>Anyone who receives a digital signature based on an MD-5 or SHA-1 hash
should be suspicious.

aw, c'mon mike - you don't *really* mean that, do you?

i mean, really - should we be suspicious, just in case they someone has been
running "SHA1-Crack-2005" for the last 3 days on their 200,000,000 cpu
cluster in their basement with each magic cpu capabable of 10M hashes per
second?  (i think i did my math right for 2^69..)

and if i understand correctly, the chinese research only showed it was
possible to generate md5 colliding hashes w/ after an hour w/ their top500
supercomputer - but they couldnt actually crack one, and md5 still hasnt
ever been cracked by brute force.

so i agree w/ jason and michael in that it's interesting that MD5/SHA1 are
"broken" by some definition of broken, and that we should be aware of this -
i just want to inject a healthy does of "so what?" into the scene.  no
disrespect intended - just healthy discourse. ;-)

Josh Coates
http://www.jcoates.org

-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] Behalf Of Michael Halcrow
Sent: Wednesday, February 16, 2005 7:20 PM
To: BYU Unix Users Group
Subject: Re: [uug] SHA-1 is probably broken


On Tue, Feb 15, 2005 at 10:18:43PM -0700, Phillip Hellewell wrote:
> <tinfoil hat>
>
> Hmmmm, come to think of it, how can we trust this so called e-mail and
> so called break of SHA-1.  I mean, hey, they didn't even really prove it
> yet, so how do we know it's not just some huge conspiracy; hey, maybe
> someone is trying to impersonate Michael and ruin his good name by
> spreading these nasty rumors!!  Since there's no signature, we don't
> really know _WHO_ sent that message!!!
>
> </tinfoil hat>

I certify that the previously sent messages with SHA-1 hashes for the
bodies:

ce57b00152bc4d28fb6d1db7c0942d234d7061c5
21d6067691cc57fd69682adc946576cac6f653a7
342abccf1e67c06f89760f0f719fd75221e87b62

Were sent by me.  You can take your tinfoil hat off now.  ;-P

BTW, you can still use GnuPG with a hash that still is not known to be
broken.  Place this in your gpg.conf or your .gnupg/options file:

digest-algo RIPEMD160

Note this is outside the RFC2440 spec, but it should be supported by
any PGP app that's worth its salt.  Anyone who receives a digital
signature based on an MD-5 or SHA-1 hash should be suspicious.  Well,
you should always be suspicious, but in those cases, you should be
especially suspicious.

Mike
.___________________________________________________________________.
                         Michael A. Halcrow
       Security Software Engineer, IBM Linux Technology Center
GnuPG Fingerprint: 05B5 08A8 713A 64C1 D35D  2371 2D3C FDDA 3EB6 601D

Friends don't let friends do Windows.

--------------------
BYU Unix Users Group
http://uug.byu.edu/

The opinions expressed in this message are the responsibility of their
author.  They are not endorsed by BYU, the BYU CS Department or BYU-UUG.
___________________________________________________________________
List Info: http://uug.byu.edu/cgi-bin/mailman/listinfo/uug-list


--------------------
BYU Unix Users Group 
http://uug.byu.edu/ 

The opinions expressed in this message are the responsibility of their
author.  They are not endorsed by BYU, the BYU CS Department or BYU-UUG. 
___________________________________________________________________
List Info: http://uug.byu.edu/cgi-bin/mailman/listinfo/uug-list

Reply via email to