lkml.org 
[lkml]   [1996]   [Jun]   [22]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: [PATCH] DES LOOP Supportx
Date
"KMB" == Kevin M Bealer <kmb203@psu.edu> writes:
KMB> DES is a fairly recent US government standard correct? The US government

Err, define recent. It has been around for a few years and has been
tested over and over again. Publically, that I know of, the only way to
break it is exhaustive key search. Of course, 56bits is becomeing more
and more anemic by the month. The US govt had a hand in its
development.


KMB> has proposed that all encryption methods have back doors so that the
KMB> government can get into them, correct? Would they have designed DES in such
KMB> a way that it is unbreakable, or in such a way that a person who knew how to
KMB> get into it, can do so at will.

That's an oversimplification of key escro policies. Weather or not they
can break it "at will" and not with an enormous amount of CPU power is
still a matter of folk lore and rumor mill. Of course, it is not as
strong as IDEA which is also being trusted more and more by the general
public.

KMB> Ergo DES is compromised, and it is only until whoever wants to bad enough
KMB> can buy from the right person(1) or figure out what the method is.

That's an ergo based on assumptions and possibilities. DES is the
reccomended algorythm by los federales. Weather this is because they
want to easily spy on you or prevent you from trusting something
emberrasingly weak and make the FBI's job harder tracking down the
criminal who robbed a bank because he could "man in the middle" attack
your ATM transactions is a toss up. There is one instance that I know
of where an old gcc2.x compiler bug trivialized the keyspace on an
implementation of DES and that does make me itchy about the possibility
of a mathematical trap door but once again, conjecture and not fact.
This leads us back to IDEA which is newer than DES but hasn't shown any
weakness that I know of.

KMB> It looks to me like anything powerful enough to be really secure will be
KMB> very slow for full system encryption.. (I am thinking mostly of RSA/MD5
KMB> based things which I admittedly don't know enough about.)

Not as such. The readme files on PGP contain a wealth of information.
Likewise, there are several good www sites out there. Among them are:

http://www.cs.hut.fi/ssh
http://www.enter.net/~chronos/cryptolog.html
and, of course, http://www.eff.org.

I, once again, bring up IDEA. It's 128 bit keyspace is much stronger
than DES, faster than triple DES and considerably faster than RSA on a
key size that would give comperable security (I think around 3100 bits
or so).
<PLUG TYPE="shameless">We need an implementation of IDEA for the
loopback device that is developed as PGP was (yes I think the ITAR
restrictions are silly on this matter, and I vote).</PLUG> Any takers?

KMB> (1) am I suggesting there may be non-angelic people in the gov't?
KMB> who might even be for sale(2)? .... shame on me.

millner@millner:~$ man 2 sale
No entry for sale in section 2 of the manual

Ok, serious again. In almost any random sample of people, there is
always an asshole. The govt is no different. This is why the
constitution tries to limit the government the way it does. It's
really inefficient sometimes and occasionally it seems downright
crippled but they knew. If the government could move with lightning
speed to do its own bidding then it would be extremely efficient at
doing all the wrong sorts of things.

KMB> (2) am I allowed to say this? (looking both ways)

The day you are not is the day I leave this country for good. The day
they would stop me from leaving because I thought the govt was going
straight to hell and there were no more options to turn it back is the
day I get nasty. Oh yea, and I stil vote.


To randomly insert something to make it pertanant to this list again.
IDEA encryption for the loopback device anyone?

Rob

--
"Do human beings have souls?"
"No. The question is inherently misleading.
Souls have human beings."


\
 
 \ /
  Last update: 2005-03-22 13:37    [W:0.079 / U:0.076 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site