tag:blogger.com,1999:blog-5304409.post4785216114466305649..comments2024-03-17T16:17:20.145-07:00Comments on Amit's Thoughts: Future anti-predictions 2Amithttp://www.blogger.com/profile/12159325271882018300noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-5304409.post-86485404248869213772009-06-29T14:37:54.895-07:002009-06-29T14:37:54.895-07:00Thanks for the comments xaminmo! I think quantum ...Thanks for the comments xaminmo! I think quantum computers will be <i>useful</i> but conventional computers will be useful for so many more problems that they'll get faster even quicker than quantum computers. General purpose seems to win over special purpose in most areas because of network effects. For example GPUs are being more and more general purpose and can take more problems that would've been solved by FPGAs.<br /><br />Replication style teleportation requires us to measure and reproduce the state of all the particles. Unfortunately quantum mechanics limits our ability to measure everything. This is why the entanglement approach is interesting — it manages teleportation without having to measure everything.Amithttps://www.blogger.com/profile/12159325271882018300noreply@blogger.comtag:blogger.com,1999:blog-5304409.post-30951411191638769692009-06-28T22:19:48.396-07:002009-06-28T22:19:48.396-07:001) I think the quantum computing will be somewhere...1) I think the quantum computing will be somewhere in the middle.. Basically, parallel multi-qubit processors. I see them being useful for pattern recognition and encryption, but not for "traditional" processing.<br /><br />2) DNA computing is useful only for such a tiny fraction of workloads. If you have a hugely parallel workload that would take weeks to run, then DNA computing is worthwile. You'll spend a day "programming" it (sequencing the components), a day "processing" (combining and replicating), and a day getting the output "selecting the "correct" answers out of the soup.<br /><br />3) I think teleportation with entangled particles is a pipe dream. A few bits per particle, and you still have to ship the particles? It would only be possible if you could reentangle remotely.<br /><br />I think it's more likely that replication style teleportation would be possible. With great amounts of energy, it's possible to create matter from energy. So, you transmit a pattern (we're what, about 15PB of data?) and parallelize the data into a fabricator. Once you've assembled the replacement, and verified that it's indeed correct, you would disassemble the source. For a person, this would mean that you copy them, then kill the original.<br /><br />This would take a ridiculous amount of energy. To store and utilize that energy compactly enough, you'd have to be dealing with matter/antimatter reactions. Antimater manufacture and containment also takes ridiculous amounts of energy. This would be best relegated to a solar or nuclear platform in space.<br /><br />4) Wormholes, I think, would not be stable. If they were stable and naturally occurring, either we would suffer their consequences already, or they avoid matter, or they would somehow not be readily usable by us.<br /><br />Even so, I think bending space is less likely than punching a programmed hole through a dimension in which our two desired points in our spacetime were already proximal. We might be able to create tunnels through which we could pass data, but probably not whole, solid objects, without the use of something powerful such as black holes. Any that we could keep, control, or maintain would be too small (microsecond life span with an event horizon big enough for an electron? How much space could you warp with this? Maybe an array of them could be used?)Anonymousnoreply@blogger.com