Die Sauerkraut ist in mein Lederhosen

I wasn’t at Quantum Information Science Workshop in Vienna, VA, but I heard that the topic of quantum computing “going black” came up at least a few times. One speaker mentioned during his talk that several of his former graduate students were now in “the black hole” of secret U.S. research programs and another expressed, during the open session, that the field is not yet mature enough to be conducting secret research.

Quantum computing is an odd field when it comes to secrecy. Since one (not the only) reason that quantum computers are interesting is that they break the most popular public key cryptosystems, there is a strong national security issue at stake in building a quantum computer. On the other hand, the cat, so to speak, is very much out of the bag, so that it’s not like you can protect the intellectual idea behind quantum computers. But to counter this, suppose that you came up with a surefire way to build a large quantum computer. Should that technology be classified?
There is, of course, only one way to answer this question. A poll, of course:


Bonus points for identifying the title of this post without using the intertuben.

33 Replies to “Die Sauerkraut ist in mein Lederhosen”

  1. Dave, the answer to your question pretty much has to be “it already is”.
    The reason is simple: the practical enterprise of building quantum computers can by no means proceed in isolation from similar math-science-and-engineering enterprises that *already* have strong classified and/or proprietary elements.
    Here are three areas of secrecy (and no doubt there are more):
    (1) Proprietary techniques for factoring large matrices. Pretty much every mechatronic simulation code in the world uses iterative techniques to solve large systems of equations … and companies like ANSYS guard their “multiphysics” algorithms carefully!
    (2) Trade secrets for integrating large-scale stochastic equations. Same as above, except that this time it’s the financial mathematicians and engineers who possess closely-held trade secrets.
    (3) Secretly-held methods for compressive sampling. Same as above, except that this time it’s the communication mathematicians and engineers who have closely-held trade secrets (especially in the private and defense sectors).
    Our QSE Group became aware of these partly-opaque worlds while assembling references for our one-page summary “Efficient Quantum Simulation on Manifolds.”
    http://faculty.washington.edu/sidles/QSEPACK/Kavli/QSE_summary.pdf
    It turns out that the mathematical prerequisites for thinking about QIS/QIT within a manifold framework are nicely summarized on Patricia Schwarz’ string theory page
    http://www.superstringtheory.com/math/math2.html
    But as for the nuts-and-bolts of writing good code that embodies these mathematical ideas (as any serious quantum computer engineering effort would have to do) … well … *those* manifold-centric ideas and algorithms often are very closely held indeed!
    Very roughly speaking, it appears that the good ideas about large-scale computation on vector spaces are mainly in the open literature, while in contrast, the good ideas about computing on large-dimension manifolds (and graphs too) are often very closely held.

  2. How ironic (or more like predictable) that the section entitled “Going Black” in the summary of that workshop (soon to be published in the QT) was removed during the final approval process – and not by the editor.

  3. Hi Dave!
    Isn’t the most realistic vote “It already is”?
    The common-sense grounds is that the key mathematical tools for simulating a quantum computer (with classical resources) are already secretly held—not so much by physicists or even by the Defense Department, as held as trade secrets by the finance, telecommunication, VLSI, and systems engineering communities.
    The reason is simple: (1) nowadays, the development of sophisticated technologies entails detailed simulation, (2) large-dimension state-spaces are generally simulated on (nonlinear) manifolds as contrasted with (linear) vector spaces, and (3) today’s highest-quality math and software tools for simulation on manifold state-spaces are mainly proprietary.
    Examples are startlingly easy to find — the CASI Corporation, for example. It’s been eye-opening to discover how much of today’s state-of-the-art in simulation mathematics is simply missing from the open literature.

  4. Well, if the hint is Val Kilmer, and it’s a ridiculous German phrase, I would guess Top Secret! (1984). Man, it’s been a long time since I’ve seen that movie.

  5. My wife and son (both writers) tell me that every great technological advance is heralded and legitimized by the appearance on the literary scene of a new genre of … techno-thrillers.
    This tradition began with Mark Twain’s techno-thriller novel Tom Sawyer Abroad, featuring Tom, Huck, Jim … embracing that technological wonder of the Gilded Age … balloon flight (yes, this novel really exists).
    In recent decades we have John le Carré’s so[histicated meditations on modern cognitive science … and the usual techno-fantasies on cyberspace, sensors, and drone technology … but otherwise nothing much really new has appeared on the techno-thriller scene (too bad!).
    Which means: probably no near-term “black” quantum computers.

  6. > Well, if the hint is Val Kilmer, and it’s a ridiculous
    > German phrase, I would guess Top Secret! (1984).
    The Val Kilmer hint gave it away instantly.

  7. John,
    I’ve been working on a spy novel for a few years that involves quantum computers. If I had more time I might actually finish it.
    On the “actually published” front, there’s always Paul Preuss’ Secret Passages.

