Re: Quantum Computers [was Read any good books lately?]

John Clark (jonkc@worldnet.att.net)
Thu, 26 Aug 1999 13:33:45 -0400

Dan Fabulich <daniel.fabulich@yale.edu> Wrote:

>You know, sometimes I wonder if I'm in John Clark's killfile.

No, not at all, you'd be one of the last. I didn't want you to feel insulted so I found some stuff to argue about.

>Clark stated it wrongly: it's not that MWI doesn't have to define a
>measurement, it's that it CAN define a measurement in a physically
>rigorous way.

All measurements split the world but, if we're consistent with the common usage of the word, not all world splits are measurements.

>it doesn't have to define what an "observer" is,

Agreed.

> measurement is an irreversible process.

Usually, not always.

>I have to agree with Moss, however, who notes that the presence of
>conventional detectors in Clark's quantum computer scenario would ruin the
>test, since conventional detectors would fail to completely revert when
>the quantum computer erased its memory.

If you're right then many world's is wrong and you have a way to prove it. Many worlds explains the 2 slit experiment by saying everything splits when a photon encounters the slits and rejoins when they hit the film and no longer exist in either universe. If the theory is wrong and the universes do not merge back together then we'll see no interference bands and we've proven the theory wrong, if we do see the bands then you're wrong and the worlds do merge. Either way an interesting experiment.

>Certainly we'll have some kind of AI in the next 50 years, but I strongly
>doubt that it'll be thermodynamically reversible.

A Drexler style nanocomputer would have to be thermodynamically reversible, otherwise the contraption would give off so much heat that the thing you'd end up with would not be a computer at all but a pretty good high explosive.

John K Clark jonkc@att.net