Re: SSN encryption
Available news archives: comp.lang.tcl - comp.lang.python - comp.security.firewalls - sci.crypt - comp.lang.php - comp.lang.javascript
Google
 
Web news.hping.org


sci.crypt archive

Re: SSN encryption

From: Peter Pearson <ppearson@nowhere.invalid>
Date: Fri Sep 30 2005 - 19:13:19 CEST

drfremove@nber.org wrote:

[regarding Paul Rubin's suggestion in]
>
> http://groups.google.com/group/sci.crypt/msg
c4bdd165ba12b92a?dmode=source&hl=en

> Could the key be determined from knowledge of
> some *other* SSN/hash pairs?

No. Resistance to a known-plaintext attack is required
of a good cipher, and Paul Rubin has given you a good
cipher. The attacker's best option is to get a known
plaintext-ciphertext pair (e.g., by recognizing his
own record in the database) and then to guess keys
until he finds one that maps that plaintext onto that
ciphertext. So be sure to choose a key that he won't
guess early on. You can choose a nice key like this,
but with a longer randomish string:

echo 'asldkjfie35nav982hu23asd;klgjiup2389hb' | md5sum

-- 
Peter Pearson
To get my email address, substitute:
nowhere -> spamcop, invalid -> net
Received on Sat Oct 15 04:38:02 2005