🚀 go-pugleaf

In article <1e6jpsINNi71@ni.umd.edu> louie@sayshell.umd.edu (Louis A. Mamakos) writes:
>In article <1992Nov15.135015.19748@cubetech.com> andrew@cubetech.com writes:
>
>>I just use ForYourEyesOnly that came as a 2.0 Demo app...  It's a
>>NeXTSTEP app, you can insert files (but not directories, sigh) and use
>>different fonts like NeXTmail, you can mail the ciphertext from the
>>services menu, and...... it's legal to use in the US (whereas PGP
>>(Pretty Good Privacy) is very ILLEGAL in the US).
>
>If FEE is legal to use in the US, then why did it mysteriously
>disappear from 3.0?  Is NeXT afraid of a legal challange from RSA?
>I've been very curious about this since the early 3.0 announcements
>made a really big deal about this capability.

NeXT could not export 3.0 if it included FEE.  In this case it was
"all or nothing."  There was no point in trying to deal with all the
hassles of exportable and non-exportable versions just for FEE.

>Also, while you may trust ForYourEyesOnly to not have any trap doors
>in it, I don't.  Will NeXT tell you how it works?  Will you believe
>them?  For example, the "Secure" application that's on the Educational
>Software Sampler CDROM looks rather cool.  Its description claims:
>
>	Description:
>
>	Secure provides private single key encryption to user files.
>	It can be used either as a standalone or Services menu
>	application.
>
>It in fact does this.  But if you are the paranoid type (and it pays
>to be if you're talking cryptography) and you "strings" your way
>through the application, you find that is uses the UNIX 'crypt'
>command!  While crypt makes your file unreadable to the naked eye, it
>certainly has no real degree of security that I'd want to depend on;
>certainly not compared to DES, which is what I expected to find.
>Search "archie" for cbw, the Crypt-Breakers Workbench.

FYEO uses either FEE or EE.  EE is a well documented algorithm and it
is recognized as secure.  While the mathematical speedups in FEE may
weaken the algoritm, it comes down to "how secure must this really
be."  Crypt is not secure.  I would say that DES is not secure as well
if you want to talk about theoretical absolutes.  I would say that DES
is "secure enough" unless you need to keep things secure from
governments with the resources to break it.


andrew
-- 
andrew@cubetech.com     | "Carpe Nocturn - I get more work done after
Andrew Loewenstern      |  2 a.m. than most people do all day."
Cube Technologies, Inc. |                          - Simson Garfinkel