Subject: Re: intel supports content protection
From: Santiago Gala <sgala@hisitech.com>
Date: Thu, 25 Aug 2005 10:22:28 +0200
Thu, 25 Aug 2005 10:22:28 +0200
El mié, 24-08-2005 a las 13:51 -0700, Jonathan Ryshpan escribió:
> >From todays San Jose Mercury:
>         Intel's new chip will feature qualities such as computing in
>         64-bit chunks instead of today's 32-bits, as well as
>         virtualization, so that more than one operating system can be
>         run simultaneously. ==>Built-in hardware encryption will provide
>         better security.<==
>         
> Does this mean that future operating systems will support actual
> **working** content protection schemes?
> 

There is an interesting academic paper (don't have the URL right now),
which demonstrates that all content protection schemes are breakable.

Basically the premise is that there is no way for a computer program to
know if it is running natively or emulated. Interestingly enough,
virtualization is quoted in the previous sentence to the one you're
highlighting.

It is fairly similar to the old "physical" anti copy devices, where a
sector of the floppy was burnt to leave a physical pattern which was
read by the program.

I remember in the company I worked for, they had one of those dBase III
floppies, and it required the original to be inserted every time you
started the program. The original wore off with time, so they had to
install a cracker, much more user friendly, and store the useless
original with the license.

The cracker was a resident program (good old DOS days...) that installed
a modified disk handler, which failed in exactly the way the program
wanted, returning the defective sector from a non existent floppy to the
happy program.

Similar cracking schemes existed for those parallel port dongles.

Regards
Santiago

> jon
> 
> 
-- 
Santiago Gala <sgala@hisitech.com>
High Sierra Technology, SLU


["application/pgp-signature" not shown]