Four updates

A few weeks ago, I was at QIP’2019 in Boulder, CO. This week I was at SQuInT’2019 in Albuquerque, NM. There were lots of amazing talks—feel free to ask in the comments section.

There’s an interview with me at the website “GigaOm,” conducted by Byron Reese and entitled Quantum Computing: Capabilities and Limits. I didn’t proofread the transcript and it has some errors in it, but hopefully the meaning comes through. In other interview news, if you were interested in my podcast with Adam Ford in Melbourne but don’t like YouTube, Adam has helpfully prepared transcripts of the two longest segments: The Ghost in the Quantum Turing Machine and The Winding Road to Quantum Supremacy.

The New York Times ran an article entitled The Hard Part of Computer Science? Getting Into Class, about the surge in computer science majors all over the US, and the shortage of professors to teach them. The article’s go-to example of a university where this is happening is UT Austin, and there’s extensive commentary from my department chair, Don Fussell.

The STOC’2019 accepted papers list is finally out. Lots of cool stuff!

21 Responses to “Four updates”

  1. fred Says:

    As an aside (Scott, I’d understand if you would block this), and interesting interview between Joe Rogan and Andrew Yang, running for president and pushing for basic universal income to counter job displacement due to the current and on-going rise of automation and AI. Lots of interesting questions and data.

  2. Peter Morgan Says:

    With apologies, an unhelpful aspect of your Gigaom conversation is the making of such a strong distinction between classical and quantum: that some things are classical and other things are quantum.
    A better distinction is between measurements that are modeled by functions on a phase space and measurements that are modeled by Hilbert space operators, with the crucial distinction being between commutativity and noncommutativity of the algebras of measurements. Finding ways to say this that are accessible enough for the lay public (or at the level of Gigaom) is not easy, but one elementary statement is that Planck’s constant is a universal constant, equally applicable to all physical models, with quantum fluctuations being at the same scale of the action as noncommutativity in quantum theory.
    The first reason why this matters is that classical mechanics can be presented in a Koopman-von Neumann Hilbert space formalism, in which noncommutativity of measurements is as natural as it is in quantum mechanics, which makes the distinction between what is classical and what is quantum much more delicate. [An alternative way to approach this, from the quantum end, so to speak, making systematic use of Quantum Non-Demolition Measurements, is through Tsang&Caves’ “Evading quantum mechanics: Engineering a Classical Subsystem within a Quantum Environment”,
    Focusing on the Hilbert space aspect instead of on “quantum” is also helpful because we can usefully use any finite N-dimensional Hilbert space, not just the 2^n dimensional Hilbert space of an n-qubit system. It seems possible we might match Moore’s law for the dimensionality of the Hilbert space, but not as likely for the number of qubits. We should also focus on how far the reliable dimensionality of the group of unitary transformations in a given implementation falls short of N^2.

  3. mr_squiggle Says:

    In the Byron Reese interview, you said this:

    Some people like Roger Penrose have seized on the observation by Godel and Turing that no machine can be a perfect oracle for mathematics. In order to say that the brain or at least the mathematician’s brain must be doing something that a Turing machine can’t, but the obvious problem with that argument is that humans are not perfect oracles for mathematics either to put it very mildly.

    Thanks for saying that. I read a book by Penrose a couple of decades ago, and it was painful. My memory of is is that he seemed to spend the whole book addressing every possible objection to his theory except the obvious one.

  4. anonymous Says:

    It is very depressing for me to see that on one hand, universities are claiming that they are having a hard time recruiting faculty, but on the other, I am on track to be rejected from every single one of the 30 postdoc positions I have applied to. The issue is of course that universities are having a hard time recruiting faculty that meet their very high standards, standards which make it impossible to increase faculty hiring when conferences only accept a certain number of papers.

  5. Tamás V Says:

    In the GigaOm interview, and in general, wouldn’t it be less confusing to say only that entanglement is a correlation, without mentioning that it’s “instantaneous”? Isn’t it rather a “timeless” correlation?

    If Alice and Bob each possesses one qubit of an EPR pair, and Alice measures hers at time t and gets |0>, is there any experiment (incl. the possibility that they match their records later when they next meet) that can single out time t as the time when Bob’s qubit really changed in a way that he’d surely get |0> if he measured? (Other than saying it’s obvious because Alice did something at time t.)

  6. Scott Says:

    Tamás #5: Well, I try to make contact with where people currently are, even if there are misconceptions in it. Indeed, if you could measure a “speed of entanglement,” or a “time at which Alice’s measurement affected Bob’s state,” or anything else of that kind, then special relativity would be wrong and you would have superluminal communication.

  7. Tamás V Says:

    Scott #6: Well, that’s why I generously allowed that Alice and Bob can check their records later as part of the experiment, to avoid superluminal communication and save special relativity 🙂

    At high school, I read lots of popsci books that included chapters that tried to explain special relativity to a wider audience, taking into consideration “where people were”. The net result was that I had zero chance to understand it, had more questions than answers in the end. I could understand the basics of the basics only 20 years later 🙂

    Your intro to QC in the interview was the most concise and precise and understandable I’ve ever seen (shared it on LinkedIn immediately, or rather, instantaneously). I’m sure you could find a way to explain entanglement too without menntioning the word “instantaneous”.

  8. gentzen Says:

    Scott, you wrote:

    I didn’t proofread the transcript and it has some errors in it, but hopefully the meaning comes through

    It would be interesting to know which are those errors. How about this?

    It is now Google, IBM, Microsoft, Intel, a bunch of startup companies, all are investing on a scale of hundreds of billions of dollars.

    That would be more than 10% of the net worth of Alphabet (Google), Amazon, Apple, … and much more than the cost of past and future supercoliders.

    This also seems not have some inaccuracies:

    At a bare minimum, so there are many different approaches to quantum computing, but if you’re doing superconducting qubits, which is maybe the most popular approach today, then at a bare minimum, you need to cool everything down to 10 mKB, or so, so that your chip superconducts and you see the quantum behavior, so that means that you need to see an enormous cooling system.

    Is “10 mKB” the new unit for milli-Kelvin bytes? Also, isn’t this just the temperature that you reach with a dilution refrigerator, i.e. they just cooled it to the lowest temperature available at a reasonable cost, not because they were stricktly forced to do that (1 K would be cold enough for most superconductors, 4 K is cold enough for mercury, and 9 K is still cold enough for Niob), but because they could do it, and because it somehow helped increase the quality of the qubits?

  9. Scott Says:

    gentzen #8: Yes, “billions” should be “millions,” and “10 mKB” should be “10 millikelvin.” And I’m certain that I said those things correctly in the interview. And they’re good examples of why I took care to forewarn people about errors in the transcript. Unfortunately, as soon as I start emailing in corrections, I’m then responsible for anything I didn’t correct, and I don’t have time for that.

  10. Scott Says:

    Tamás #7: If you want to read me explaining entanglement “the right way,” why don’t you check out my undergrad lecture notes or my American Scientist article or any of the innumerable other places where I’ve tried to do that!

  11. Joe Shipman Says:

    I’m still waiting for someone to express the progress in building computers using the metric “what integers can be factored by Shor’s algorithm, without cheating, today?”

  12. Scott Says:

    Joe #11: Then you’ll be waiting a long time, because that’s a stupid metric. (Or rather: anyone willing to use that metric today, is probably someone you shouldn’t trust.) It’s almost exactly analogous to judging the Manhattan Project in 1942, 1943, 1944 by the metric “how big an explosion can you make, today?” I.e., it’s something that’s going to be basically flat for a long time and then undergo a step change, under the assumption that everything is progressing as it should—because you need a critical mass in the one case and to exceed the fault-tolerance threshold in the other.

    No, that doesn’t mean scalable QC is around the corner; it just means you can’t conclude anything one way or the other from the abandonment of factoring circus-stunts at the end of the liquid NMR era almost 20 years ago. If you want to have a real discussion about experimental progress (or where it’s stalled), look at coherence times and 2-qubit gate fidelities, and at the ability to maintain those fidelities in integrated systems.

  13. Raoul Ohio Says:

    Scott #12.

    Agree that biggest number factored is a stupid metric.

    On the other hand, it is totally easy to understand. So there’s that. Also dimensionless – always good.

    Does anyone know if they ever got past 15 = 3 * 5 ?

  14. Scott Says:

    Raoul #13: 21=3*7 was done with some precompiling/cheating. But none of this matters. No really, it doesn’t. The obsessive focus on an “easy to understand” but currently meaningless metric is what set naïve people up to get impressed by the claims of factoring 6-digit numbers or whatever using quantum annealing—something that not only had nothing to do with Shor’s algorithm, but likely had nothing to do with quantum mechanics at all. So it’s done real damage to the field.

  15. Tamás V Says:

    Raoul Ohio #13: There is also a practical aspect. If D-Wave could suddenly factor 2048-bit integers within 1 day, the CISO wouldn’t be amused to know it’s not even a quantum computer. And most likely many people would be very interested to find out where exactly the speedup comes from, even if it’s only a constant factor.

