Yes, in general that's true, but the authors are considering the case
where weak measurements are made and the information is fed back into
the system. Quoting from the paper:
"...The time evolution of the system with feedback by weak measurement
is accordingly given by
*rho prime* = U_delta(*rho*) *rho* U_delta_dagger(*rho)
The remainder of this paper will be devoted to exploring the
implications of equation (1). This equation has a variety of interesting
features. The first, perhaps most obvious, is that it can be nonlinear
as a function of *rho*..."
James Donald wrote:
>> At 12:26 PM 5/21/99 -0600, Mike Stay wrote:
>> There are actually two papers on the LANL site that say
>> nonlinear quantum computers imply quantum polynomial time
>> solves NP:
>
>We have, however, quite overwhelming reason to believe that
>the Hamiltonian is linear. Indeed, it is difficult to
>conceive of what it would mean for the Hamiltonian to be non
>linear. If the non linearity was sufficient to have
>perceptible effects in everyday life, reality would be
>unimaginably different.
>
> --digsig
> James A. Donald
> 6YeGpsZR+nOTh/cGwvITnSR3TdzclVpR0+pr3YYQdkG
> G/LpGDPngLqKyrmf7Ku71BC5elnBLW/776UC22OC
> 4G5g4KqHTfsL8trhga5Z6ZLJ/FmZKMWoyJ6GBH9qz
--
Mike Stay
Cryptographer / Programmer
AccessData Corp.
mailto:[EMAIL PROTECTED]
