CodesInChaos

less info
meta site favicon meta user | stack exchange network profile network profile
5,964 reputation
1927
bio website github.com/CodesInChaos
location Munich, Germany
age
visits member for 1 year, 10 months
seen 1 hour ago
stats profile views 109

1,106 Actions

suggestions reviews revisions comments badges posts accepts all
Aug
30
 comment Proof of work for standard computers
@DavidSchwartz The please post an answer that describes an algorithm. I for one see no way to make collisions of any kind work. My impression is that a computer with 1GB memory and a billion cores would solve your problem a billion times as fast as a computer with a single core, since there are n^2 possible pairings he can pick from to compare.
Aug
30
 comment Proof of work for standard computers
@DavidSchwartz You need to make it work well on standard computers, and hard on specialized computers. And at least for hamming-distance I don't see what kind of algorithm you envision.
Aug
30
 comment Proof of work for standard computers
For hamming distance, I don't know what data structure the worker should use to efficiently check if a near-collision occurred. If you think hamming distance works well, please post an answer describing how it would work in practice @DavidSchwartz.
Aug
30
 comment Proof of work for standard computers
For prefix collisions it's properties are much worse than a prefix pre-image with the same hash function(scrypt or plain hash). An attacker who as large shared memory and many small parallel If you want to use high memory requirement to give an advantage to a CPU, the worker must not be able to share that memory between cores. That's because a PC has a lot of memory per core, but the total amount of memory isn't larger than that of the attacker.
Aug
30
 revised Proof of work for standard computers
added 9 characters in body
Aug
30
 comment Which blind signature schemes exist, and how do they compare?
The problem is finding a scheme that's actually secure/production ready. If you're referring to A Novel Untraceable Blind Signature Based on Elliptic Curve..., it gives me a bad feeling. It seems like there is a way for the requester to obtain two valid signatures in one operation. And it's generally very sloppy.
Aug
29
 reviewed Approve suggested edit on ntru tag wiki
Aug
29
 reviewed Approve suggested edit on proof-of-work tag wiki excerpt
Aug
29
 reviewed Approve suggested edit on ntru tag wiki excerpt
Aug
29
 reviewed Approve suggested edit on proof-of-work tag wiki
Aug
29
 comment Random Sequence Generator function
Please don't cross post. Nobody suggested you should. Someone suggested a migration, and even that was dubious.
Aug
29
 comment Is this a good way to encrypt a file?
Your salt encryption step doesn't make much sense. Why not just include the plain salt?
Aug
29
 comment Why is ciphertext from low entropy plaintext not compressible?
The problem is that the actual entropy of the ciphertext isn't higher than the entropy of plaintext and key. It's just difficult to exploit this redundancy for compression.
Aug
27
 comment Secure Hash Function based on AES
You can build a MAC with a similar technique (CBC-MAC) but building a good hash function is harder.
Aug
27
 awarded  Enlightened
Aug
27
 awarded  Nice Answer
Aug
27
 comment Storage of Private Keys
@JordanArseno For spending the coins you use a smaller hot wallet. There is no reason to have users spend their own coins.
Aug
27
 comment Is it OK to use a data-encryption key for key wrapping, too?
@StoverFlo I assumed the anonymous edit was you not being logged in. If it wasn't feel free to revert.
Aug
27
 reviewed Approve suggested edit on Is it OK to use a data-encryption key for key wrapping, too?
Aug
27
 revised malleability wiki excerpt
deleted 1 characters in body
1 … 32 33 34 35 36 … 56