On Mon, May 08, 2006 at 03:59:29PM +0200, Michal Suchanek wrote: > >> Ehm, the chip has two functions > >> -secure key storage > >> -OS attestation > >> > >> If you use a free OS and a piece of hardware that is not completely > >> horrible you should be able to just ignore it. > > > >Or emulate it... > > > > No, the point of attestation is that it cannot be emulated.
We're talking Fritz-chip here, right? Its specs are available; and you can code against that specification. Unless that component becomes integrated into the CPU itself (which is likely in the future); there's always a way to write a driver that would emulate the primitives that are part of that spec. Once we have fritz-on-die, it will become necessary to intercept the opcodes that query it through some virtualization technique, and replace them on-the-fly before sending them to the CPU/fritz-combo (that's what I'm hinting at). Developing this may require some reverse-engineering, and all this may thus be illegal in some countries, but that's merely a social, not a technical obstacle. I'm convinced it *will* happen, and there will be drivers floating around when this stuff becomes ubiquitously mandatory. This chip will stop the unwashed masses from casual copying, that's likely (and that's all what I.P. holders really care about), but it's no real barrier to determined hackers. We have no need to get all too nervous about this. The general public ought to, but it's too passive and uneducated about issues like these, to really care. The public will take whatever politics will dish out, no matter how restrictive that will be. It happened before, it happens today, and it will happen tomorrow too. All in all, TC is an inconvenience; but the legal framework that is being currently assembled accross the world is a much greater danger to our freedom than TC could ever turn out to be. Let's concentrate on the technical issues; at least here we can make a difference. In the mean time, politics will go their own crooked ways, which only big bucks can influence. That's not our domain. > Thanks > Michal Regards, -Farid. _______________________________________________ L4-hurd mailing list [email protected] http://lists.gnu.org/mailman/listinfo/l4-hurd
