Re: Should be in crypto for criminals Re: just stupid?

Crypto_at_S.M.S
Date: 07/19/05 

Date: Tue, 19 Jul 2005 12:04:33 +1000

Joe Peschel wrote:
> Crypto@S.M.S wrote in news:11doh4u3b1e119d@news.supernews.com:
>
>
>>Joe Peschel wrote:
>>
>>
>>>Crypto@S.M.S wrote in news:11doef71iqve79d@news.supernews.com:
>>>
>>>
>>>
>>>>Joe Peschel wrote:
>>>
>>>
>>>>>It's not a matter of my winning a discussion; it's a matter of your
>>>>>refusing to accept the profundity of Ashwood's cryptanalysis.
>>>>>
>>>>
>>>>If it was so profound, it would have revealed the clear text.
>>>
>>>
>>>No, you have ignored and misunderstood what Joe actually said.
>>>
>>>
>>>
>>>>>>that you would have posted the results if you had any.
>>>>>
>>>>>
>>>>>I'm not working on a ciphertext-only crack.
>>>>>
>>>>>J
>>>>>
>>>>
>>>>In real life, a cipher text only attack is what you need
>>>>to mount to uncover the clear text being sent between your enemies.
>>>
>>>
>>>That's not true. Often you can mount known-plaintext,
>>>chosen-plaintext and chosen-ciphertext attacks. Enigma messages, for
>>>instance, were often cracked using such techniques.
>>>
>>
>>In general, yes; with CryptoSMS in particular, no.
>>
>>Multiple encryption prevents the use of known and chosen plain text
>>attacks, because the plain text is hidden from each layer of
>>encryption. Same goes for chosen cipher text attacks:
>>
>>http://www.ciphersbyritter.com/GLOSSARY.HTM#AlgebraOfSecrecySystems
>
>
> You are layering, however, weak ciphers and hashes, and you have done no
> examination of how a specific layer will affect secrecy. You're only
> guessing that layering must be stronger.
>
>>And do not forget we are only talking about messages transported by
>>SMS.
>>
>>
>>>>Since this discussion was originally framed as "crypto for
>>>>criminals", let's use that as the example. Do you think that law
>>>>enforcement would be given cribs of any kind, when they attempt to
>>>>break a criminal's encryption?
>>>
>>>
>>>Law enforcement, or any attacker for that matter, doesn't need to be
>>>given cribs. Cribs are often "probable word" attacks. An attacker
>>>might expect to find the name of a correspondent, high frequency
>>>words, and expected words encrypted in the messages.
>>>
>>
>>Statistical attacks require large volumes of encrypted messages
>>to work with. No such thing exists when all that was sent were a few
>>short messages.
>>
>>
>>>>In a true, real world attack, you would be faced with cipher text
>>>>only, and have no clues about message content and/or pass phrases.
>>>
>>>
>>>That's just isn't true. See my comment above.
>>>
>>>J
>>>
>>>
>>
>>That just is true. See my comments, just below yours.
>
>
> No, probable word attacks, chosen-plaintext and chosen-ciphertext attacks
> are still possible. Layering one cipher on top of another cipher does not
> necessarily prevent these attacks.
>
> J
>
>

Yes it does:

http://ciphersbyritter.com/NEWS5/RISKRELY.HTM

Relevant Pages

  • Re: Should be in crypto for criminals Re: just stupid?
    ... >> Multiple encryption prevents the use of known and chosen plain text ... layering weak layers of materiel to form a much stronger composite ... >> Statistical attacks require large volumes of encrypted messages ...
    (sci.crypt)
  • Re: How to choose an IDS/FW MSS provider
    ... detect attacks by inspecting layer 3 headers for prohibited IP ... Layer 4 firewalls detect ... facility with an IDS or IPS deployed. ...
    (Focus-IDS)
  • Re: Layer 2 Trace
    ... no equivalent in the various layer two protocols. ... Audit your website security with Acunetix Web Vulnerability Scanner: ... Hackers are concentrating their efforts on attacking applications on your ... Up to 75% of cyber attacks are launched on shopping carts, forms, ...
    (Pen-Test)
  • RE: Re: mcafee DDOS solution
    ... ISP can block volumetric DDoS attacks (layer 2/3). ... The right strategy is ISP service for volumetric attacks and on premise DDoS ...
    (Security-Basics)
  • Re: unsuccessful hacking attempt at my machine
    ... the reasons I tend to use the word cracker and rarely say hacker as far ... needed such as Denial of Service attacks. ... and this is what white-hat hackers are about. ... Watchfulness and logging is another layer. ...
    (comp.os.linux.security)