Re: Algorithm suggestions
- From: daw@xxxxxxxxxxxxxxxxxxxxxxxx (David Wagner)
- Date: Wed, 27 Sep 2006 20:25:46 +0000 (UTC)
Geoffrey Summerhayes wrote:
I expect the number of packets passing between any device
and the server to be around 100 per day at peak usage, so
in theory at least, it could require some time to gather enough
data to reverse engineer an encryption scheme.
It's worth reading up on Kerckoff's law (I am sure I misspelled
his name). If reverse engineering your crypto suffices to break
the security of your system, then you're probably in trouble. We
normally try to build systems that remain secure even if the adversary
knows how they work; modern cryptographic schemes are good enough
that you don't need to keep the algorithms secret.
I expect that every time someone invents a better lock, people
start working on better lockpicks.
Things are different in cryptography.
I think you probably should hire a cryptographic consultant, if
security matters. If you try to roll your own, the result may be
insecure. It's probably too hard to teach everything you would
need to do a good job, over a newsgroup. My apologies.
.
- Follow-Ups:
- Re: Algorithm suggestions
- From: Geoffrey Summerhayes
- Re: Algorithm suggestions
- References:
- Algorithm suggestions
- From: Geoffrey Summerhayes
- Re: Algorithm suggestions
- From: David Wagner
- Re: Algorithm suggestions
- From: Geoffrey Summerhayes
- Algorithm suggestions
- Prev by Date: Re: Algorithm suggestions
- Next by Date: Re: Algorithm suggestions
- Previous by thread: Re: Algorithm suggestions
- Next by thread: Re: Algorithm suggestions
- Index(es):
Relevant Pages
|
|