4
$\begingroup$

Words have meanings. Some words have multiple meanings. Some can mean one thing when used by reasonable people, and quite another when used by the less reasonable. One of those words is supremacy.

The above was written by none other than our very own James Wootton in his Medium article A quantum of supremacy.

'Quantum supremacy' is a phrase originally coined by John Preskill in 2012 in his paper Quantum computing and the entanglement frontier and is used to describe "the point where quantum computers can do things that classical computers can’t, regardless of whether those tasks are useful".

However, it is often considered a controversial term, due to its association with white supremacy, as both James Wootton and Roger Penrose point out.

As such, this question is twofold: Are we happy using the term 'quantum supremacy'? and, if not, What alternative(s) should we use instead?

$\endgroup$
3
  • 3
    $\begingroup$ it might also be worth linking this comment chain on scirate. I hope people will give their opinion about this here. I personally don't really care about which terminology is used, so I'm fine either way. It's worth noting though that regardless of what we decide here, most people asking/answering questions won't be aware of this discussion, and will just use the terminology used in papers anyway. We can edit the questions to correct them but that can only go so far $\endgroup$
    – glS Mod
    Commented Oct 16, 2019 at 10:30
  • $\begingroup$ I thought it was John Preskill? $\endgroup$ Commented Oct 18, 2019 at 0:02
  • $\begingroup$ @MarkS Ha! Of all the mistakes to make! Fixed now, thanks! $\endgroup$
    – Mithrandir24601 Mod
    Commented Oct 18, 2019 at 7:37

5 Answers 5

5
$\begingroup$

I agree with Auden that if there's an abuse of the term then it should be dealt with, and there would probably be other indicia/other issues that are manifest.

I've consistently been using "quantum supremacy" on this site without thinking that it might be offensive to viewers on this site. I wouldn't put up a stink if they were revised to another consensus term. I also absentmindedly tag questions "[quantum-supremacy]" but they are auto-retagged as [quantum-advantage].

However, I believe the term "quantum advantage" fails to convey the monumental goals/achievements being reached.

Note that when I'd used "supremacy" when talking with others about what's going on in the field of quantum computing, I have commented that it's a term that was coined before Charlottesville.

Added later:

It appears that Preskill, who coined the term, has moved to referring to "quantum computational supremacy". This narrowing of the term seems meaningful.

$\endgroup$
4
  • $\begingroup$ Also, Google has adopted the term "beyond-classical" (computation, device, etc). $\endgroup$ Commented Jan 5, 2022 at 7:20
  • $\begingroup$ I've been hearing a lot of talk on "post-classical" computations. $\endgroup$ Commented Feb 3, 2022 at 1:40
  • $\begingroup$ I think "post-classical" jars a bit with "post-quantum" (as in post-quantum cryptography) in that both refer to the situation where quantum computers are available, but use opposite adjectives. Also, it seems to me that "post-classical" carries a suggestion that quantum computers are to replace the classical ones which is of course misleading. IOW, it seems weird to refer to a world where classical computers are the main type of computer, occasionally equipped with a quantum coprocessor, as "post-classical". $\endgroup$ Commented Feb 5, 2022 at 18:33
  • $\begingroup$ OTOH, "beyond-classical" simply suggests ability to do something that classical computers can't, which seems most apt. $\endgroup$ Commented Feb 5, 2022 at 18:34
2
$\begingroup$

"Quantum supremacy" and "quantum advantage" mean two different things. These are terms of art used regularly and publicly by Google, IBM, academics and pretty much everyone else in the field.

This is a site for technical discourse. Not using technical terms of art correctly would unnecessarily introduce ambiguity and confusion.

QCSE is not the right forum to introduce debates regarding social issues. I respect James Wootton for raising issues that concern him on Medium though. That's exactly the type of forum where these sorts of issues should be hashed out.

Out of curiosity I plugged "supremacy" into Google. The three "Top Stories" were:

$\endgroup$
1
  • 4
    $\begingroup$ Couple of comments on this: One is that 'quantum supremacy' isn't remotely well defined in terms of actual numbers, as far as I know (while 'quantum advantage' is). The other is that I'd argue that even 'quantum computational supremacy' (or 'quantum communication supremacy' etc.) is an improvement over 'quantum supremacy' $\endgroup$
    – Mithrandir24601 Mod
    Commented Nov 15, 2019 at 9:38
-1
$\begingroup$

I'm taking my mod cap off for this answer.

I think a few simple principles will help us navigate this.

