From: Adrian Tymes (wingcat@pacbell.net)
Date: Tue Apr 30 2002 - 20:14:49 MDT
Harvey Newstrom wrote:
> On Tuesday, April 30, 2002, at 04:14 pm, Adrian Tymes wrote:
>> This almost exactly duplicates Descartes' infamous musing. If you were
>> uploaded into a VR that *perfectly* replicated your current reality, in
>> a way that you did not remember the upload, then no, it would not
>> matter: by definition, there would be no difference to you. The nagging
>> suspicion that it might matter comes from the insistence on perfection
>> here: we intuitively know that, no matter what we may rationalize,
>> replications (at least from non-digital states like "reality as we now
>> know it") are actually rarely perfect, even if done by very clever
>> people assisted by the best of modern technology (and thus, one suspects
>> despite all rationalization, replications made by even very advanced SIs
>> would suffer from the same limitation in practice), that one state can
>> often be told from another...and in that difference, is what matters.
>
> Isn't this the same as asking if you lie to people and they never find
> out, does it matter? Remember the Truman Show? The guy was kept in a
> lie his entire life. Wouldn't you be mad if you found out? What about
> all the suffering around us. What if that could have been prevented,
> but was simulated just for realism. What if our lives could have been
> so much better, but our boring lives were maintained just to maintain
> the lie.
>
> I sure would be mad, and think it matters to me. The question of how
> could it matter if I don't know, simply means what I don't know won't
> hurt me. People can act unethically and deceitfully as long as they
> don't get caught. I don't believe that. The deceived person is being
> deprived of larger possibilities through fraud. If they were free to
> explore and get out into the larger universe, to make their own
> decision, their lives may be so much better.
Ah...but reread the latter half of my arguement. Yes, logically,
_IF_IT_WERE_PERFECT_ it wouldn't matter. But we know intuitively that
the most likely result of any attempt to achieve it would be far from
perfect, with negative consequences. One of the morals of Truman is
the perils of deluding oneself into relying on technical perfection in
such cases: even with state-of-the-art *everything* shielding from
reality, reality still slipped in.
This, arguably, is what the Luddites are arguing - but they, themselves,
slip into relying on perfection: the fear of perfect disasters,
unthwartable by all of modern technology no matter what we try. Or
perfect terrorists, undetectable until it's too late to stop them. And
so forth.
When one deals with reality, rationality demands accepting imperfections
and dealing with them. There are cases where technical perfection can
be achieved (I know: I've created a few professionally), but these are
usually only the simplest of cases. For anything sufficiently complex,
improving the odds will usually get better results in practice than
completely ignoring odds-improving tactics in favor of perfection.
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 09:13:44 MST