On Thu, 9 Mar 2000 GSLevy.domain.name.hidden wrote:
> jqm1584.domain.name.hidden writes:
> > That's not such an easy question, but all approaches must deal
> > with it eventually. F'rinstance, if you believe your measure will be
> > split if you have 2 conscious continuations, and not if there's just one.
> > But actually in my case a conditional 'probability' distribution
> > is fine, so for example if one could identify all human-like conscious
> > computations, one could ask for the conditional effective probability
> > distribution within that set. Choose any set of computations you want,
> > really, but be careful to note the meaning of the chosen denominator. It
> > will still be an objective, well defined feature of reality.
> > Remember, too, that with a finite regularization in place the
> > measures in question are finite integers. For decision making and so on,
> > it's fine to work with a regularization, get the answer as a function of
> > that, and then take the limit.
>
> This is porcine effluent as well as bovine coprophilia, enough to give
> abdominal discomfort to a coprophagic melanogaster.
It's getting a bit too unwieldy for my taste. I'm afraid I'll
have to return to the tried and true. I gave it a shot.
> This is definitely the weak point in you argument. You seem to assign the
> "consciousness" property with the same abandon and arbitrariness of a voodoo
> grand master. You have no OBJECTIVE way of assigning consciousness which is a
> purely SUBJECTIVE property. In addition, your argument depends on identifying
> all human-like conscious computations over the scope of the Plenitude, which
> is ludicrous because 1) you don't know how to characterize "human-like"
> consciousness, 2) the human-like property may not be relevent - there may be
> other types, and 3) the size of the Plenitude makes any evaluation
> impossible.
Bullshit. You have so much, do you own a cattle farm?
0) Consciousness is an objective property. If it did not objectively
exist, there would be no one around to see this argument.
1) Cognitive science would be useful in trying to guess which computations
have the property; the Turing test would not suffice.
2) So? It was just an example, one chosen because it would be useful in
making many predictions of interest.
3) I have already explained how to use a regularization. The fact that
the right way to do it for the full plenitude is unknown is a problem for
the argument that the plenitude is real, one we should all hope to solve.
> Why not use the differential approach already in use in Physics for dealing
> with infinite quantities such as vacuum energy. In Physics, the only energy
> of relevence to the world we live in, is the difference between the actual
> energy and the ground state. Should you decide to apply this concept to
> "measure," all you'll have to concern yourself with is the difference between
> the measure of two objects, or the relative value of their measures.
See 2) above and realize that I already was doing that.
- - - - - - -
Jacques Mallah (jqm1584.domain.name.hidden)
Physicist / Many Worlder / Devil's Advocate
"I know what no one else knows" - 'Runaway Train', Soul Asylum
My URL:
http://pages.nyu.edu/~jqm1584/
Received on Wed Mar 15 2000 - 13:05:48 PST