Re: bit-split, or: the schizophrenia of trusted computing

[Marcus Brinkmann]

At Mon, 01 May 2006 12:47:14 -0400,
"Jonathan S. Shapiro" <address@hidden> wrote:
> 
> On Mon, 2006-05-01 at 18:05 +0200, Marcus Brinkmann wrote:
> > At Mon, 01 May 2006 12:01:55 -0400,
> > "Jonathan S. Shapiro" <address@hidden> wrote:
> > > The second point is fundamentally a value judgment, and it cannot be
> > > decided by purely technical means. It implies that there may exist
> > > *some* forms of information encapsulation are not only acceptable, but
> > > may be ethically mandatory.
> > 
> > But as you said, it can not be decided purely by technical means.  A
> > consequence is that in many circumstances, and I have in mind
> > particularly those circumstances which have a broader public impact,
> > technical means may not be the appropriate means to enforce encapsulation.
> 
> In the absence of technical means of support, encapsulation *cannot* be
> enforced in a share-access computing system.

I used the word "enforce" above in the same sense as in "enforcing a
law".  Using your definition, one can not enforce any law.  That may
be true, but I think that the term is nevertheless commonly used in
this way, and rightfully so.

> If you can offer a demonstration of how to do it, you will overturn
> decades of security research going all the way back to the Anderson
> Report and earlier. If you can actually demonstrate this, then trust me
> when I say that your life is completely *wasted* in the area of
> mathematics.

I hope its clearer now.

Thanks,
Marcus