Quantum Advantage

Update (22 May 2017): This Scirate thread seems to have touched a nerve. Since this was previously buried in the comments here, it’s worth promoting to the top of the post. I think that “quantum computational supremacy” addresses the concern. Basically, we use “quantum” as an adjective for our peer group, which makes the analogy to “white” too strong. Adding “computational” emphasizes that it is the computation, not the people, that are supreme.

I’ve had quite a few conversations lately about a comment I left on Scirate. The paper at that link, “Quantum advantage with shallow circuits” by Sergey Bravyi, David Gosset, Robert Koenig, shows a provable separation between analogous classes of quantum and classical circuits, even when the quantum circuit is restricted to nearest-neighbor gates on a 2D grid. This is a fantastic result! My comment, however, wasn’t regarding the result, but rather the title of the paper. I’m just happy that they called it a “quantum advantage” instead of using that other term…
The term “quantum supremacy” is the fashionable name for the quantum experiments attempting to beat classical computers at some given task, not necessarily a useful one. According to current usage, the term (strangely) only applies to computational problems. The theoretical and experimental work towards demonstrating this is wonderful. But the term itself, as any native English speaker can tell you, has the unfortunate feature that it immediately calls to mind “white supremacy”. Indeed, one can even quantify this using a Google ngram search for *_ADJ supremacy over all books in Google’s corpus between 1900 and 2008:

None of these terms has a particularly good connotation, but white supremacy (the worst on the list) is an order of magnitude more common than the others and has, on net, been growing since the 30s. For almost every native speaker that I’ve talked to, and quite a few non-native speakers as well, the taint of this is hard to escape. (For speakers of German or French, this word is a bit like “Vormachtstellung” or “collaboration” respectively.)
The humor surrounding this term has always been in bad taste — talking about “quantum supremacists” and jokes about disavowing their support — but it was perhaps tolerable before the US election in November. Given that there are several viable alternatives, for example “quantum advantage” or even “quantum superiority”, can we please agree as a community to abandon this awful term?
This isn’t about being PC. And I’m not trying to shame any of the people that have used this term. It’s just a poor word choice, and we don’t have to be stuck with it. Connotations of words matter: you don’t say someone is “scrawny” if you mean they are thin, even though my thesaurus lists these words as synonyms. Given the readily available alternatives, the only case I can think of for “supremacy” at this point is inertia, which is a rather poor argument.
So please, say it with me now: quantum advantage.
Update: Ashley Montanaro points out that “advantage” should potentially be reserved for a slight advantage. I maintain that “superiority” is still a good choice, and I also offer “dominance” as another alternative. Martin Schwarz suggests some variation of “breaking the X barrier”, which has a nice feel to it. 

28 Replies to “Quantum Advantage”

  1. Hi Steve,
    I think it’s a good idea to have this debate, and as with many other people I agree that the word “supremacy” is far from ideal and leaves a bad taste in the mouth, for the reasons you give. However, unfortunately I don’t think that “advantage” or even “superiority” mean the same thing, so I don’t think they’re good replacements.
    My feeling is that one can demonstrate quantum advantage without demonstrating quantum supremacy. Even Deutsch’s algorithm is an example of quantum advantage, but quantum supremacy is doing something which you could never do in a reasonable time on a classical computer. For example, something that would need 100 years of supercomputer time. Also, I think that “quantum supremacy” specifies an event: the first time that quantum computers dramatically outperform classical ones. I’m not sure that “quantum advantage” can be interpreted as an event.
    The use in the paper you mention of “advantage” is completely right, in my opinion, because they show that you can do something with constant-depth quantum circuits that you can’t do with constant-depth classical circuits. But this isn’t an example of quantum supremacy, because you can also do it with a polynomial-time classical algorithm.
    If there were an alternative word that would be an immediate drop-in replacement with the same meaning, but not the same unpleasant connotations, that would be great. But I’m not sure there is, although any suggestions would be welcome! I think it would need to be a noun, and one that can specify an event. Of course, we could all agree not to use any term at all, but this seems a shame, as having something specific to use does seem useful.