  8. Austrians would kill you for the title, as Vienna is not a part of Germany (though German-speaking), let alone Bavaria.
    For your poll: If the choice is between the NSA being the only ones with the knowledge how to build a quantum computer and everyone sharing this technique, I’d prefer the latter.

  9. Since the US doesn’t seem to be putting a lot of money into academic quantum computing, it is either the case that the research has already “gone dark”, or that the US will be beaten to the punch by a country that actually is putting a lot of money into the field, i.e. Canada. Either way, the question is moot.

  10. Don’t believe it, this is just cover – Dave already has a large scale quantum computer working in his basement

  11. Obviously, this phrase was encrypted in the stoned sequences (Kilmer wore special contact lenses that made his pupils seem dilated in the scenes where Morrison was stoned) of “The Doors” (1991). Prior to the audition, Val Kilmer memorized the lyrics to all songs written by Jim Morrison. He also sent director Oliver Stone a video of him performing a few Doors songs, which Stone claimed hurt Kilmer’s image as Morrison. We all know how accurate Oliver Stone is on all matter Historical and Governmental in his documentarries (i.e. JFK, Salvador, and Alexander, 2004, about Alexander the Great, starring Kilmer as Alexander’s father). Further evidence of NSA involvement in the film: “Mr Mojo Risin” is an anagram of “Jim Morrison” — a clue to transposition cyphers and the like. John Travolta was considered for the role of Jim Morrison in 1986, but objected to Oliver Stone’s paranoia about religion, given that Jim Morrison (as portrayed by Val Kilmer) is frequently linked with and even directly called Dionysus throughout the film. In Alexander (2004), Kilmer plays Philip, whose wife (portrayed by Angelina Jolie) is a devout worshiper of Dionysus. Finally, Oliver Stone states in the commentary for the Director’s Cut that the concert sequences were based on the orgy scene in The Ten Commandments (1956) — a reference to modulo 10 arithmetic.

  12. Ian, I would be very interested to hear more about your novel, and also about the ideas you chose to embed in it (and I will note that Patricia Scharz of “schooloftheuniverse.com” also has a novel coming out).
    Perhaps your Quantum Moxie blog should launch a topic “The scientific ideas that I (am embedding)/(will embed) in my novel …”
    Now, that’s an topic that pretty much everyone among the blogosphere’s technorati is qualified write about!

  13. John, now that’s an idea! I’ll give that a try. And I’ll e-mail you a PDF copy of the first few chapters of my novel and you can tell me if it is pure garbage or not.

  14. The main reason that I don’t think there is any black program in quantum computation, is that no-one on this blog (or anywhere else AFAICT) has offered any concrete suggestions as to *what* would be kept secret in such a program.
    In contrast, it has slowly been dawning on me (in the course of a literature search) that are plenty of computational disciplines that do keep secrets (in addition to the obvious discipline of cryptography, I mean).
    For example, just try to learn how C&SI’s PCGLSS iterative preconditioner works — and yet this C&SI algorithm is at the heart of much of the world’s large-scale simulation software.
    Another great example are the stochastic integrators methods of the financial engineering community … if your market prediction algorithms run better and/or faster than the competitions, then you are strongly motivated *not* to publish them.
    This even extends to quantum simulation software — just search for the phrase “banned by Gaussian”.
    Thus, an important but under-recognized virtue of QIS/QIT community is that present-day QIT/QIS simulation algorithms are open … this openness is *not* true of most of the world’s cutting-edge simulation algorithms.

  15. I expect the NSA is working on this like mad, and it’s a spinal reflex for the NSA to classify just about everything. (You probably need at least a “Secret” level clearance just to read the Fort Meade cafeteria menu.) Ergo, some of the work is being kept secret, whether or not it really deserves to be yet.
    From the NSA’s paranoid point of view, this actually does make some sense; they don’t merely want the details of the Brooklyn Project kept secret, but the precursor stages that led up to it, to increase the time for the competition to re-engineer The Device, and thus the time the NSA has a commanding advantage. The atomic bomb is a lot harder if you only know that E=mc2, and not that some naturally occurring radioisotopes can sustain a fission chain reaction (to say nothing of which ones).

  16. David, I think you and I are basically on the same channel.
    History’s most significant science-and-technology secrets have mainly come under the heading of (1) process technology, (2) simulation technology, or (3) management technology.
    In my bibliographic database I find:

    In a hotly contested Cold War race for technical superiority, the extreme environment of space exacted its toll in numerous failures of extremely expensive systems. Those funding the race demanded results. In response, development organizations created what few expected and what even fewer wanted — a bureaucracy for innovation. (Samuel Johnson, The Secret of Apollo)

    and also

    Tactics, organization, training, leadership, and other products of an effective bureaucracy are necessary to realize the full potential of new inventions … This underscores a theme running through this volume: Having an efficient bureaucracy is the key determinant of whether a country manages to take advantage of a technological revolution. (Max Boot, War Made New)

    From this (admittedly not-too-dramatic) operations-oriented point of view, the contributions (already) of QIT/QIS to advances in process and simulation technology amply justify the DoD’s investment.

