Steve Jurvetson, on Rose’s Law for quantum computers

If you are a nerd and you're not following Steve Jurvetson on Flickr, you should correct that. Why? Posts like this one, in which the VC and tech-thinker explores interesting things in interesting ways. "Barring a fracture of physics, we may be able to build quantum computers more powerful than the entire universe within 3 years. They harness the refractive echoes of many trillions of parallel universes to perform a computation, unlike anything we have seen before." Check out the full post, with annotations and more thoughts.

• [Boundegar]

  That’s interesting, but I don’t think there is any evidence for the existence of parallel universes.  Is it possible he’s more VC than scientist?

  • [The Dour Salmon]

    I don’t think he’s talking about real, actual parallel universes. It’s a metaphor.

    • [Gary Fredericks]

      Either is plausible, depending on his interpretation of quantum mechanics (see MWI).

      • [Boundegar]

        Interpret all you want.  It’s not evidence.

        • [Mantissa128]

          The Elitzur-Vaidman quantum bomb tester.

          It measures something that happens in a counterfactual world. You’ll have a hard time convincing me that this counterfactual world isn’t real, since it has real, measurable consequences. In other words, if that world weren’t real, the quantum result wouldn’t exist.

  • [redesigned]

    There is a ton of evidence for parallel universes in the math, and in quantum interactions … there is even macro evidence, we just left it in the other universe by accident.  oops.

  • [Martijn]

    Mining other universes for their computing power sounds like an extreme form of colonialism. Though I’m sure there’s a nice SF story in there.

• [Stefan Jones]

   Theoretically faster. Once you load Windows 8 on it . . .

• [jimbeaux]

  They’ll know it’s working properly – if the answer is 42

  • [Steve Jurvetson]

    ….42 hmmm…. 4 bases == 2 bits.
    A,T,G,C == 0,1 
    The secret to life, the universe and everything is the digital code. It’s the question that drives us.

    • [YamaraTheGod]

      . . . Tricky.

• [Jacob Henner]

  I don’t exactly unserstand how something can be “faster than the universe” – does the universe have a speed?  anyway, i’ll go read the article to find out

  • [ldobe]

    I think the idea is along the lines of “harnessing the power of sunspots to generate cognitive radiation”

    The whole definition of universe means anything outside it is inaccessible from within it, since the universe contains everything.

    Parallel universes sounds stupid to me in the same way religious belief sounds stupid. There’s no evidence they exist, and whether or not they do exist is untestable anyway. And if it were testible, they wouldn’t be separate and parallel so much as, connected up in some way.

    • [Gary Fredericks]

      It’s not really a religious issue. Quantum mechanics involves things that look a lot like parallel universes, and some people choose to assume that means that’s what’s actually going on.

      • [ldobe]

        I understand your argument, but I’m not sure how parallel universes would solve any problems or be a valid way to explain quantum computation.

        My informal studies of QM don’t seem to indicate that the many worlds interpretation of QM would in any way allow for interactions between forked universes. But it might be allowed if we ever get a workable theory of quantum gravity. I’ll concede that I’m not trained in QM formally, but occam’s razor would suggest that the VC is blowing smoke and using buzzwords, rather than physics finding a way to interact with trillions of parallel universes in only a few years.

        It just seems deceptive to say a bunch of parallel universes when all that’s being talked about is the mathematical model of the many worlds interpretation.

        • [redesigned]

          parallel universes != multiple worlds.  in m-string theory most the parallel universes are tightly curled in on themselves, bubble universes such as our own are the exception not the rule.  at least that is my megar understanding.

        • [ldobe]

          Ah I see.  I was thinking four dimensional interpretations of reality, while it seems everyone else is thinking in 11 odd dimensions or possibly more.

          I’m pretty skeptical of String Theory.  I think Loop Quantum Gravity is a better research direction, since it makes testable predictions.  String theory has had a long time to develop, and still doesn’t have any predictive power, nor is it feasibly testable.  The String Theories we understand, we know are wrong, and there are 10^500 or so others that haven’t been characterized.  The landscape and anthropic principle are no better than creationism in their logical consistency, and don’t belong in science so much as in a class on mythology.

          I know, let the hate mail begin.

        • [redesigned]

          @Idobe:disqus you’d probably enjoy the book flatland.  a classic and a fun story about dimensions and perception.

          my understanding about multiple universes is different from what people would quantify as the classic dimensions, our universe has 4 dimensions for certain and probably around 11.  The only relation between dimensions and parallel universes is you’d have to travel in a dimension outside the classic 4 to reach a parallel universe.  people often interchange the 2 words in sci-fi when they don’t mean the same thing at all.

          …but to be honest the math of multiple universes and most the math used in quantum computing is at my upper limits or beyond what i’m currently able of wrapping my noodle around, so i don’t really have any answers, just some sorta understandings.  others may be able to better answer your questions.  cheers. :-)

        • [Colin Rosenthal]

          “My informal studies of QM don’t seem to indicate that the many worlds interpretation of QM would in any way allow for interactions between forked universes.”

          There is, It’s called “interference”.

        • [ldobe]

          Yep.  Interference.  Duh.  I remember that now, but if you don’t think about things specifically and often enough, you tend to forget.  I need to re-read some textbooks. 

      • [gibbon1]

        My rough understanding is it only looks like parallel universes if your mathematical model isn’t taking into to account the interactions between macroscopic collections of particles.

        Simple model, the a playing card stood on end will fall in both directions. Complex model, a chain reaction will cause the card to fall one way or another.

    • [Nathan Hornby]

      From what I gather the existence of parallel universes is ‘proven’ by association – i.e. if they didn’t exist then a whole bunch of maths and physics falls over.  Like with the Higgs Boson.

      Religion, on the other hand, is just made up – nothing relies on it to be true, other than faith.I’m a total arm-chair scientist though so trust me at your own risk.

  • [Steve Jurvetson]

    I figured that one needed a definition, so i gave one in the post

  • [cfuse]

    In short, probably several.

    Speed is just a measure of movement over time. The upper limit for movement is the speed of light, and I couldn’t tell you what the upper limit of time is (only that time itself is a variable force, in much the same way gravity is).

  • [Boundegar]

    So we’ve solved that thing about calculating the position and momentum of every particle in the universe, and can do it faster than the actual universe.  The future is therefore known completely.  The rest you can now quit, the game is over.

• [technogeekagain]

  “Faster than the universe” — No. Consider that the universe *IS* a quantum computation.

  And I simply do not believe the projection. It looks very much like the AI projections which were ludicrously, and perennially, wrong.

  Do I think we’ll make something practical out of quantum computing soon? Maybe. Do I think we’ll make anything which gets much beyond practical soon? No.

  I would be delighted to be wrong. But I think it’s a pretty safe bet.

  • [Flyne]

    It should be read as ‘faster than the classical universe’, as is clear in the article: ‘Meaning, it could solve certain problems that could not be solved by any non-quantum computer, even if the entire mass and energy of the universe was at its disposal and molded into the best possible computer.’

    Predictions about quantum computers are of a qualitatively different kind than those about AI: We have a mathematical model of quantum computing, we know (and can measure) what needs to be done, we just have to do it. Moreover, we can measure exactly how quickly we are making progress. This was not the case for AI: what failed us there was that we did not understand the problem and had no way of measuring how close we were getting.

    • [Tynam]

      Flyne has a point here.  While “3 years” is wild optimism for getting any useful quantum computing done, it is at least fundamentally an engineering problem with which we can make progress.

      AI isn’t an engineering problem, it’s a conceptual one – even if we had infinite resources, we don’t know what we should build with them.  And we didn’t have the slightest idea how hard the problem was until we tried it.  Possibly we still don’t.

• [Rider]

  Can we repost this every three years until this happens so that it will dumber and dumber and dumber as time passes.

• [danegeld]

  Does the D-wave system actually hold water? I thought that they’re one gate short of having a general quantum computer. But I’ll agree they have nice graphs which echo Moore’s law and certainly resemble what i’d expect progress would look like. I think that if you want a genuine, honest-to-god-it-actually-works-and-is-scalable superconducting qubit implementation, check out the research of Martinez, DiCarlo, Schoelkopf. Word on the street is that D-Wave are a bunch of crooks who are good at getting funding, and will be righteously blown out of the water in about five to ten years by something based on transmon devices that actually works. (But I might be wrong) It’s an exciting time

• [elix]

  They harness the refractive echoes of many trillions of parallel universes to perform a computation, unlike anything we have seen before.

  I’ll be willing to eat humble pie quite happily if 2018 rolls around and we’re cracking 4096-bit RSA keys in a couple seconds for fun, but I’m going to call skeptical tech-buzzword bullshit on this guy. How many qubits are we going to need to be able to use in order to handle “many trillions” of quantum outcomes? This sounds like rosy-cheeked blue-sky daydreaming after a gram of bubblegum kush and some pizza.
  Don’t get me wrong. If we can get a handle on quantum computing to the point of mass production (even if it’s only in the double digits initially), this potentially will be a new chapter in our civilization. But this guy seems like he’s trying to sell it too hard. Just say what you think: “It’s possible that in a few short years, advances in quantum computing will give us near-incomprehensible computing power.” To be fair to him, he’s a VC, and according to his Flickr post, he’s been invested in this guy for almost a decade, so I suppose that evangelizing the rosy future he’s gotten in on early is understandable.

  I’d love to be proven wrong, but my optimism for the future is smaller only than my faith in the power of humans to oversell and underdeliver.

  • [Gary Fredericks]

    Well given just 48 qubits you’re dealing with over 250 trillion possible outcomes.

  • [Mantissa128]

    Chill out, old son. The ‘trillions’ bit is from David Deutsch, not some breathless PR department.

• [Steve Jurvetson]

  To be fair, the caveats and qualifiers are important to the context.  There are also definitions of some of the text on the graph.   If you click through to the post, perhaps the bottom half of it should be in bold.    

• [Daniel Rodgers-Pryor]

  No!

  I’m actually studying physics and soon to be working on QC and none of that is right at all!

  1) Quantum computers won’t replace normal computers. Ever. They don’t do the same thing. All quantum computers can do is solve a particular class of problems much more quickly than ordinary computers can. They’re slower, more difficult to use/maintain and MUCH more expensive than classical computers for anything outside this problem-class.

  2) The numbers in that chart are rubbish; they’re based off of the D-Wave system which even they admit isn’t a quantum computer. We are decades away from even a small quantum computer being usable.

  3) There are no parallel universes in quantum computing.

  • [Boundegar]

    Yes but you don’t have millions of dollars.  Your argument is invalid.

  • [Dlo Burns]

    Aww, I was hoping to play Crysis in THE FIFTH DEMENTION

  • [Mantissa128]

    Wow, I feel like a quantum apologist ITT.

    1) If you mean that D-Wave does not have a general-purpose quantum computer, you are correct, but the rest of your point is misleading. A general-purpose quantum computer is one that could act as a universal Turing machine, and there are no logical or physical obstacles to that.

    2) People have been arguing for years that Moore’s law is rubbish. And yet the exponential increases still continue, despite everyone’s hot air. The future will tell.

    3) That depends on your interpretation of the quantum world, Daniel. How does a single photon of light interfere with itself? Even the simplest accounts of quantum phenomena can’t get away from the idea that there is more than one copy of the world that is encountering itself at the smallest scales. Do you know something that scientists around the world do not?

    Harrumph.

  • [AnthonyC]

    I agree with (1) and (2), but (3) is matter of debate. Replace “Parallel universes” with something like “non-zero amplitude at multiple vectors representing possible universal wavefunctions which are distinct but not quite so far apart in configuration space that the interaction between them is zero” and it is much more reasonable.

• [ujin]

  Cynical, jaded comment!!!

  Hopeful caveat.

  • [rocketpjs]

     Smug agreement.

    • [Tynam]

      Mildly sarcastic rejoinder.

    • [Nathan Hornby]

      Misinformed refutation.

      • [musesum]

        Innocuous non-sequitur 

• [stephenl123]

  Shenanigans!

• [redesigned]

  i think this is the graph you meant to use.  it much more clearly shows the progression of Qberts.

• [zog]

  What about the potential negative effects of the Turing-Lovecraft Theorem? wait sec, there a knock at the door…

• [Dillo]

  Just remember gang, it’s all fun-and-games until we’re condemned to our own pocket universe by The Galactic Council for running the cosmic equivalent of a fork-bomb on the universe(s).

  “Sigh, every civilization that discovers quantum computing inevitably does this at least once when they first figure out how to bang the rocks together…”

• [Guest]

  Am I alone in thinking that statements like “more powerful than the entire universe within 3 years” are both meaningless and simultaneously misleading? On a whole host of levels “within 3 years” is reminiscent of the worst of Wired’s breathless futurism/mania, and “more powerful than the universe” is meaningless and silly. Don’t get me wrong, I want him to be right! I’m waiting for the promised Rapture/Singularity/Ascent to the great Cloud Eternal just like the next citizen of the future…but I’d like it to be real, or at least tethered to the real in some useful way. Otherwise, it’s just noise.

• [Ron Chase]

  But will it be fast enough to tell us what women really want?