    1. These are good points, Ashley. I think “superiority” does achieve the goals that you want for the term. Even though there are some phrases that have negative connotations with superiority, it also has more neutral meanings as well, like in the phrase “moral superiority”. “Dominance” is perhaps another choice; what do you think of that one?
      Or maybe we could go with “classical inferiority”. 🙂

      1. I’m not sure… I think the classical folk see us as acting overly superior as it is 🙂 And to me it seems like “superior” might mean “only a little bit better”.
        “Dominance” is interesting because in mathematical usage it seems to make sense (eg. one term dominating an expression) – though the dictionary suggests it may have similar problems with connotations (Merriam-Webster’s first definition is “controlling, prevailing, or powerful position especially in a social hierarchy”…).
        In many ways I like Raphael’s suggestion of quantum victory, but perhaps this too seems too antagonistic. Quantum triumph? But then there’s no notion of comparison…

  2. So should we also stop saying “naval supremacy” and “commercial supremacy” and so on?
    If Nazis talk about “racial hygiene” laws, should we stop using the term “hygiene” in other contexts?
    I get how the word “supremacy” reminds you of something unpleasant but I don’t think anyone would interpret it as an endorsement of white supremacy.
    A different objection you could raise is to the term “quantum.” These experiments really demonstrate “computational supremacy” analogous to earlier quests for things like maritime supremacy.

    1. I think your last point is very insightful, and it helps to highlight why the first point you make is a weak argument. You first ask the rhetorical question of whether we should stop using terms that have been tainted by unpleasant associations. Well, I’m not saying that “yes” or “no” follows logically; however, we have a choice. Given the readily available alternatives like “superiority” or “dominance”, why would you actively choose a word that did have negative associations? Regarding the terms “naval supremacy”, etc., I have also heard “superiority” used in that context (much more often in my experience). And regarding the feeling of righteousness (“I won’t let them take my word, so I’ll keep using it in protest!”), it’s not like we’re a marginalized group that is reclaiming a term. So we shouldn’t adopt the term as a form of protest or resistance.
      In addition, your analogy is weak because word pairs like “oral hygiene” are well established and have no negative connotations. “Supremacy” is definitely distinct in that regard since it is strongly associated with just one additional qualifier. And I’m certainly not saying that using it is an endorsement of white supremacy, or could even reasonably be interpreted as such. Again, the point is that we have a choice. Why continue doing something unpleasant if you don’t have to?
      Your last point, that it is really “computational supremacy”, greatly clarifies why “quantum supremacy” is a bad term. The word “quantum” is often used in our community as the name of our tribe! To say “X supremacy” where X is the name of a group of people is definitely crossing a line in my mind. The fact that people mention “quantum supremacists” and so on really reinforces this reading. Whereas if X is a resource, the connotations are a lot milder. Saying “quantum computational supremacy” or even just “computational supremacy” is orders of magnitude better to my ears since it completely removes the tribal associations.

  3. Since the scientific aim of “quantum supremacy” experiments is to provide experimental evidence against the Extended Church-Turing thesis (see e.g. arXiv:1011.3245), why don’t we use “breaking the Church-Turing barrier” instead of “achieving quantum supremacy”. “Breaking a barrier” is shorter and sounds slightly more catchy than disproving (or providing evidence against) an (extended) thesis, yet it’s quite to the point and not an exaggeration. It has desirable positive connotations of speed (as in sound, or speed-of-light), practical hardness, and entering a new regime where new rules apply, and no negative ones, as far as I can tell.

    1. I quite like this suggestion Martin! However, it’s a little awkward to use in practice, e.g. “breaking the quantum barrier(edness ? :-)) is achieved with fidelity x and number of qubits y” is a little tricky. I do think it’s in the right direction.
      In talks lately I’ve been trying to use “beyond classical”, “unambiguous quantum advantage”, and “supra-classical”. All of them have drawbacks. I like the word “beyond” a lot as it has connotations of exploration. On that, “beyond the quantum frontier” is a nice turn of phrase – also more-or-less due to John Preskill.
      I should say I agree with pretty much all of what Ashley is saying above. Though, we have (obviously) discussed this a lot.
      In practice that “quantum supremacy” has fallen into common usage in the science media without too much complaint.
      Finally I should say that bconnotations remain bad until they are challenged. I believe that conceding, or reserving, words for extremists (esp. when they have other uses) might not be a great idea for improving political discourse. On the other hand I’d like a more positive term, like “beyond”, as it captures the imagination a bit better.

