3 ms·
I lean towards a sort of death. But perhaps we're all dying and reanimating in between each Planck time instant anyway, in which case we shouldn't worry about i
by Wintamute 13y ago
I lean towards a sort of death. But perhaps we're all dying and reanimating in between each Planck time instant anyway, in which case we shouldn't worry about it too much :)
- goldfeld 13y agoThat quite defines how I think about it. The current "me" is as different from the me within a second from now as from a theoretical me at the other end of a teleportation device that delivers me there within a second. The three selves are all distinct from each other and in fact are as distinct (and as close) to each other as from any other human being, with the caveat that the 'me's a second from now inherit all the memories and physical state from the current me, whereas a different person does not. But in terms of selves, of consciousness, they're all isolated phenonema. I think consciousness is merely an illusion, we think we have continuity because we inherit the memories and thought processes from the self from the moment before, but all that exists is a moment of awareness (the "now"), which is analogous to a clock of a computer processor.
- Zikes 13y agoDepending on granularity, your comment was written by a number of people approaching infinity?
- goldfeld 13y agoYes. And I really do think it approaches infinity, or else our wetware has a limited clocking speed and that would define a single atomic moment of awareness (seems more likely since we're highly advanced computers, but computers nonetheless.) Even as I was writing the comment several clocks might have been spent not on it, but picking up ambient sound and other stimulae. To the extent that I'm half conscious to these other things, I was actually fully conscious to them though for only a few number of clocks, whereas things on which I seem to be fully conscious are demanding the majority of my (multi)processor clocks. There's the case for whether we can truly multitask--are we multi-core? We definitely are as to the many functions done subconsciously, but what about the conscious regions of the brain? But to simplify the thought experiment, we can simply think that we can only really spend any single moment of awareness on a single atomic thing, and like a computer we juggle attention between various things so fast as to make it seem we're doing them all concurrently (like the various applications simultaneously running on a computer.) The upside, of course, is that my philosophy frees me up for gultfree teletransportation. Bring it on!
- nemo1618 13y agoFun fact: there are 18.5 million billion trillion quadrillion Planck times per second. So you can think of the universe as a simulation that runs at 1.85*10^43 FPS, where each "frame" sees all light-speed particles move one Planck length in their direction of travel.
- deadairspace 13y agoIs there any actual justification for viewing spacetime as being discrete, with unit 1 Planck length/time? As far as I'm aware, the Planck length is just a length you get when you multiply a load of constants together.
- jules 13y agoYou're absolutely right. The plank time is just a time such that events that happen in sufficiently smaller time intervals probably need new physics to describe. This is comparable to how very fast moving objects needed more than Newtonian physics to describe, namely relativity, or how very small objects needed quantum mechanics. It may be that that new physics is discrete, but at this point we have no idea.
- pbhjpbhj 13y ago>a time such that events that happen in sufficiently smaller time intervals probably need new physics to describe // Physically¹ events happen instantaneously we just can't tell at which precise instant they happen. This is true even with quantised time. - - - ¹ Perhaps that should say "in the mathematical model of the physical universe we base our science on".