Log in

Previous Entry | Next Entry

The first rule of physics club...

I'm starting to think there may only be two laws of physics:

1.) Everything which is not forbidden, is compulsory.
2.) Nothing is forbidden.

Why make things any more complicated?


( 6 comments — Leave a comment )
Nov. 15th, 2004 09:00 am (UTC)
How do you mean, compulsory?

(nice icon btw, I always love your icons)


Oh, well its always nice to hear that nothing is forbidden;)
Nov. 15th, 2004 05:14 pm (UTC)
The first rule is not mine, it's called the "totalitarian principle". I searched for it on Wikipedia, but there was no entry for it, so I created one just now. Hopefully this is all correct; to the best of my knowledge this is what it means.

Heh... my first Wikipedia entry... I didn't realize it was so easy to make one. I should probably get an account now.

As for the second rule, well that's just my middle-of-the-night insomnia telling me to post wild speculations about what the future of physics holds. I say "nothing is forbidden" would be a good end to the quest for physical laws since it seems every time we discover a conservation law, it ends up being broken later on. So perhaps we will eventually reduce them to a single law, or a delightful anarchy of no-laws. Of course, not all processes are equally likely... things can be so heavily supressed that they are effectively forbidden. Which is necessary to avoid complete chaos, a mininmal requirement for sentient beings to have developed. I don't know how likely my "second law" is but if it turns out to be true, then our little corner of the multiverse would be special in that it happens to have very orderly things going on which has been conducive to evolving life.
Nov. 15th, 2004 05:49 pm (UTC)
Law of accelerating returns
I'm still pulling for Kurzweil's "Law of accelerating returns" to hold. Put simply, it says things always get more interesting, and that this increases the rate at which things continue to get more interesting.

I've never understood much about the many worlds view. Seems fine as long as your inside the superposition, but I'm incredulous towards all this stuff about events with negligible probability actually occurring.
Nov. 15th, 2004 05:57 pm (UTC)
Re: Law of accelerating returns
Nov. 16th, 2004 08:51 am (UTC)
Re: Law of accelerating returns
You've been pushing this Kurzweilai guy pretty hard lately. I am all for accelerating change, and I think it is the natural way for technology to go in the long run, but I can't stand when people who use the word "singularity" to describe what they think is going to happen. Exponential growth is natural, not hyperbolic growth. Exponential growth never produces a singularity, it keeps growing exponentially forever. I think by throwing that term around he is making transhumanists look stupid. (And to be fair, he's not the only one doing it, it's become fairly widespread.) Which is not good for our case.

The other thing that bothers me about his graphs (at least one or two of which are clearly linear) is that he's chosen all the steepest graphs he can find and ignored ones that show boom-bust cycles (which are also a natural phenomenon, although they act on a more short-term scale), or periods of stagnation. The truth is, there are many things in nature which oscillate, there are many which attenuate (decay), and there are also many which accelerate. It's no mystery why, it all depends on whether you have positive or negative feedback. Systems with positive feedback are always going to be exponential, unless they pass into another phase where there is a damping factor or another regime of behavior. There isn't anything more fundamental about exponential growth than there is about exponential decay. It all depends on what you're talking about. So yes, it applies to evolution as well as technology, but no it is not more fundamental... and no, there will not be a singularity, and (perhaps this is too obvious to point out, but...) there is no such thing as infinite wealth or infinite techonology.
Nov. 16th, 2004 11:23 am (UTC)
Re: Law of accelerating returns
I like Kurzweil because I like the long term arguments for an exponential trend.. not the moore's law stuff. Yes, singularity is a stupid word. As you say, its not meaningful within Kurzweil's own moving frame of reference its only meaningful from a fixed point in the past with fixed perceptions. Anyway, we live in the singularity from the point of view of people even 100 years ago.. in the sence that we do not understand the vast majority of scientific developments around us.

As for the boom bust cycle, I think your missing Kurzweil's real mistake here: he ignores discontinuities. The rate of evolution probably does grow in a generally exponential trend, but there are massive discontinuities, both upwards and horizontal (or even downwards). These discontinuities are the results of the importance of random events being magnified, and I doubt any amount of evolution can outgrow the importance of random events. This is why all the other stuff I've sent you is worried about avoiding horizontal discontinuities.

I don't think "boom and bust" is all that relevant really. Its probably easier to think in terms of diversification (when evolution "does not really know what to do" and everything is managing to sruvive just fine) and extinction (when one thing breaks ahead and whipes everything else out).

Of course, Kurzweil other major flaw is thinking that life will be all happy. You need conflict and "death" at the most important levels of evolution. We don't cry when a scientific theory dies due to lack of evidence, but as the theories increase in complexity, "humanitarian concerns" may one day apply. And in the short run, I'd expect that people will become far more disposable, as the precentage of people being trained to think increases. I mean who would bother with civil liberties unless thinking people were hard to replace. :)

Anyway, we could very well see a leveling off of many aspects of technological progress well before the additional exponential curve brought about by AI kicks in. It depends on a lot of factors, like just how complex the algorithms governing our brains are, and just how able we are to use machines as part, but not all, of our thinking. I just really have no way of knowing. I do however think that older human endevers, such as philosophy, can be brought up to speed.

I should probably read some other more moderate transhumanists.
( 6 comments — Leave a comment )


domino plural

Latest Month

May 2017


Powered by LiveJournal.com
Designed by Lizzy Enger