dr jimbob

12,009 reputation

1659

bio	website
	location	Brooklyn, NY
	age	31
visits	member for	2 years
	seen	8 mins ago
stats	profile views	295

Good Morning how are you, I'm dr jimbob
I'm interested in things.
I'm not a real dr,
But I am a real jim bob.

Have a PhD in Experimental High-Energy Physics, but left academia in mid-2010 to program professionally.

Mostly program/script in python, django, and jquery these days doing mostly web apps.

Also have experience programming in C, C++, java, haskell, php, and (bash) shell more in the past.

Linux as primary OS since 1999, ubuntu user since 2005 (Hoary).

	12,009 reputation	bio	website		visits	member for	2 years
1659 badges		location	Brooklyn, NY		seen	8 mins ago

summary answers questions tags badges favorites bounties reputation activity

891 Actions

suggestions reviews revisions comments badges posts accepts all

4m	answered	RSA key length vs. Shor's algorithm
11h	revised	Why does PGP use symmetric encryption and RSA? added 7020 characters in body
11h	revised	Why does PGP use symmetric encryption and RSA? added 7020 characters in body
15h	comment	How likely/possible is it that the NSA have broken common encryption techniques such as SSL/TLS? Asymmetric is the bigger problem, as Shor's algorithm would break PKE schemes based on integer factorization/discrete logs (like RSA and El Gamal even with elliptic curves) in O(b^3) instead of sub-exponential. So QC capable of factoring say 1024-bit numbers means a scale-up of only ~4 in number of qubits and ~64 in time would break 4096-bit RSA. (The classical scale-up is 10^11 from 1024 to 4096 bit RSA). Granted, there are proposed PKE schemes that are resistant to Shor's algorithm (none in wide use though).
15h	comment	How likely/possible is it that the NSA have broken common encryption techniques such as SSL/TLS? @AJ - Agree, it is relevant to bring up, who knows when large-scale QC will be feasible. It still seems a far way off, though it would not surprise me if the NSA etc has a headstart (though rumors would likely leak out). I try and double the lengths of my symmetric keys to be secure against quantum algorithms (e.g., 256-bit from 128-bit) as its easy (Grover's algorithm speeds up brute force attacks from O(N) to O(sqrt(N))).
1d	comment	Why does PGP use symmetric encryption and RSA? (If you use don't use random padding in RSA, any attacker with your public key can recreate the stream and decrypt. If you do random-pad there's no way for someone with the private key to recreate the same stream.) I'm not trying to prove you wrong. The more I think of it, I was wrong with my first comment. ECB should still be avoided as eavesdroppers could reorder/drop parts of the longer message if it aligned properly. On further thought, you could probably do something like CBC using RSA (if you don't care about being slow) and wasting space if you properly pad (e.g., OAEP) each block.
1d	comment	Why does PGP use symmetric encryption and RSA? Let's try and stay civil and my repeated part was me agreeing with that part of the argument. I never downvoted you, but disagreed with bringing up chaining modes (and linking to block cipher modes of operation). Disagreement of ideas is good; how else do we learn? My point was that its weird to discuss chaining modes for RSA. E.g., applying CTR mode to RSA won't work as normally you XOR the N-block plaintext with something like `E(k,0)++E(k,1)++...E(k,N-1)` (++ being concatenation). This simply won't work in a secure manner with asymmetric keys (similarly OFB and CFB make no sense).
1d	comment	Why does PGP use symmetric encryption and RSA? Or see for example Thomas Pornin's SO answer: "Chaining mode such as ECB makes no sense for RSA, unless you are doing it wrong."
1d	comment	Why does PGP use symmetric encryption and RSA? Sophisticated chaining modes only need to be introduced for block ciphers. A block cipher means it is both deterministic and symmetric. RSA is not deterministic (in practice as randomized padding needs to be used) or symmetric (all the block cipher diagrams assume the same encryption/decryption key). All of the analysis of block ciphers is meaningless when attempted on RSA.
1d	comment	Why does PGP use symmetric encryption and RSA? RSA is not a block cipher. There's no need to do any sort of sophisticated block cipher chaining mode with RSA. Granted if you had a message too large for RSA and wanted to avoid hybrid encryption for some bizarre reason, you could in principle do something similar to ECB (which is weak for any deterministic block cipher) by splitting up the plaintext, but it would be slow.
1d	answered	Why does PGP use symmetric encryption and RSA?
1d	comment	How likely/possible is it that the NSA have broken common encryption techniques such as SSL/TLS? Agree with @CodesInChaos. DWave can only do quantum annealing algorithms (not Shor or Grover) and only for one very specific type of problem (the DWave Problem) and its qubits are very noisy with a high-error rate (ok for Monte Carlo simulations). The evidence seems to be that it does have quantum effects, but there's no suggestion whatsoever that its in any faster than optimized classical annealing algorithms. Quantum computing is very interesting work (Umesh Varizani has a good MOOC at edx.org on quantum computing).
2d	comment	Are very short messages secure using Public Key Encryption? @CodesInChaos - Agreed RSA with a good random padding scheme (RSA-OAEP - resistant to Adaptive Chosen Ciphertext Attacks) can encrypt small messages directly. Granted, in practice when you use RSA (e.g., in TLS ) you will use encrypt a symmetric key as asymmetric encryption is slow and expensive and can only encrypt short messages on the order of the length of the RSA modulus (n).
2d	revised	Are very short messages secure using Public Key Encryption? added 2419 characters in body
Jun 15	awarded	hash
Jun 15	revised	Are very short messages secure using Public Key Encryption? deleted 2 characters in body
Jun 15	revised	Are very short messages secure using Public Key Encryption? added 320 characters in body
Jun 15	answered	Are very short messages secure using Public Key Encryption?
Jun 14	answered	Why are MD5 collisions dangerous?
Jun 11	comment	Is it legal to find bugs on a website and report them to the website's owner? +1 for the advice to cover your ass (don't search for security holes unless you have permission from the owner to audit). Laws are vague and the threat of legal action will complicate your life. But have "many people gotten jail time"? The wikipedia list of computer criminals is small (though likely incomplete), and none seem to fit the pattern of "I found a security hole and didn't use it". Most cases the sentence for virus creators/bot net owners/website defacers are relatively light (e.g., community service).