  1. Try to use "quantum advantage" when you can.
  2. It makes sense to use "quantum supremacy" when the resource being asked about uses the term.
  3. If someone says something is bothering them, Be Nice and try to avoid it. - e.g., if someone tells you "quantum supremacy" bugs you, try to switch.

The key here, I think, will be understanding that both "quantum supremacy" and "quantum advantage" are used reasonably widely in papers (though I think the former is used more widely in "pop" media) and that because of this, both are going to show up on this site.

So, in summary: if you think someone is abusing the term, then let a mod know there's a problem - it'll probably be a part of a broader set of issues anyway. Otherwise, try to preference "quantum advantage" in your writing, but there's places where it just makes sense to use "quantum supremacy", and some may be more familiar with that term.

$\endgroup$
-1
$\begingroup$

tl;dr- While there's nothing wrong with "supremacy" as a word, the term "quantum supremacy" is a misnomer. "Quantum essentiality" would seem more accurate.


1: "Quantum supremacy" is a cheesy misnomer.

Right now we're in a regime of classical supremacy. This is, classical computers are generally superior to quantum computers in almost all contexts.

To clarify definitions:

  • Superiority (being superior) is about being better than another in some way.

  • Supremacy (being supreme) is about being better than all others in some broad sense.

    • Though like many strong descriptors, we often use this term in a weaker sense so long as it's mostly true, barring some exceptions.

Examples of supremacy:

  1. The Supreme Court of the United States has supremacy over legal disputes in the US. This is, even if all other courts in the US rule one way, the Supreme Court can overrule them all in a single decision.

  2. Humans seem to have supremacy over the Earth, as far as animals go.

  3. The "supreme leader" in an authoritarian power structure has supremacy over authority in that power structure, superseding all subordinate leaders' authority.

  4. The Supremacy Clause of the Constitution of the United States asserts that the federal legal structure has legal supremacy. For example, if a city or state law conflicts with a federal law, the federal law wins.

  5. The supremum characterizes the point at which all elements at/above have supremacy over all inferior elements.

If we get to a point where quantum computing methods are generally superior to classical computing methods, such that no one cares to think of classical computing ever again, then that'd be quantum supremacy.


2: "Quantum essentiality" would be more accurate.

Once quantum computers can do things that classical computers can't, they'll become essential to working with general computational workloads. Essential things have the property of essentiality. Presumably we could call this quantum essentiality.

Terminology like this is already used in other technical contexts, e.g. in biochemistry. For example, "essential nutrients" fill roles that couldn't otherwise be filled, so they have the property of "essentiality" in biological functioning.

By contrast, a "supreme nutrient" would be a nutrient that'd be generally superior to all others in a broad sense. So if you're low on Vitamin C? Just eat the supreme nutrient. Low on iron? Ditto. How about Vitamin A? Yeah – you get the picture.

But, we don't have a supreme nutrient because there's no one nutrient that does it all; instead, we have essential nutrients as they're all necessary in the big picture.

That's what quantum computers will become when they can solve problems that classical computers can't – essential, not supreme.

In short, "quantum supremacy" is a cheesy misnomer for quantum essentiality.


3: Regarding the political/politeness concerns..

First, white supremacists believe in white supremacism, not white supremacy. The misnomer may be common in popular media, but.. I dunno. I'm sure all of us hold ourselves to higher standards than that.

Second, I think the worst that could be said of "quantum supremacy" is that it sounds like a cheesy exaggeration that reflects a relatively poor vocabulary. But not bigotry.

But, more to the point, we ought to select our terminology to be as accurate and expressive as possible. So while we should replace "quantum supremacy", e.g. with "quantum essentiality", we should do so for conceptual consistency – not to legitimize misunderstandings.

$\endgroup$
1
  • 2
    $\begingroup$ I think this is off-point though. I don't think we should come out with our own terminology for this. This post should be about deciding which of the terminologies already used in the literature we want to use here. Coming up with another term would only make things more confusing (regardless of it being more or less suitable). $\endgroup$
    – glS Mod
    Commented Jan 24, 2020 at 16:56
-2
$\begingroup$

I'd suggest that we could start using the terms "Quantum Advantage" to mean quadratic or polynomial speedup and "Quantum Super(polynomial) Advantage" for super-polynomial speedup.

$\endgroup$
1
  • 2
    $\begingroup$ we shouldn't invent our own terminology here. Can you point to sources that use the notation you are referring to? $\endgroup$
    – glS Mod
    Commented Jun 13, 2020 at 5:53

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .