Talk:Quantum computing

Merge: Q17995793 Quantum Computing / Q176555 Quantum Computer Reason: Language links

They lead to the exact same article, but the Quantum Computing language links seem to supersede the Quantum Computer one. So there are about 30 links language links missing from the Quantum Computing article. Justcheckingin (talk) 05:22, 1 April 2023 (UTC)

I see the issue that you are talking about, but I am not very familiar with Wikidata, so I am not sure about the best course of action. It seems that other languages of Wikipedia have separate articles for the two, like pt:Computação quântica and pt:Computador quântico, so I do not know whether it would be best to merge d:Q17995793 and d:Q176555 or just to add the missing language links to d:Q17995793. It might be worth asking at the help desk or teahouse (or maybe asking at the Wikidata project site instead).  — Freoh 11:11, 2 April 2023 (UTC)

If I understand the wikidata guide page wikidata:Wikidata:Sitelinks to redirects correctly, the solution is to make sitelinks to redirect pages and tag them with a badge “intentional redirect”. I wonder if there is a tool to do this since otherwise it involves a lot of manual labour. Jähmefyysikko (talk) 15:51, 2 April 2023 (UTC)

I made a query here: wikidata:Wikidata:Project chat#Quantum computing and quantum computer Jähmefyysikko (talk) 16:47, 2 April 2023 (UTC)

Received a reply: It should indeed be done by adding sitelinks to redirects, but unfortunately there is no tool. Also, I noticed that in some wikis, there are no redirects to their Quantum computing/computer page. In this case it is not possible to make a sitelink directly, but it would require a creation of a new redirect page to that wiki. Jähmefyysikko (talk) 11:59, 3 April 2023 (UTC)

Jähmefyysikko, if I understand correctly, the best solution here would be to ensure that every language of Wikipedia that has a quantum computer page also has a quantum computing page (and vice versa), creating a new redirect for languages that have one but not the other. I do not speak most of these languages, so I do not feel comfortable making dozens of redirect pages with titles that I do not understand. Do you think that it would be appropriate to add Wikidata entries pointing directly to the existing pages? For example, adding an entry to d:Q17995793 linking to als:Quantencomputer? Then, we could leave it to people who speak those languages to create the quantum computing redirects and update the Wikidata entry to the (more precise) sitelink to redirect.  — