Technical Difficulties from on Top of the Mountain
2015-02-08
The power in thinking about hard problems.
I am an engineer by trade, not a scientist, because I have a certain amount of impatience with thinking for thinking's sake. But as I have spent great amounts of time solving simple problems, I have more and more appreciation for big thinking. It can be a great place to go steal ideas to use tomorrow.
One big thinker was Feynman. He made many contributions to Physics, finding ways to solve problems that were intractable with traditional tools, but he was also curious about a great many things and was able to predict the future just by wondering about what would happen if you took things to their extremes. He had predicted molecular machines (or MEMS) by just thinking about a serious of 10:1 reduction levers that just kept getting smaller and smaller. But it turned out he also had been pondering the future of computing and its intersection with physics way back when.
The thread for this started from watching the quantum computing talk (LIQUi) from LangNext 2014: (channel9)
There was a reference to a list of publications at the end of the talk with a link: arXiv.org Search Results for Dave Wecker, and skimming through the paper 'Improving Quantum Algorithms for Quantum Chemistry', I noticed that the first paper was by Feynman: "Simulating physics with computers". This actually was a keynote talk at a conference in 1981, but back not everything got recorded for uploading to youtube, so you'll have to be satisfied with a
scan of the original paper (or you can pay $40 to Springer Publishing to get a copy of the transcript for which they paid nothing.)
Basically, almost 40 years ago, Feynman did the thought experiment about whether you could simulate quantum physics--and to be complete he considered both classical discrete computers as well as then non-existent quantum computers. Short answer? Classical computers would never be able to tackle big enough problems in a scalable way, but if engineers ever figured out how to build usable quantum computers for the physicists to use... well, we might just be in for more interesting times.
I am an engineer by trade, not a scientist, because I have a certain amount of impatience with thinking for thinking's sake. But as I have spent great amounts of time solving simple problems, I have more and more appreciation for big thinking. It can be a great place to go steal ideas to use tomorrow.
One big thinker was Feynman. He made many contributions to Physics, finding ways to solve problems that were intractable with traditional tools, but he was also curious about a great many things and was able to predict the future just by wondering about what would happen if you took things to their extremes. He had predicted molecular machines (or MEMS) by just thinking about a serious of 10:1 reduction levers that just kept getting smaller and smaller. But it turned out he also had been pondering the future of computing and its intersection with physics way back when.
The thread for this started from watching the quantum computing talk (LIQUi) from LangNext 2014: (channel9)
There was a reference to a list of publications at the end of the talk with a link: arXiv.org Search Results for Dave Wecker, and skimming through the paper 'Improving Quantum Algorithms for Quantum Chemistry', I noticed that the first paper was by Feynman: "Simulating physics with computers". This actually was a keynote talk at a conference in 1981, but back not everything got recorded for uploading to youtube, so you'll have to be satisfied with a
scan of the original paper (or you can pay $40 to Springer Publishing to get a copy of the transcript for which they paid nothing.)
Basically, almost 40 years ago, Feynman did the thought experiment about whether you could simulate quantum physics--and to be complete he considered both classical discrete computers as well as then non-existent quantum computers. Short answer? Classical computers would never be able to tackle big enough problems in a scalable way, but if engineers ever figured out how to build usable quantum computers for the physicists to use... well, we might just be in for more interesting times.
