If I'm reading the documentation for crypto_stream_xor and crypto_secretbox correctly, both functions use XSalsa20. The former does not have any authentication whereas the latter does. Pursuant to this I would expect the first few bytes of crypto_secretbox to be the same as the first few bytes of...
The question is really about the algorithm implemented by crypto_secretbox and how it is related to crypto_stream_xor; that there was PHP code involved is a bit of a red herring.
@MaartenBodewes I make systems to detect munitions. I have a team currently detecting Nazi bombs from Holland so they can be later removed. I need these stupid cables so we can unplug and plug in sensors underwater, which can be everywhere in Holland.
After getting my Ph.D. paid for by kill vehicles and the NSA, I am doing humanitarian de-mining for net positive karma. :)
@Maeher It's about the relation of a widely used stream cipher, XSalsa20, and a widely used authenticated cipher, crypto_secretbox_xsalsa20poly1305, which are both implemented in many libraries in many languages and used in many protocols. The phrasing in PHP is not a substantial part of the legitimate question about how the algorithms are related.
It's not promising. It gets so much unwarranted attention and it is so bad that we joke about it. But we really wish it wasn't that way. It's a viral idea and it keeps getting reinvented. Academics (usually people who study something in a different field) that promote it likely haven't put significant effort into studying modern cryptography. To use an analogy: These are not chemists, they're members of a school alchemy club. And they don't even realize it because they're not humble enough to consider that there is a huge legitimate body of collective knowledge to base ideas off of. — Future Security1 min ago
Comment on "chaos"-based cryptography. Is it a fair statement?
@Maeher NaCl is widely used in many applications and protocols. Maybe you don't like it but it's a serious cryptography design and widely deployed, and it is a good choice for many applications because it is widely available in many languages without disasters like OpenSSL's API and ABI. Essentially the same algorithm—with one difference: 256 bits of the first 512-bit block are skipped—is used in TLS today, ChaCha/Poly1305 of RFC 7539.
@SqueamishOssifrage I have no strong feeling whatsoever about it. But everything you say seems to point to the fact that it's an implementation question.
Consider this question:
https://stackoverflow.com/questions/56186844/generating-hchacha20-internal-state-with-existant-chacha20-implementation
How is that question off topic when reusing blowfish algorithm to implement bcrypt isn't? Does the fact that I include code samples mean it's automatica...
According to @Maeher, we are not allowed to illustrate questions on crypto.stackexchange.com with executable code—such questions are permissible only on stackoverflow.com:
https://chat.stackexchange.com/transcript/message/50330566#50330566
Squeamish Ossifrage: Are we forbidden from illustrat...
@Maeher It is sometimes the clearest way to phrase something. I've seen plenty of papers with what were once straightforward algorithms transcribed into math formulas that are completely impenetrable piles of symbolgoo. I've also seen plenty of math formulas that cut to the heart of a pile of spaghetti code. What strikes me as bizarre is a proscription on the particular form of expression for a question.
According to @Maeher, we are not allowed to illustrate questions on crypto.stackexchange.com with executable code—such questions are permissible only on stackoverflow.com:
https://chat.stackexchange.com/transcript/message/50330566#50330566
Squeamish Ossifrage: Are we forbidden from illustrat...
@SqueamishOssifrage I don't see how the overhead of having executable code has ever added anything useful to an algorithmic description. But you do you.
It's better if the description of an algorithm is unintelligible to a machine, and only interpretable by human foibles of typesetting conventions and guessing inconsistent meanings of symbols across editorial styles?
According to @Maeher, we are not allowed to illustrate questions on crypto.stackexchange.com with executable code—such questions are permissible only on stackoverflow.com:
https://chat.stackexchange.com/transcript/message/50330566#50330566
Squeamish Ossifrage: Are we forbidden from illustrat...