  16. Andrei Says:


    In the interview with Adam Ford, in regards to the double-slit experiment you say:

    “To say that again, like decreasing the number of paths that the photon could take to reach a certain spot, you can increase the chance that it gets to that spot. This is the thing that violates any conventional understanding of probability.”

    In my opinion, the difficulty of understanding the double-slit experiment classically has its origin in using the wrong classical model. This model is Newtonian mechanics of the rigid body. The particles are represented by bullets or marbles and the barrier as a concrete wall. Indeed, explaining the single-particle interference pattern using such a model is impossible.

    The problem is that there are a lot of physical phenomena that cannot be explained with bullets either, like magnets, electromagnetical induction, black-holes. You need a field theory for them. Let’s see how strange the two-slit experiment looks when a field model is used.

    So, we have an electron approaching the barrier which consists of a great number of electrons and nuclei. The trajectory the electron takes depends on the electric and magnetic fields existing at its location, and those fields are determined by the positions and velocities of the electrons and nuclei inside the barrier. It seems obvious for me that the electric and magnetic fields that are generated by a barrier with one slit will be different than those generated by a barrier with two slits. Just like the quantum amplitudes, those fields may add up or cancel each other. So, I see no reason to be perplexed by the fact that closing one slit determines the electron to go into a region where it could not arrive before. The fields, that are different, exert a different force on the electron thereby changing its trajectory.

    I know that you used photons in your example but those are EM waves classically so I think that the experiment with electrons is more interesting to explain.

  17. Gerard Says:

    @Andrei #16

    I don’t see how that makes sense.

    The double-slit experiment would normally be done with an electrically neutral and non-magnetic barrier so there are no EM fields outside of the barrier material.

    Also it is known experimentally that low energy electrons do not penetrate any significant distance into solid matter.

    So the barrier material is doing just what it appears: allowing particles to pass outside of it and preventing them from passing inside it.

  18. Andrei Says:


    “The double-slit experiment would normally be done with an electrically neutral and non-magnetic barrier so there are no EM fields outside of the barrier material.”

    The barrier is made out of atoms. Atoms consist of charged particles (electrons and protons/quarks). On average, the barrier will be neutral (same number of electrons and protons) but those particles do not occupy the same position so what you get is a large number of dipoles. Those dipoles produce electric and magnetic fields of unlimited range.

    The EM interactions between two macroscopic objects are difficult to see at large distance because the attractive and repulsive forces between very large groups of particles average out. But for a single electron the interaction should be noticeable.

    “Also it is known experimentally that low energy electrons do not penetrate any significant distance into solid matter.”

    The electrons do not penetrate the barrier, they interact from a distance with the fields produced by the charged particles in the barrier.

    “So the barrier material is doing just what it appears: allowing particles to pass outside of it and preventing them from passing inside it.”

    The electron is prevented to enter the barrier because there is a large electric repulsion between it and the electrons in the barrier. But this interaction takes place from a distance. The electrons don’t just bump into other electrons, but their trajectory curves as a result of the Lorentz force. The way it curves will obviously depend on the electric and magnetic fields present at that location and those fields depend on the position and momenta of field sources.

  19. Ajit R. Jadhav Says:

    Andrei #18 and #16:

    The fact that you appreciate the existence of atoms in the barrier is, say, appreciated! Your general line of reasoning has some really good points to it, too.

    However, to understand the QM mysteries right, don’t focus too much on the double-slit experiment. Instead, look for the light-matter interactions.

    Light, of course, can be seen as a local and propagating disturbance within an EM field (at least the basic element of analysis, i.e. the plane-waves can be). But as people soon figured out, if you apply the EM theory to the cavity radiation problem, an inevitable implication is that the ultraviolet catastrophe is predicted, but doesn’t occur in reality. This problem (for one) brought out the limitation of using the classical EM theory to model the light-matter interactions. Also the photo-electric effect, and the curious dependency on the frequency of light, not on its intensity. The classical EM fails to explain light-matter interactions. That’s why QM was born.

    To cut a long story short, the outcome was that people figured out that it’s the classical EM phenomena that should be explained on the basis of the QM principles, rather than the other way round, i.e., rather than trying to explain the QM phenomena on the basis of the classical EM principles.

    Demonstrating how the limitations of the EM approach leads to problems in the double-slit experiment is relatively more difficult, because the moving electron’s contribution to the overall EM field in the interference chamber doesn’t actually interact a lot with matter.