      1. I like “breaking the X barrier” for some X, but I’m not sure that X=Church-Turing works – after all, it’s only the extended Church-Turing thesis that’s falsified by the quantum experiment, and neither Church nor Turing actually proposed that.
        Maybe we’ll just end up with a variety of phrases for different contexts (e.g. “supra-classical” as an adjective, “supremacy” as an event, and “beyond the classical frontier” as a longer phrase). It’s interesting that John Preskill didn’t use the word “supremacy” in the title of his original article, but went for something longer and more positive.
        I also agree with Mick that it’s a shame to just give up on a perfectly good word because of its usage by a group of extremists.

        1. This discussion led me to think about other alternatives, months later: “Church-Turing intractability” or “Overwhelming Church-Turing complexity”. The following example sound good to me, but I’m not an impartial judge :
          – This experiment solves Church-Turing intractable problems;
          – This machine allows us to explore the regime of overwhelming Church-Turing complexity which, by definition, is not reachable in practice by classical computers.

  4. Variations on supremacy feel less harsh: supreme court, supremeness (is that in Bill & Ted?), chicken supreme, Diana Ross and the Supremes… Thinking about it, I’d pay good money to see a Mel Brooks production of Baby Love by Diana Ross and the Supremacists.
    Anyway, my suggestion is for Quantum Transcendence.

    1. Yeah, these other variations have no negative connotations. I think Aram’s comment has really clarified for me that the reason “quantum supremacy” sounds so bad is that we really do use “quantum” as the name of our community, and “(group of people) supremacy” is just a bridge too far.

  5. At QIP 2017 (in Seattle) the phrase “quantum ascendancy” was heard.
    This phrase (as it seems to me) aptly describes the study of quantum dynamics in respect to worthy pursuits like “computational ascendancy”, “observational ascendancy”, and “metrological ascendancy” (and more).

  6. On quantum s?t?u?p?i?d?t?y? supremacy: whatever the term happens to be, we are going to use it as soon as some minute and marginally convincing improvement is found for an otherwise pointless task. We’ll all shout hurrah and Google stock will jump half a point. Then, when pressed by the media for soundbites about how awesome we all are, we’ll be forced to throw our tail between our legs and backtrack on yet another hype orgy in our field. This whole exercise is annoyingly arrogant and embarrassing. Show a little humility and shut up about the idea of beating your strawman competition into submission.

  7. I cannot believe we, as scientists, are having this discussion.
    I’m sorry, I don’t want to sound aggressive, but when I first saw Prof Wiesner’s paper, I thought of a bad joke.
    Seriously? Do you advocate for modifying well-established scientific language every time the political or social situation hints at some unpleasant association with the term? Did anybody in the world, before the Trump obsession started hammering our heads, ever associate “quantum supremacy” with “white supremacy”? What about “black hole”? What about the Alice/Bob dichotomy? What about “spectrum”? What about “genetic algorithm” and “discarding genes with low fitness function”?
    This is so much worse given that, yes, there are problems of inclusion in our (and most of any other) community. And the discussion we are having doesn’t do anything good for solving these problems. It’s a moral placebo, which makes us scientists look silly, and makes it harder establishing serious efforts trying to improve things.
    Even worse, it’s an anger catalyst, a deliberate quest for aggression. It’s the PhD equivalent of a pub brawl between football hooligans. A little of experience on “The Internet” can teach anyone that the only possible outcome for this discussion is that sooner or later the comments are going to be closed.
    This paper, and all the discussion around it, is clickbait. It’s first-world problems, it’s social media drama, it’s a shame as disconnected from science as “ancilla qubit” is disconnected from “female slavery”.
    I’m feeling embarrassed. 10/10 for making me reply.
    As for the “quantum computational supremacy” proposal, I lode Steve’s effort to find a good enough common ground. It’s not my task to wage war against one or the other term (if I had to, I would choose “post-quantum” as my arch-enemy), and I will eventually refer to the most widely accepted term in the community. However, I feel that “quantum computational supremacy”, instead of being a good compromise, summarizes the worst possible options:
    – it would not make Prof. Wiesner happy, because “supremacy” is still there;
    – on the other hand, as already pointed out by others, it adds an unnecessary and slightly misleading adjective.
    But hey, again, whatever you people like. Language should serve science, not the other way around.

