originally posted in:Secular Sevens
On the topic of the digital consciousness thing, I don't think our current silicone processing technology can handle human complexity. Transistors rely on positive/null binary programming, which works great for calculations but lacks the complexity of context. Neurons are not positive/null binary systems. They activate through a bunch of different means, cascading signals through one another. There are null/positive integers there, but its waaay more complex than just those logic gates. Its like asking a person which way to go to the market, expecting a left or right direction to be given, but told north by northwest instead. Until we can make some sort of computing technology which can duplicate that at least 4 way system of logic gates in contrast to our current binary systems, we won't ever be able to ever start to duplicate neuronal data in any meaningful terms. In order to gain contextual perspective, something more than the binary system will have to be derived.
English
-
Edited by A3LeggedBurrito: 3/1/2014 5:12:03 PMIt is apparently possible to model human brain activity with modern computing capabilities in a technical sense, but it is so difficult to do so it might as well be considered impossible (and this article shows just how difficult it is). And when you add to that the necessity for the computer to be relatively small, that is being able to fit inside some sort of (probably) bipedal body, it is clear that we would need to discover a new type of computing. However, I think we could definitely do so by the time this issue becomes relevant.
-
Interesting. Thanks for the link. Though I will have to contend what its doing is merely modeling the structure, using binary. ITs not actually modeling the processes. Transistors lacking the multitude of ways to int data will simply not be capable of contextual calculations. And their assumption that the exascale computers will be able to simulate human thought is...well absurd. It probably will finally match the brains in as afar as processing capacity, but lacking the proper programming language it won't actually work like a brain. Its a step in the right direction, and a very necessary one. I'm sure it will be a part of whatever tool set is developed towards that goal, and I hope thy get such a computer up and running as the first step by 2018.
-
Could quantum computers potentially fulfill that role?
-
Wonderful question. We don't know, but the answer is probably no. Quantum computers are not actually any faster than traditional transistors at making calculations, its just a point in using the analysis that you can use a trick with entanglement to preclude certain possibilities, and do certain KINDS of calculations much much faster. As far as we can tell, quantum computers wont have the versatility to be used in such a way. They won't replace traditional computer. IT will just be another tool in the toolbox, it probably won't really break any new ground or redefine any technology. Not by itself at least. Interestingly though, I suspect we will need some sort of chemical computing just to accurately analyze quantum computing data. Our brains are already chemical computers, so understanding our brains on a mechanical level, and designing similar things to be used as data processing in decoding the data for quantum computers, we would then have precisely the technology necessary to make a complex digital persona who will understand context. I personally think that since Moore's Law is on a hard deadline, in that it will no longer be applicable in about 10 years, we had damn sure better figure out how chemical computers might work. If we can use chemical computers for processing quantum data and store it on transistor based hard drives, we could have the processing power easily capable of simulating entire universes, much less a human brain. That is no small feat.