1997-06-24 - Re: New PGP signatures

Header Data

From: tzeruch@ceddec.com
To: Lucky Green <shamrock@netcom.com>
Message Hash: 8a5d8e89be15f11eb6d5da6698c8a0dd67396383621bae390810e9a39d760c5d
Message ID: <97Jun24.124732edt.32257@brickwall.ceddec.com>
Reply To: <Pine.3.89.9706231540.A13824-0100000@netcom2>
UTC Datetime: 1997-06-24 16:53:48 UTC
Raw Date: Wed, 25 Jun 1997 00:53:48 +0800

Raw message

From: tzeruch@ceddec.com
Date: Wed, 25 Jun 1997 00:53:48 +0800
To: Lucky Green <shamrock@netcom.com>
Subject: Re: New PGP signatures
In-Reply-To: <Pine.3.89.9706231540.A13824-0100000@netcom2>
Message-ID: <97Jun24.124732edt.32257@brickwall.ceddec.com>
MIME-Version: 1.0
Content-Type: text/plain

On Mon, 23 Jun 1997, Lucky Green wrote:

> On Mon, 23 Jun 1997 tzeruch@ceddec.com wrote:
> [On PGP switching to SHA-1/ElGamal]
> > They could have supported RSA/SHA if they wanted to.  They also put 3DES
> > and IDEA support as options in the manual (but not in the freeware or
> > tryit versions that I can see).
> I guess suggestions such as the one above prove that CP is still attracting 
> newbies. That's a good thing. Now if they only read the FAQ.
> The patents for DH (which cover the public domain ElGamal) expire this
> Fall. By using SHA-1/ElGamal, PGP is moving to a technology that will soon
> no longer require paying large sums of money to RSA for the use of a
> one-line mathematical formula. This is *a good thing*. 

PGP 5.0 is still covered by patents since it still has RSA in there to
read old messages.  Also (except for the legal pettifoggery), PGP is
supposed to have licenses to RSA. 

I was speaking technically, not politically.  If MD5 is broken, replacing
MD5 with SHA1 is the fix.  If my transmission is broken, I don't also need
a new engine.  The antecedent post mentioned the reason to change PGP was
a technical flaw.  My point (which echos the FAQ) is that there were other
forces at work - legal and economic factors.  So prepend "If all they were
interested in was fixing TECHNICAL weaknesses in PGP " to my comment.

If abandoning one standard and requiring the world to change to a
radically changed incompatible new one is "a good thing", it is only
because paying extortion is not, which is a sentiment I agree with.

I do think it is a shrewd move, since now the worst thing RSA can do to
them is not allow them to sell backward compatible versions in the US,
were I in their position I would probably do the same thing.

On the other hand, if PGP5+ messages are not covered by any licensing
restrictions on the technology, then PGP5+ can be cloned (i.e. a fully
interoperable version without any code from PGP inc.).