On Thu, 28 Jan 1999, Wei Dai wrote:
> On Thu, Jan 28, 1999 at 03:50:18PM -0500, Jacques M Mallah wrote:
> > A lot of things are illusions, but you need to start out with
> > something that is not too trivial either. Strings seem too trivial;
> > dynamics and decisions seem to supply the missing ingredient. Of course
> > there are plenty of people who would object at this point and say that
> > even computations are too trivial. I know what my intuition has to say
> > about where to draw the line. I think there is a deep reason, but I am
> > not able to explain it in words.
>
> It is your responsibility to explain and justify your intuition.
> We can't continue this discussion if you are going to appeal to an
> intuition that only you have access to.
I'd hardly say that only I have access to it, since there are a
lot of computationalists around.
It applies both ways. I do agree that it's not the sort of
thing we can expect to resolve. Like reductionism vs. dualism, there are
always going to be disagreements, I think.
> > Of course your way has rather counterintuitive consequences which
> > computationalism avoids. The recently dead, for example, could have about
> > as much measure in your scheme as the living, since the brain would have
> > about the same chemical structure but would no longer compute. What do
> > you think it would feel like to be such a consciousness? I doubt it would
> > feel like anything, but that's just a restatement of my above position.
>
> It doesn't feel like anything because consciousness depends on the
> structure of neuron firing patterns. Since the brain-dead have no brain
> activity they obviously lack this structure. I don't see how this is a
> problem for my proposal.
There are very likely telltale signs that a neuron has recently
been firing, which is just as good in your proposal. But suppose it's an
AI with non-volatile memory. It would still have experiences when turned
off, or just stored on a CD, in your proposal.
> > I doubt that. Remember, in these artificial neural nets, brain
> > size is already present (easy to make em' big). All that remains is for
> > them to start exploiting the niche of intelligence, and since it would
> > have advantages, it should be a fairly steady and stable progression,
> > with plenty of time to let it happen.
>
> I'm not convinced that if you add in the code for evolving the neural nets
> (what are you going to replace DNA, proteins, etc. with?) and simulating
> the external environment (what environment are you going to use?), and
> interfacing the neural nets with the external environment, that it's going
> to be shorter than the program form I proposed. But I agree that without a
> more authoratative estimate this discussion probably won't get very far.
You obviously don't need to replace proteins and DNA, just the
neural net.
So where are we going to get that authoritative estimate? I don't
really know any computer scientists, though NYU has them of course. It
would have to be someone interested, but not so interested as to take sides.
> > That doesn't make much sense to me. If there's something that
> > it's like to be a string, why should that be ruined just because it's
> > surrounded by other strings? There are no external observers here, just
> > internal observers. Is your idea that the concious part of the tape must
> > be printed out at the beginning of the tape?
>
> Why should it be ruined if it's interspersed with a 0 every other bit? Why
> should it be ruined if it is intermingled with this other string in this
> particular way? But notice that in all three cases you're introducing new
> information that is not present in the original string, namely the
> algorithm for extracting the substring from the original string. Clearly a
> line has to be drawn somewhere and the natural place to draw it is to
> disallow any extra information. Thus a substring cannot be conscious
> because it is not a self-contained carrier of information, and it can only
> be made into one by the introduction of new information.
Again, I'm not seeing it the same way as you. You still have not
answered my question: are you proposing that the program must print out
the conscious part of the string at the beginning of the tape, then erase
the rest of the program? Otherwise it is still a substring. What about
the fact that it is surrounded by other bits in the orthogonal direction
from the other Turing machines?
If a computation was interspersed with 0 at every other bit, it
would still be implemented. This is exactly the sort of question the
implementation criterion will solve. In my proposal, it would still be
implemented because the shortest program that could label the bits used
would still label them in the same order.
I can see why it's more problematic with just strings, since you
no longer have the requirement of causal relationships which greatly
constrains what mappings can be considered for computations.
> > > That is not the model I have in mind. It doesn't seem sensible to think of
> > > a universe with a continuous infinity of Turing machines.
> >
> > Why not? The set of all possible programs - what could be more
> > natural than that? It seems closest to Max's 'everything' idea as well.
>
> Sorry, I was wrong. It's not insensible to think of a universe with a
> continuous infinity of Turing machines. All you need is a three
> dimensional Turing machine and a universe with four spacial dimensions.
That's a mysterious remark. All I need is what I suggested, which
if you like amounts to a universe with one continuous spactial dimension,
one discrete spactial dimension, and one discrete time dimension.
(Assuming the heads' internal variables are not counted.)
Regardless, the point remains that it's more natural to start out
with all possible programs than it is to 'clone' the machines and add
extra tape at each time step.
- - - - - - -
Jacques Mallah (jqm1584.domain.name.hidden)
Graduate Student / 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 Fri Jan 29 1999 - 12:13:58 PST