|
|
 Up |
|
|
  |
Author: Scott ContiniScott Contini
Date: Dec 12, 2006 20:51
Simon Johnson wrote:
>> As an aside, I mention that Ron Steinfeld had a great result at this
>> year's
>> Asiacrypt for RSA generator with known modulus. HE modified the
>> Fischlin-
>> Schnorr reduction to get provable security while outputting log(M) bits
>> per
>> iteration. In my opinion, this is a breakthrough.
>>
>> Scott
>
> I would be interested if you could provide me with a reference to this
> work?
>
> There does not appear to be a paper with such a title on the guy's
> site.
>
> Simon
|
| |
|
| |
no comments
|
|
|
Author: UnruhUnruh
Date: Dec 12, 2006 13:31
>> Edible One-Time Pad Books (Pair). CAN BE EATEN IN AN EMERGENCY, or a
>> page at a time as each page is used. 5000 high-quality random numbers.
>> Decoy cover conceals true purpose of book. Discreet palm-size 7cm x
>> 4cm. Ideal cryptogift.
>Hang on... I can see a couple of problems here:
>*) 5000 "high-quality random numbers" per book, each number intended to
>encrypt a single character - if you take a look at your "sent mail" folder,
>this would probably only cover about 10 emails per book!
>*) How do we know you don't work for a TLA, and have your own copy of the
>numbers?
I have no idea what a "high quality random number" is if it is generated by
someone else.
>Only kidding - I think they'd make a pretty *cool* xmas present for a
>cryptogeek, though I wouldn't "write home to my aunty in Russia" using them
>;)
|
| Show full article (1.08Kb) |
|
| |
1 Comment |
|
|
Author: UnruhUnruh
Date: Dec 12, 2006 12:33
"Simon Johnson" gmail.com> writes:
>Mark Probert wrote:
>> Hi, all.
>>
>> Has anyone done any testing to find out how good the .Net pseudo-random
>> number generator class RNGCryptoServiceProvider is?
>>
>> Regards,
>>
>> -mark.
>It is the Windows equivalent of /dev/urandom is considered
>cryptographically secure.
And you know this how?
|
| |
|
no comments
|
|
|
Author: David TaylorDavid Taylor
Date: Dec 12, 2006 12:29
On 2006-12-12, framarks@ molcho.com molcho.com> wrote:>> Edible One-Time Pad Books (Pair). CAN BE EATEN IN AN EMERGENCY, or a
>> page at a time as each page is used. 5000 high-quality random numbers.
>> Decoy cover conceals true purpose of book. Discreet palm-size 7cm x
>> 4cm. Ideal cryptogift.
>
> Hang on... I can see a couple of problems here:
>
> *) 5000 "high-quality random numbers" per book, each number intended to
> encrypt a single character - if you take a look at your "sent mail" folder,
> this would probably only cover about 10 emails per book!
If you're manually performing a OTP operation on the entire message, you'll
probably want to send abbreviated messages in any case. And you could
always buy more than one...
> *) How do we know you don't work for a TLA, and have your own copy of the
> numbers?
|
| |
|
no comments
|
|
|
Author: Mark ProbertMark Probert
Date: Dec 12, 2006 11:32
Hi, all.
Has anyone done any testing to find out how good the .Net pseudo-random
number generator class RNGCryptoServiceProvider is?
Regards,
-mark.
|
| |
|
7 Comments |
|
|
Author: UnruhUnruh
Date: Dec 12, 2006 11:04
"Simon Johnson" gmail.com> writes:
>> Hello,
>>
>> Now, I know that cracking DES brute force takes forever if you don't have
>> custom chips. And, my intention is not to crack DES but to find working
>> implementations of brute force DES attack (yes, it's for my diploma paper)
>> :-) That is, I am looking for a windows or linux application which would
>> crack DES if I ran it for few decades on a PC (or a network of PCs, and I
>> think that my mentor could possibly provide me with a grid too). Any
>> ideas?
Not at all sure what you are asking for. Assuming you have an encrypted
text and the plaintext for it, then
|
| Show full article (1.15Kb) |
|
no comments
|
|
|
Author: secretcodebreakersecretcodebreaker
Date: Dec 12, 2006 10:25
John,
This is off topic.
I tried to send you an e-mail at both the ecn and freenet addys.
Both bounced - "...no mailbox here by that name."
Can you send me an e-mail with a valid return address, please.
Sent it to my secretcodebreaker.com domain address. Use
secretcodebreaker as a user name (mailbox name).
Thanks.
Bob (Reynard)
jsavard@ecn.ab.ca wrote: > Anne & Lynn Wheeler wrote:
|
| |
|
no comments
|
|
|
Author: SergeiSergei
Date: Dec 12, 2006 09:50
Kristian GjЬsteen wrote:
> Sergei gmail.com> wrote:>> Just want to hear your opinion: Does Privacy Preserving Data Mining make
>> any sense to you?
>
> I think it does. I'll ramble on for a couple of paragraphs, but I'll
> state up front that I'm not familiar with the litterature. Also, what
> I'm saying is either obvious or wrong.
>
> I think multiparty computation is a nice topic to compare with. We
> define security of multiparty computation in terms of an ideal process
> where every participant hands his private data to a trusted third party,
> the TTP computes the result and hands it back. The protocol is secure if
> it its "effect" is indistinguishable from the ideal process. The idea is
> that if the result of the computation somehow reveals the private input
> of some party (perhaps when combined with the private inputs of some
> other parties), that's impossible to avoid. But we don't want any other
> compromises of private input, and there are none in the ideal process.
>
> Given this, a natural approach to privacy-preserving data mining is ...
|
| Show full article (3.63Kb) |
|
no comments
|
|
|
Author: Nikola SkoricNikola Skoric
Date: Dec 12, 2006 09:30
Hello,
Now, I know that cracking DES brute force takes forever if you don't have
custom chips. And, my intention is not to crack DES but to find working
implementations of brute force DES attack (yes, it's for my diploma paper)
:-) That is, I am looking for a windows or linux application which would
crack DES if I ran it for few decades on a PC (or a network of PCs, and I
think that my mentor could possibly provide me with a grid too). Any
ideas?
Why do I need that? Because I implemented DES (in C#, lisp and did a FPGA
chip in VHDL), warped it in a for loop and started searching key-space to
match my plaintext and my cyphertext. And, of course, it takes forever.
What I want now is just compare the speed of my implementation to that of
best (although, for start, any would be just fine :-) ) existing software
brute force DES cracker. I tried to justfuckingoogleit, but there's so
much background noise. I'll keep trying, but if anybody here knows a link
or two I could use, I'd be more than thankful...
|
| Show full article (1.21Kb) |
|
3 Comments |
|
|
|
|
|
Author: Diego TaylorDiego Taylor
Date: Dec 12, 2006 07:44
Hi,
I am using a feistel network with a symmetric algorithm known function
on each round to generate output of length lesser than a standard
cipher block. (i.e: 60 bits)
My question cames about using it for odd number of bits (i.e: 59 bits)
requiring an unbalanced feistel network. I have seen just the abstract
of this paper: "Generalized Birthday Attacks on Unbalanced Feistel
Networks": http://www.springerlink.com/content/6thw7lq7el11qefj/
I would like to know if using rounds of 30 and 29 bits will bring me
too much insecurity to the proposed scheme.
Thank You,
Diego
|
| |
|
no comments
|
|
|
|
|
|
|
