Re: To Jonathan [backdoor]

[Marcus Brinkmann]

At Wed, 10 Jan 2007 02:08:07 +0300,
Anton Tagunov <address@hidden> wrote:
>
> Pierre>  I think the backdoor is an *unsafe* default, actually.
> 
> You're quite right.
> 
> We should decide what is more dangerous to an average Joe:
> data loss/vendor lock-in or data disclosure?

This generally depends on the type of data, even for average Joe.

If you assume a right for absolute privacy, even if only under some
conditions, then there should be a mechanism which allows to keep some
information absolutely private.  It seems that most people think that
there should be a right to absolute privacy under some conditions.

This does not affect the "trusted computing" (more specifically:
remote attestation) scenario, where data is not held private but
disclosed under tightly controlled circumstances.  In this sense,
"trusted computing" is the anti-thesis of privacy.

What I don't understand is what you mean by backdoor.  To me, an
utmost secure OS and the ability to inspect are not in contradiction,
and where you seem to want to see a backdoor I just want to see
authorized access.  One example is Owner Override described in:

http://www.eff.org/Infrastructure/trusted_computing/20031001_tc.php

Thus, I am not sure what this is all about, I am looking forward to
your mail describing backdoor design.

Thanks,
Marcus