[117613] in Cypherpunks

home help back first fref pref prev next nref lref last post

Re: Build a better OTP?

daemon@ATHENA.MIT.EDU (Anonymous)
Tue Sep 7 03:31:30 1999

Date: Tue, 7 Sep 1999 09:15:14 +0200 (CEST)
Message-Id: <199909070715.JAA19616@mail.replay.com>
From: Anonymous <nobody@replay.com>
To: cypherpunks@cyberpass.net
Reply-To: Anonymous <nobody@replay.com>

>>>>> Anonymous  <nobody@replay.com> writes:

  > The point is, it is absurd to raise paranoia to the
  > point that you suspect a back door in the RNG chip, if
  > you're not going to be equally concerned about all the
  > other components you are forced to trust.

No, the *point* is that Intel has only provided access to
the RNG through an API that runs on a single, proprietary,
closed-source OS.

As Mr. Gillogly just pointed out, there's no point in
providing a bias-free, uncompromised hardware RNG if it can
only be used with an OS whose security model can only be
regarded as a joke (by anyone but Phillllllllll, that is).

If Intel was forthright and honest about the whole
situation, serious crypto folk might take them seriously.
As it is, they're making some pretty ambitious claims for a
company apparently unwilling to let anyone see what's really
behind the curtain.

So, for all *practical* purposes, these processors must be
regarded as having *no* hardware RNG.


home help back first fref pref prev next nref lref last post