1,820 reputation
1517
bio website emboss.github.com
location Germany
age
visits member for 3 years, 5 months
seen Jul 15 at 0:11

Developer of PKI solutions, web applications and security consultant. Committer to CRuby (MRI) and maintainer of the Ruby OpenSSL extension. Author of krypt. @_emboss_ on Twitter.


Jul
7
awarded  Yearling
May
15
awarded  Great Answer
Jan
1
awarded  Popular Question
Sep
25
awarded  Popular Question
Jul
7
awarded  Yearling
Mar
20
awarded  Guru
Jan
13
awarded  Nice Question
Nov
10
awarded  Good Answer
Jul
7
awarded  Yearling
Jun
11
comment Which cipher suites with AES cipher provide forward secrecy?
@HubertKario OK, now I get it! Sorry for the confusion :)
Jun
10
comment Which cipher suites with AES cipher provide forward secrecy?
@HubertKario You've got to be kidding me? security.stackexchange.com/questions/14731/what-is-ecdhe-rsa ?
Jun
5
comment Using digital signatures to identify users
That entirely depends on what you want the entity impersonating the client certificate to represent. On the server side, the certificate is tightly bound to the particular server it runs on, due to hostname validation and all. But the situation is more relaxed on the client. There are basically no real naming restrictions. So if you want multiple client devices to represent the same entity, that's entirely possible - just let them share the same key/certificate. If you need to able to distinguish each device, then you would issue a distinct certificate to each, but it's not mandatory.
Jun
5
comment Using digital signatures to identify users
@James: You would use mutually authenticated TLS, which still requires certificates/keys for both the client and server, as in your originally planned scheme. But with that provided, it will do all the (a)symmetric crypto for you and you can be sure that it's secure.
Jun
5
answered Using digital signatures to identify users
Feb
27
awarded  Nice Answer
Dec
4
awarded  Benefactor
Nov
27
awarded  Promoter
Nov
27
revised Timing attacks on password hashes
fix typo
Nov
27
comment Timing attacks on password hashes
I thought a bit about your proof, and I think it's even possible to leave out the oracle at all? Assume the existence of A, then show we can conduct a preimage attack in less than exponential time in contradiction to the assumption of H being collision-resistant? I'm going to raise a bounty on this one, it's really important to me, so I would love to hear other impressions. But it's not that I wouldn't like your answer, I like it a lot! Just hoping that maybe someone can provide results for existing functions. Thanks again!
Nov
26
comment Timing attacks on password hashes
@PaĆ­loEbermann: Thanks a lot!