Re: grub to help refund of pre-installations

[Robert Millan]

On Sun, Jul 13, 2008 at 03:45:18AM +0200, Yoshinori K. Okuji wrote:
> On Tuesday 08 July 2008 07:57:30 Jean-Christophe Haessig wrote:
> > Le mardi 08 juillet 2008 à 07:32 +0200, Yoshinori K. Okuji a écrit :
> >
> > <snip>
> >
> > In fact, you want to write a GRUB module which deciphers a partition
> > when given a key. That's no DRM in any way. In the DRM world, users
> > don't (and can't) get the real keys.
> 
> You are completely right. I don't try to mean that it is a sort of DRM by 
> itself. My concern is rather about making it too easy to implement DRM on the 
> top of GRUB. For example, if the logic is present, it is very straightforward 
> to get a key from somewhere else but an user input (e.g. by using a BIOS 
> call), and hide it from the user. I can think of many nasty ways to hide the 
> key, even if GRUB is free.

TC/DRM proponents tend to mix legitimate security features with their
handcuffware in order to confuse the public.

I think the best response to this is to implement any security features
as long as they're legitimate and don't rely on treachery-encumbered
technology (such as TPMs).  This would contribute to split both things,
and when they're only left with their illegitimate features, they've
already lost (because nobody will buy them).

What's happening nowadays is that people start to use TPMs as if they were
a security feature, which is really bad as it reduces opposition to them
dramaticaly.  I think the technical benefit they get by reusing our code is
minimal in comparison.

-- 
Robert Millan

<GPLv2> I know my rights; I want my phone call!
<DRM> What good is a phone call… if you are unable to speak?
(as seen on /.)