2,050 reputation
2829
bio website stackexchange.com/users/…
location
age
visits member for 3 years
seen 9 hours ago

Moderator♦ pro tempore on French Language, Computer Science and Software Recommendations. I'm also a unix amateur, and a developer with a computer science background and security leanings by trade.

Interested in embedded systems? Join the Embedded Programming and Design proposal!

Programmeurs francophones : participez à Stack Overflow en français.

Avatar picture slightly adapted from a photo by Luridiformis (Zonda Grattus).


Jun
14
reviewed Leave Open What are the consequences of not checking the server mac in a TLS connection?
Jun
14
awarded  Nice Answer
Jun
14
comment Strategy for random CTR initial counter values
@RichieFrame While the device can receive and verify signed messages, it needs to be able to emit encrypted messages without having to wait for the control server.
Jun
13
asked Strategy for random CTR initial counter values
Jun
13
comment RSA decrypting of a huge file by parts
Your question is not clear: are you designing a storage format for that file, or do you have the encrypted file already stored? If the latter, you need to tell us how that file was produced. As the answers here explain, what you're describing, with a huge file somehow encrypted with RSA, is highly atypical. RSA only lets you encrypt a few hundred bits at a time, and slowly, so it's normally used to encrypt a symmetric key which is used to encrypt the rest.
Jun
13
comment Given $n$ bits, how many “truly random” sequences/numbers can be constructed?
@NeilSlater Indeed, I inverted the meaning of pass/fail midway (pass = have the correlation property vs pass = look random). I've reworded that sentence.
Jun
13
revised Given $n$ bits, how many “truly random” sequences/numbers can be constructed?
Corrected sentence which inverted the meaning of pass/fail midway (thanks Neil Slater)
Jun
12
answered Given $n$ bits, how many “truly random” sequences/numbers can be constructed?
Jun
8
reviewed Reviewed ECDSA: How to retrieve a non-random k
Jun
8
revised ECDSA: How to retrieve a non-random k
added 91 characters in body; edited tags
Jun
8
reviewed Approve suggested edit on RSA encryption with private key and decryption with a public key
Jun
6
reviewed Approve suggested edit on modes-of-operation tag wiki excerpt
Jun
4
comment Padding to increase entropy and size of (otherwise) small JSON objects
@aross Asymmetric encryption is never used for ordinary data directly — it would be technically possible but extremely slow and libraries aren't set up for it. What you do is hybrid encryption: generate a random AES key, use an AES mode to encrypt (and authenticate) the data (your JSON string), and use public-key encryption on the AES key. You can keep using the AES key as long as the other party still has it. JWT only standardizes RSAES-PKCS1-V1_5 for asymmetric encryption, which is not perfect.
Jun
4
answered Padding to increase entropy and size of (otherwise) small JSON objects
Jun
1
reviewed Close How small are we talking about when defining the small public/private key exponent
Jun
1
reviewed Leave Open How Does Progressive Hashing Work?
May
31
reviewed Reject suggested edit on Security of the iterated Hill Cipher
May
29
reviewed Leave Open Why is elliptic curve cryptography not widely used, compared to RSA?
May
27
comment Does data authenticity always, implicitly, provide data integrity?
@otus These are different meanings of integrity (or, if you prefer, different settings which both have an integrity property).
May
27
comment Does data authenticity always, implicitly, provide data integrity?
@otus Yes, that's my point. In this sense, HMAC achieves authenticity (if the HMAC is correct, the file has to be mine) but not integrity.