Monday, August 20, 2007

My non-science post

I was going to post about not-science. It was going to be novel. However, I started writing the post and then I started reading the post and it was rambly and discontinuous and not well argued or thought out. So I'm putting it on the back-burner, and maybe I'll post it when I've thought about that stuff some more. But I'm pretty pre-occupied with my biology stuff now, so maybe I'll just keep talking about that.

So. This New York Times article is neither exciting nor news, but it is interesting. Better than this New York Times article which is just plain silly (and which we discussed last night over pizza and beer, hah).

Article #1: Americans have to drastically rethink two things: the label "Dumb Jock" and the idea that your brain stops developing at an early age. It turns out that in mice and in humans, excersize stimulates the hippocampus and prompts the development and maturation of new neurons. Aerobic excersize is better than stretching in the elderly. So take a jog, and you'll be able to do that problem set better!

Other things that supposedly increase bloodflow to the hippocampus and prompt the regeneration of neurons are: moderate alcohol use (although too much and you're hurting your brain), marijuana, a social life, and chocolate. Things that hurt your brain are stress and fast food.

Article #2: If you accept the "reasonable" hypothesis that some civilization reaches a point when they can create computers capable of modelling thousands of brains, then it is likely we are all simulated. There are several problems with this idea, and more problems with it being in the science section of the New York Times. Really, it is dubious that the assumption that thousands (or really millions) of brains will ever be able to be well modelled in a computer simulation. Especially not well enough to lead to the complexity and diversity we see in nature. At the moment, we have trouble modelling three electrons on the largest supercomputers. AI is incredibly basic and does not come anywhere near an actual sentience or even a facsimile of sentience. We can program supercomputers to play chess and solve checkers, but in the end we cannot make a computer that can have a normal conversation. Imagine not just the processing power but the skill in computer science necessary to overcome that barrier and model not just one being, but millions.

It's funny that one of the explanations given as to why this was a reasonable assumption to make began with the phrase "assuming a planet-sized computer" which is another assumption that I hope is self-evidently absurd. People can imagine planet-sized computers, but imagine trying to build one. And that ignores all the problems of finding a place to put such a computer where it would stably remain and be easily accessable but wouldn't interfere with life on earth. Basically, to do that it seems we would need faster-than-light-speed data transfer, unless it wasn't really a planet-sized computer but rather a moon-sized computer, and even then it would be incredibly unweildy to make and position and so on. Again, not only do we not have the technology, but we are nowhere close to having the technology.

We really will need to reach the technological singularity before any of this will happen. And given that the human race has a maximum of 7.8 million more years to live, and 1794 more years to travel in space, (taken from this, another nice pseudo-scientific article by the same dude) well, we'd better hop to.

This is all ignoring the fact that, well, presuming that some society did make it to a point where they could model the world, they wouldn't have come to similar conclusions as ourselves and realized that modelling sentient beings is, well, kind of sadistic. And if the question comes down to "Are we sentient or do we just think we're sentient?" well, then I ask you why you are asking such a silly question. Because in the end that is a bit meaningless.

Besides, the only conclusion you can really come to from the idea that maybe we're all just simulated in a computer is the fact that, okay, maybe we're all just simulated in a computer, but that doesn't have any repercussion on my every day life, what's right or wrong or in between, and so forth.


ayn said...

If I had a planet sized computer, I would be using it while sitting on the couch next to you. But I just have my laptop, so I am sitting on the couch next to you using it, instead.

Duff said...

Ugh. The simulation idea is just silly. It seems like a bad and pointless pop techno-goth rendition of the old theological move that we are just ideas in God's head.

Duff said...

Also, the technological singularity idea seems to me to attract mostly people who wish they were scientists (or their idea of scientist, which is far less glamorous than the movies), but aren't. It is also a poorly concieved idea. Does a bicycle count as trans-human augmentation? I can under my own motor power travel farther in a day than any human possibly could 200 years ago under their own motor power. But I am not sure that means much, particularly about my ultimate metaphysical humanity. By programming a computer, I can solve physics problems or math problems they would have spent centuries solving using their available means. Does that make me transhuman? In the end, I am not sure how much faster information relay between man and computer can be achieved thru means other than our visual cortex (the most intense information processing unit in our brain). So how will wires in my scalp do anything?

The technological singularity people always get under my skin. They rarely understand people or science that well. It is just bad sci-fi.

Elizabeth said...

About the technological singularity: You totally have a point. I hadn't really thought of it that way. It's basically a fancy way of saying "SUPER FUTURISTIC TECHNOLOGYXORS" I think. But it's possibly better (or at least more honest) than the simulation idea, which pretends to be new and scientific when really it's, as you said, old theology.