I liked the book and thought he did a good job presenting the paradigm. I felt disappointed with the number of solid examples or concrete tests (what kind of experiments or observations would validate this paradigm over others?), but I should probably go back and look up the references he points to.

Did you see Jeff Hawkins' little speech he gave at the Beyond Belief conference? This guy made some pretty outrageous claims. He said that he knows how the brain works. From the language he used, it sounded like he meant the whole damn thing. (I only watched it once, perhaps I'm mistaken.) Experience tells me that people who are as excited about an academic study as much as this guy usually turn out to be wrong on a few key issues that they overlooked in their enthusiasm. But part of me hopes he's right. Or at least on the right track. Either way, I'm probably going to have to read this book now.

I've been doing some introductory reading on Neuroscience and Philosophy (specifically, books about neuroscience written by philosophers, and not the other way around). It's a difficult subject to jump into, even when you're in the shallow end. Neuroscience seems to be one of those fields that is completely irreducible to anything resembling layman's terms.

What exactly do you mean by non-linearity of neurons?

What I'm referring to as the non-linearity of neurons involves a couple characteristics. One is that if one neuron begins firing, others that it has connections to will have a greater probability of starting to fire, and the original signal can be passed to neighbors without the initial neuron "losing energy". The passing of the original signal to different neighbors can be said to be statistically independent. The global features of non-linearity start to enter the picture when you consider that the number of neighbors that are triggered affects the chances that the original neuron will have of being triggered again after it saturates and recharges. You can imagine how this feedback could produce semi-static states, semi-static features that propagate, and other interesting behaviors. These are characteristic of non-linear dynamical systems like a glob of oobleck (cornstarch and water suspension). If you haven't seen this stuff, check out some videos online to see the creation of life from dirt. And I highly recommend trying it for yourself with a speaker and some friends. My friends Rajiv, Scarby, Briggs, and I accidentally stayed up all night messing with the stuff on a 12" subwoofer hooked up to an amp and a laptop for generating signals.

Another thing, I am not a neuroscientist. I took a Computational Neuroscience class in which I learned how individual neurons behave and can be modeled and how they interact with others, although this is where the model starts to get significantly more complicated and probably slightly incomplete or inaccurate. It's unknown whether any existing model of neuronal behavior and interactions accurately approximates what we have observed in the brain on a large scale, much less if it possesses all the dynamics to contain the information of "what's going on up there". The field is certainly getting closer at an increasing rate, but there are, as you know, still many gaps.

The brain is such a complex thing with shit going on at so many different scales that it almost certainly will always make sense for us to be able to describe it on several different levels. Right now, our top-to-bottom, bottom-to-top paths are too bumpy (and then there's all those people drinking and driving) for us to be satisfied, and this is the case with in areas of science. Neuroscience, cognitive science, psychology made sense awhile ago, but now there's a bunch of divisions "in between" those that make sense and qualify as separate within the field. Maybe we'll be satisfied with our patchwork eventually, but I like to think that we will just repeatedly discover that everything is infinitely complex.

I thoroughly enjoyed the book, On Intelligence by Jeff Hawkins, and I think you would be interested.

we need to start a decision-race a la the space race

I'm almost certain I want to do that, but what is it?

Some more undecidable things!

5. some Collatz (3n+1)-style conjectures
6. whether there's a set larger than the integers but smaller than the reals (i.e., there is a surjection from it to the integers, but no injection, and there is an injection from it to the reals, but no surjection).

I guess this sixth one you wouldn't say is Gödel-style, though.

And here's one thing I can't decide if is decidable or not (without Choice):

7. whether there are two subsets of the reals each smaller than the reals (in the sense of cardinality, as above) whose union is the reals

Figuring I'm wrong about this, could you explain why the rationals and the irrationals don't decide question 7? are the irrationals smaller than the reals?

Not smaller in the sense of cardinality, as in question 6. I'll define a function f(x) whose domain is a subset of the irrationals. Let it map all of the irrationals in the open interval (-1,0) to their negatives. Then let it take all multiples of sqrt(2) which are multiples by a rational number from 0 to 1 exclusive and divide them each by sqrt(2). This clause will hit all the rational numbers in (0,1), and we have already hit all the irrational numbers in (0,1). f's domain is a subset of the irrationals and its image is the whole interval (0,1).

And there is a function g(x)=tan(pi(x-1/2)) from (0,1) to all the reals.

Then h(x)=g(f(x)) is a function whose domain is a subset of the irrationals and image is all the reals.

i think it has something to do with being the decider

bryan, i think you're right.

Undecidable decidability...let us bow our heads.

condoms?

Yes. I said to him "I hope they aren't lubricated."

In response, I got

"They are used.

j/k"

Dude.. let's hang out at your dad's place. Seriously.

Ribbed for her pleasure.

I WIN!

I could survive for 1 minute, 3 seconds chained to a bunk bed with a velociraptor

Created by Bunk Beds.net