most people should own). It's a good book. Just be careful working from it. (I'm not trying to call you out; I just think this is a fun thing to talk about.)

pbsd 13y ago
> RSA (and Rabin) and conventional DH, both of which are looking increasingly sketchy in 2013. May I ask why you say this? (Not agreeing nor disagreeing, just curious.)
- sillysaurus 13y ago
  Elliptic curve cryptography isn't vulnerable to quantum computing. EDIT: This is mistaken. I was thinking of lattice-based cryptography, which isn't currently vulnerable to quantum computing. In the video game Final Fantasy, there's a spell called "Doom" which places a countdown over your head. When it reaches 0, you die. RSA is Doomed in exactly that sense: it's just a matter of time before RSA offers zero security, whereas elliptic curve cryptography remains (for now) unbroken.
  - pbsd 13y ago
    It is [1]. Quantum computers, using Shor's algorithm, polynomially break any specialization of the abelian hidden subgroup problem; see [2] for a fairly complete list. Whatever reason to prefer elliptic curves over integer factorization or discrete log-based schemes must be classical. [1] http://arxiv.org/abs/quantph/0301141 http://arxiv.org/abs/quantph/0301141 [2] http://pqcrypto.org/quantum.html http://pqcrypto.org/quantum.html
    - sillysaurus 13y ago
      Oh. Then which (if any) algorithm is currently believed to be safe from quantum computing? I think I was thinking of lattice-based cryptography.
      - tptacek 13y ago
        There's a bunch of them. There's no known quantum algorithm for quickly decoding binary linear codes, so McEliece is one. The Clostest Vector Problem in linear algebra is another trapdoor that may be QC-resistent. You didn't ask, but it's worth saying: block ciphers, stream ciphers and hash functions aren't thought to be fundamentally threatened by QC the way IFP and DLP number theoretic cryptosystems are.