    1. Since you want to make the dichotomy, let me illustrate what *you people* look like.
      Wiesner brought these examples up to make a point. These examples support that point. Full stop.
      The outrage is on *you*. You don’t want to believe what is blindingly obvious, so you attack. You drum up support from the worst of your kind by concern trolling.
      In fact, the examples are now wholly irrelevant because you all have made the point better than the examples could.
      You say you can’t believe we are having this discussion. This is not a discussion, this is a bunch of people who look like fools by arguing what terminology best describes a superiority they haven’t even achieved, and all while under the illusion that they are successfully refuting a claim against themselves that was never made in the first place.
      You have all missed the point. Wiesner did not advocate changing terminology. If you want to change the terminology, good. But do so of your own accord—not because boo-hoo the boys club has ended and PC brigade won, but because you understand the deeper problem it supports.

  8. Why was the Scirate thread closed? It seemed like it was just a fun discussion.
    I oppose “quantum supremacy” because it reeks of self-promotion, it sounds like something you’d read in a bad Wired article. (I’m not trying to pick on Wired.) For me, the connotations of “supremacy” in science are more about hucksterism than racism.
    For “ancilla,” if a term once had a bad meaning—which isn’t clear in this case—it is good to use it in a different context to take the term back. (This doesn’t work for “supremacy” because white supremacy is still a major political, oppressive force.)

  9. Hey All,
    I closed the SciRate thread. Here’s why:
    I think that no discussion on this topic on SciRate will be useful. Partly due
    to people being anonymous, but partly due to lack of representation of people
    that are actually upset by the words “supremacy” and “ancilla”. It doesn’t
    matter to me who they are, but they should be part of any discussion, and I
    didn’t see anyone, really, express that they had a problem with the terms,
    aside from people saying that, irregardless of social issues, the words were
    bad choices technically. So I don’t think both sides were well represented.
    And the big problem is you don’t get people chiming in with “Hey, I find this
    really upsetting, but I’d love to stick around and hear why I’m wrong!”;
    instead, those people just silently leave, and stop participating, and feel
    I think that engaging with trolls, like those whom were present in the
    thread, is not productive, as they are not particularly interested in rational
    and productive discussion.
    I want SciRate to feel like a welcoming and supportive community to everyone,
    and I didn’t feel that the discussion was constructive or going to result in
    a clear and agreeable conclusion.
    My personal opinion is that a few key people should get together and set the
    direction for the community. I believe Bravyi, Gosset, and Koenig [1] are the
    most recent positive example of this. It might be good if the author of the recent
    paper could get together a few members of the community and make a
    constructive statement with concrete proposals; but that’s not up to me.
    More generally, I think one of the most important things is to listen to the
    people that are explaining why they find particular terms upsetting, instead
    of explaining why _you yourself_ do _not_ find it to be upsetting.
    In any case, I hope this explains why I closed the SciRate thread.
    [1]: https://scirate.com/arxiv/1704.00690

    1. Following the very first comment, I am wondering, maybe the mentioned paper [1] just corresponds to “an advantage slighter than supremacy”, e.g. (if to use CS terms) does not conflict with Extended Turing-Church ideas, being more powerful than only some restricted classical model?

  10. Dr. Wiesner’s writing touches upon a very useful observation – “Science is downstream from culture”. That is, while the *results* of science (data) are objective, derived from measurements of the world, the *practice* of science is less so, as it is performed by humans. Aspects of practicing science which are influenced by culture include determining the set of problems to be investigated, and the use of language in scientific writing.
    The broad topics science ends up focusing on are determined, in part, by the cultural norms of the times. For example, medical testing on humans is now more closely monitored and regulated, whereas in the past it was not. The emergence of climate science as a discipline is probably related to efforts to increase environmental awareness in society at large. Generally, the lines of inquiry the dominant cultural norms sanction as being permissible determines what science investigates.
    As a corollary, as times change, scientific language itself may be altered. What was once acceptable, becomes not, and what was never considered, becomes acceptable. When the words you use are re-interpreted as a result, you must make a choice about whether to change them.
    Reading this paper with such a perspective in mind, perhaps a good question to ask ourselves is “How do my norms and values compare to the ones being explicitly (and implicitly) presented here?”. If there’s substantial disagreement, a natural follow-on question suggests itself: “I’m being asked to change who I am; am I OK with doing so?”.
    Quantum information is a small community (relatively speaking!). Even so, there are different norms and values held by its members. This paper helps make that clear, and Dr. Wiesner has done us a favor in writing it.

  11. The advantage of “supremacy” over “advantage” is that in many cases where quantum computers boost the computation time this is by a huge amount and for that advantage is too weak. Indeed “quantum computational supremacy” sounds good.

Leave a Reply

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