    In case you didn’t know, in actual experiments with electrons, there is no chamber of the kind they sketch in textbooks. It’s more like the insides of a TEM (transmission electron microscope): a very huge and vast cavity (compared to the effective size of the electron), with a very thin wire serving as the middle separating portion of the barrier wall, with huge, huge open spaces on both sides of this wire. In other words, the “width” of the open “slits” on the either sides of the wall is far bigger than the opaque (to electrons) spacing provided by the wire in between the two “slits”. Effectively, rather strong and nonuniform EM fields act as barriers on the _other_ sides of the “slits”, by way of “wall” on that side. But they are so distant from their atomic sources and so strong, that the lone electron’s tiny field would have next-to-no chance to interact with any piece of matter. About the only interaction the traveling electron ends up having is with the detector material, when it lands there. But that event leads to its absorption, and so, there is no chance further “downstream” to see the effect of _this_ interaction of the fields of the electron with matter. All in all, a tough proposition for an analysis from the viewpoint of a classical EM field’s interaction. [And hope that I got the description of the actual chamber not too wrong.]

    Much better it is to see the situations where the matter-light interactions are more extensive, as in atomic gas emission/absorption spectra. That leads you to Bohr.

    Finally, just as an aside, it’s often very fruitful to figure out why developments occurred in the order in which they historically did. You learn not just interesting bits of history, but also get an invaluable chance for (re)organizing the conceptual structure of your knowledge—if you want to.


    PS: Sorry for yet another long reply… But his idea and approach was so interesting….

  20. Gerard Says:

    @Andrei #18

    If electric or magnetic dipole fields were produced by ordinary solid matter those fields would be measurable and there existence would be well-known, which they obviously are not.

    I’m not going to try to give a complete explanation for why they don’t exist but I’ll mention a couple of points.

    First the charge distribution of atoms tends to be spherically symmetric with the positive charges in the nucleus surrounded by an electron cloud, so there’s not normally a net dipole moment.

    Some molecules do have dipole moments, for example water molecules, and these play a role in things like the Van der Waals force, but that is a very short range effect.

    Secondly even if you have a collection of polarized molecules, such as a glass of water, the distribution over orientations will be uniform so there will still be no dipole field any significant distance away from the material.

  21. Andrei Says:


    “The fact that you appreciate the existence of atoms in the barrier is, say, appreciated! Your general line of reasoning has some really good points to it, too.”


    “However, to understand the QM mysteries right, don’t focus too much on the double-slit experiment. Instead, look for the light-matter interactions.”

    I have formal studies in QM and I am aware of the history regarding the discovery of QM. Yet I find most arguments against classical physics (the general framework of classical physics not necessarily a particular theory, like classical electrodynamics, which could be certainly improved or changed) unconvincing.

    Take for example the case of the stability of atoms. I’ve learned that the classical atom cannot be stable because the electron accelerates, radiates energy and falls to the nucleus. So, classical EM is doomed! Then I have discovered a classical model of the atom, the “free-fall” model, published in good quality peer-reviewed journals. The idea was that once you take into account the spin of the electron the atoms become stable because the spin determines the electron to avoid the nucleus. There is not much literature on that subject, it was mostly ignored.

    A different idea is to assume the existence of a external EM field. A theory named stochastic electrodynamics makes use of this and made some progress in explaining the quantum behavior. The point is that when the electron accelerates it takes energy from the field so that it never falls on the nucleus. This external field also determines the property of spin, wave-like behavior, etc.

    I cannot comment on your arguments regarding the relative strength of various fields encountered by the electron in a two-slit experiment, I think that a simulation would be required to see what is significant and what is not, but even if you are right there is always the possibility of developing a new classical field theory that does the job. So, the problem is not that one cannot imagine a classical explanation for this experiment. The problem is to see if one can reproduce quantitatively the results.

Leave a Reply

Comment Policy: All comments are placed in moderation and reviewed prior to appearing. Comments can be left in moderation for any reason, but in particular, for ad-hominem attacks, hatred of groups of people, or snide and patronizing tone. Also: comments that link to a paper or article and, in effect, challenge me to respond to it are at severe risk of being left in moderation, as such comments place demands on my time that I can no longer meet. You'll have a much better chance of a response from me if you formulate your own argument here, rather than outsourcing the job to someone else. I sometimes accidentally miss perfectly reasonable comments in the moderation queue, or they get caught in the spam filter. If you feel this may have been the case with your comment, shoot me an email.