On naturalism and consciousness

[The Grand Nudger]

@Rhythm

To long to respond in kind, so let me make a couple brief observations.

1) There are a lot of value statements: robust, useful, reliable, complexity, and so on. But the universe is robust and complex, usefulness is an imposition by an existing consciousness (us) as is reliable. But I think the simplest definition of a gate is any mechanism which has inputs, and which is capable of subsequently triggering other mechanisms in a deterministic way.

More like technical statements that can have a variety of meanings outside the setting, sure. A simple gate being a benchmark. A gate with redundant inputs, extra outputs, steady states (memory) would be "more robust". Their usefulness only describes what sorts of logical operations they can perform - and whether or not they have to opportunity to do so. Reliability is the insistence that the mechanism be able to abstract -logical operations- If we can't rely on the gate producing the proper output it's not a useful gate - or it is a "less" useful gate.

But yes, the universe is all of those things - but we may be talking about different "things".

If something must seem robust, useful and reliable TO US

Not to us, in reference to -any- computational system. It must be robust and useful to perform functions, it must be reliable because computation is a process. You can't give the sum of 1 + 1 in an absolute sense if either variable changes from pulse to pulse. That's where most things fall as gates. They might have the structure, but they don't have the function (for a variety of reasons).

--It doesn't, or it's not knowable (and therefore not robust, useful, etc.) to us?

Not directly useful to architectures that rely on deterministic operations. Think of it like a compatibility issue between systems. The architecture is different. You're jumping ahead a bit. I think that all of these things have to be present to describe -any computer - any computational system that would be fully accounted for.

Maybe you should describe exactly (or again?) what you mean by a gate. I take it to mean only that a given input should produce the same output every time. To me, this includes simple logic switches (OR, XOR, etc.), but also could include mappings: 0--> 0, 1-->10, 2-->3, etc., something perfectly possible with a simple ANN.

In electronics, a logic gate is an idealized or physical device implementing a Boolean function;
http://en.wikipedia.org/wiki/Logic_gate

What you're talking about is a specific -function- that we could implement -with- a gate, btw.

But given a deterministic universe, any given system must necessarily produce the same output every time it has the same input.

No, it "mustn't". A program counter receives the same input pulse (the output of a clock - for synchronization) over and over, but works precisely -because- different outputs can flow from that interaction. -Architecture- "How can I get multiple disparate outputs from a single input?" - wouldn't that be a powerful little mechanism eh? Isn't it fortuitous that we were just talking about comparators? In a computational system, you get the output you define with whatever system you've created. The system only recalls the same output when you make that request. You can specifically engineer a system to give a different output with any number of repeated identical inputs. The gates (in one model) need to exhibit this behavior, but not systems comprised of gates.

But still, I'd argue that EVERY subdivision of matter must be a gate in the sense I just described,

The principles of gate construction are the principles of abstracting logic - so, gates would be possible at any level or subdivision in which logic was possible - in theory. Wherever logic gets dicey, different architecture (from the type I'm always droning on about) would be required/possible. Wherever we're just interested in effect rather than process - my comments would also lack substance (or rather, I'd offer you a different line.. - I'm not as familiar with NN or QM computing- it was after my time, lol).

and that the choice of specific systems is still a matter of our arbitrary conceptualizations, rather than a description of any existential reality.

Its the choice of effects, not systems. You "experience" a mind, you assume that I do as well (and it;s a bit more than "just" an assumption). After all, you can watch me do it, I appear to be a thinking thing, no? I do all of this "stuff" that seems like mind simply must be behind it. When we compare what sorts of systems we might expect to find "mind" in, we naturally favor effects such as those we observe in others, or experience ourselves ( I think, in part, because that would be particularly compelling evidence to us, eh?). We don't have to (and shouldn't), I've raised that point many a time. But, I'll say this, -if- we both have mind (the beach and myself) then we're going to need a different word for whatever subdivision of mind I must have - because it's clearly not the same as any mind that beach sand might have, eh? Any theory that describes beach sand mind doesn't seem to describe human minds - we just don't see similar effects, we have no reason to even make the assumption. I think we'll need another theory.