Re: When Programs Benefit

From: Louis Newstrom (louisnews@comcast.net)
Date: Thu Jun 13 2002 - 06:45:33 MDT


From: "Eugen Leitl" <eugen@leitl.org>

>> Quoting Louis:
> > Although this is true, I wouldn't jump to the conclusion that destroying
one
> > out of two has practically no effect. You are cutting in half the
number of
> > people who can read a book.
>
> No, you're cutting in half the amount of books people could read. Or,
> movies, rather.

I disagree. I AM cutting in half the amount of Star Wars films, I could
watch, but from my point of view watching one is the same as watching
another.

I was referring to the number of theatres that could carry the one existing
film. And the number of people who could watch it in that one theatre. I
really meant that at any showing date, only ONE threatre-full of people can
watch it, as opposed to TWO theatre-full of people. I.e. half the people
could see the film.

> On the other
> hand, since you'll never see those people (otherwise they would produce
> nonidentical input, making you fork), you can safely ignore it.

This argument always drives me crazy. It's like a murderer saying if he
doesn't get caught, it doesn't matter that he committed the crime. Not
seeing your consequences does NOT mean that there are no consequences. It
only means you didn't see them.

As to the "(otherwise ... making you fork)", I have always contended that
"making you fork" is the proof that these "synched" thought experiments are
faulty. If you can't allow anything that would cause a diference between to
objects, that would take a lot of effort. And, yes, it can be shown that
that amount of effort is directly related to the number of objects. If your
experiment requires bodyguards to keep Star Wars fans from talking to the
person in the experiment, you would need at least one bodyguard for each
"instance" of your subject. In other words, the number of "instances" of
the subject makes a big difference.

I think this is getting back into the "if two objects are identical in every
way are there one or two of them" argument. I have always maintained "two",
but this thought experiment hinges on a person's opinion of that argument.



This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 09:14:46 MST