  17. David, I hope you don’t battle against “Heroes and Folklore” too hard … `cuz heck … every enterprise (technological or otherwise) needs its heroes and legends.
    But it is equally true that the *last* person you want leading a big enterprise is a hero or a legend.
    Because as Dirac said “A golden era is when ordinary people can make extraordinary contributions.”
    IMHO, Dirac is expressing what the highest levels of leadership (in technology and most other enterprises too) are all about.

  18. I would hazard that the kinds of information that would be classified on such a program would be the things alluded to by abb3w. From what little I really know about QSE, I offer these kinds of categories:
    – Quantum circuit designs [a la IP cores]
    – Manufacturing processes of quantum circuits [my uncle-in-law related to me that in one year, they had a yield of zero of a certain kind of sensor in almost 1000 attempts; and that was on the most obvious of those cold-war-projects-which-shall-not-be-named]
    – Simulation verification of quantum circuit designs [but now that I look back, I see that you mentioned it]
    – Realized or the lack of performance improvements from quantum computers
    – People who might actually understand enough to participate in the project

  19. There are probably a dozen tangents to go from here, but I’ll stay on topic. With respect to the type of project we are talking about, it matters to me who is actually running. This is sad because I work very hard at work to battle the ‘Heroes and Folklore’ assault on technological problems.

  20. David, please come visit our QSE Lab anytime!
    At our QSE Group meetings the interface of technology and narrative is a chronically tendentious topic …
    This week: “Ten reasons why Minerva McGonagall was a better Hogwarts headmaster than Albus Dumbledore.”
    Next week: “Resolved: Thomas Huxley was the role model for Stephen Maturin.” 🙂

  21. I see that John Sidles is a fan of the film “Master and Commander: The Far Side of the Triwizard Tournament.”
    “Darwin also explained the concept lucidly, and with volumes of evidence, to such a degree that Thomas Huxley would say ‘How extremely stupid not to have thought of that!’ upon learning about it.”
    I can SO imagine Stephen Maturin saying that!
    Darwin is already dead, and we know it
    http://scienceblogs.com/pharyngula/2009/02/darwin_is_already_dead_and_we.php?utm_source=sbhomepage&utm_medium=link&utm_content=channellink
    Category: Evolution • History
    Posted on: February 10, 2009 11:57 AM, by PZ Myers

  22. Indeed, our UW QSE Group maintains an in-the-lab “Library of Subversive Literature” — which does include works by JK Rowling and Patrick O’Brian.
    These books are what we call “confections” … warm-ups for the truly subversive books. If we aren’t sure what qualifies a book as being truly subversive … well … heck … mightn’t this mean that we’re not yet ready to be subverted? 🙂

  23. http://bookraft.com/blog/node/182
    In the 20th century, the British Isles have produced many great series of fictions in which one can follow a group of familiar and often endearing characters in their adventures. Many like Tolkein’s Lord of the Rings or J.K. Rowling’s Harry Potter are fantasy worlds of no small entertainments and fancy. Others like Agatha Christies Hercule Poirot or Ian Fleming’s James Bond have a mix of murder and sinister spying intrique (see Agatha Christie’s The Big 4 for a Hercule Poirot in the spy game). Thus Patrick O’Brian’s historical novels set in the Napoleonic War era and featuring Captain Jack Aubrey and his Boat’s Doctor, Steven Maturin are a departure from the the novel norm.
    And as the stories are almost always set at sea with all physical aspects of sailing being the beckoning backdrop of every story. And Patrick O’Brian is far from romanticising the often harsh surrounds, relentless work and sometimes brutal life aboard a British Sailing Man-of-War. What makes these stories so interesting is the subplots of diplomatic and spying intrigue that Steven Maturin is carrying out on each mission. Also the books convey the spirit of the times as France and England now compete most directly for dominance of the seas and the emerging sea trading routes and patters [JVP: patterns?]. These books with their naval warfare nature and spy subplots would appear to be for manly Men only reading until you consider the constant role that strong women play in any number of the stories….

  24. O’Brian’s scholarly biography of Joseph Banks (the same Sir Joseph Banks who was the Royal Society’s longest-serving secretary) is a good stepping-stone to the serious literature of the Enlightenment.
    O’Brian’s regard for historical accuracy is sufficiently scrupulous that it is safe to read the those portions of his novels that deal with natural history (and intelligence … and finance … and naval battles …) not as fiction, but as lightly abridged historical accounts.
    The actual historical accounts are even more fascinating.

  25. And as the stories are almost always set at sea with all physical aspects of sailing being the beckoning backdrop of every story. And Patrick O’Brian is far from romanticising the often harsh surrounds, relentless work and sometimes brutal life aboard a British Sailing Man-of-War. What makes these stories so interesting is the

Leave a Reply

Your email address will not be published. Required fields are marked *