Antonomasia writes:
> From: "Carl Ellison" <[EMAIL PROTECTED]>
>
> > Some TPM-machines will be owned by people who decide to do what I
> > suggested: install a personal firewall that prevents remote attestation.
>
> How confident are you this will be possible ? Why do you think the
> remote attestation traffic won't be passed in a widespread service
> like HTTP - or even be steganographic ?
The main answer is that the TPM will let you disable attestation, so
you don't even have to use a firewall (although if you have a LAN, you
could have a border firewall that prevented anybody on the LAN from
using attestation within protocols that the firewall was sufficiently
familiar with).
When attestation is used, it likely will be passed in a service like
HTTP, but in a documented way (for example, using a protocol based on
XML-RPC). There isn't really any security benefit obtained by hiding
the content of the attestation _from the party providing it_!
Certainly attestation can be used as part of a key exchange so that
subsequent communications between local software and a third party are
hidden from the computer owner, but because the attestation must
happen before that key exchange is concluded, you can still examine
and destroy the attestation fields.
One problem is that a client could use HTTPS to establish a session
key for a session within which an attestation would be presented.
That might disable your ability to use the border firewall to block
the attestation, but you can still disable it in the TPM on that
machine if you control the machine.
The steganographic thing is implausible because the TPM is a passive
device which can't control other components in order to get them to
signal information.
--
Seth David Schoen <[EMAIL PROTECTED]> | Very frankly, I am opposed to people
http://www.loyalty.org/~schoen/ | being programmed by others.
http://vitanuova.loyalty.org/ | -- Fred Rogers (1928-2003),
| 464 U.S. 417, 445 (1984)
---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
