Re: Potential use case for opaque space bank: domain factored network stack

[Anton Tagunov]

Pierre THIERRY wrote:
Pierre> Could someone with better understanding of TC could tell if 
support of
Pierre> opaque memory in Hurd would really change anything for 
implementors of a
Pierre> DRM system using TC in Hurd?

Let me try. Here's the task:


Some commercial vendor produces
(1) software implementing proprietary algorithms
(2) DRM protected data
The vendor wants to keep (1), (2) secret.


We, the Freedom Fighters what to empower the user to
(1) reverse engineer algorithms from software
(2) make a backup copy of DRM protected data


How does the vendor go about keeping (1), (2) in secret?
He encrypts .exe files and data files.


The vendor keeps keys to decrypt (1) and (2) in the TC chip.
TC chip shall only let the keys to be used if "proper" software is run.


How could we empower user to decrypt (1), (2) in presence of a TC chip?
We could empower the user to inspect memory of any running process.
Without the process being able to tell if it is under inspection or not.


Marcus, Pierre, Jonathan, how do you like this setting?
cheers, Anton