1997-01-29 - Crypto verification needed for Pretty Safe Mail

Header Data

From: Anonymous <nobody@replay.com>
To: cypherpunks@toad.com
Message Hash: 06ea1bc64256c28b3b01d131d9b50415bb7e1f9608c17e61e9565ebb731c7950
Message ID: <199701290300.TAA15110@toad.com>
Reply To: N/A
UTC Datetime: 1997-01-29 03:00:19 UTC
Raw Date: Tue, 28 Jan 1997 19:00:19 -0800 (PST)

Raw message

From: Anonymous <nobody@replay.com>
Date: Tue, 28 Jan 1997 19:00:19 -0800 (PST)
To: cypherpunks@toad.com
Subject: Crypto verification needed for Pretty Safe Mail
Message-ID: <199701290300.TAA15110@toad.com>
MIME-Version: 1.0
Content-Type: text/plain


There have been questions raised on the newsgroups regarding
Pretty Safe Mail, arguably the best implementation of a PGP product
for the Macintosh.

In response, the product manager at Highware has requested that
a trusted third party verify the security of Pretty Safe Mail. However,
a NON-U.S. cryptologer is needed.

I'm including the post below. Please forward to any people you know
who may qualify.

Pretty Safe Mail is a GREAT product. If it can be shown to be safe
and effective, it will do wonders to spread the use of strong crypto.

=====

From: axel@highware.com (Axel de Landtsheer)
Newsgroups: comp.security.pgp.discuss
Subject: Re: Concerned about Pretty Safe Mail for Mac


>  I'm concerned about the product "Pretty Safe Mail" for the Macintosh,
>by a company called Highware. I was wondering whether anyone here had
>tried evaluating it at all.
>
>  It is a complete PGP implementation (not a front-end). They claim
>to have licensed some of PRZ's code from PGP. However, as far as I
>can tell, they are not making any of the source code available.

The source code for PSM is indeed not available. We are however eager to
have the code checked by any trusted source. These sources cannot be US
companies because of the new US regulations which state that US companies
must not give such support to overseas companies. Us being a Belgian
company, this makes things a little more difficult. Does anybody have a
suggestion for such a trusted source outside the US?

>  As someone on the comp.security.pgp newsgroups pointed out, writing
>a wonderful user interface on a PGP trojan horse that either crippled
>the session key generator or used the session key to leak random
>portions of secret key primes would be a perfect tactic for a
>government wishing to penetrate PGP security. With such a great
>interface, compared to the original PGP, it can't help but become
>widely used.

PSM is not a Trojan Horse. Does any trusted source want to check that?

>  I realize that without the source code, it's a major hassle, but
>has anyone looked at Pretty Safe Mail (previously called Safemail)
>at all for suspicious behavior? For example:
>
>  1) non-random session key generation?
>  2) non-random key pair generation?
>  3) unnecessary disk access to secret keys?
>  4) anything else?

Many people are worried about the random-key generation because they do not
need to bang away on the keyboard for five minutes when they create a key
pair. Apparently, this seems to loosen their nerves (I heard some people
created about 10 key pairs a day - just a joke).
In short, we use all events that happen on the Mac (clicks, typing, opening
of windows, display of icons, ...) together with the time they happen, and
encrypt these to fill a table which is the starting point for the
random-number generator. Why make people type away on the keyboard if you
have enough random events to start from?

Again, we want to have this checked by a trusted source asap...

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Axel de Landtsheer | Highware, Inc.
                   | 109 av. H. Jaspar, 1060 Brussels, Belgium
Product Manager    | voice: +32 2 537-6810 fax: +32 2 537-5155
axel@highware.com  | http://www.highware.com, ftp://ftp.highware.com
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To obtain my PGP key, send me a message with subject "Send PGP key"








Thread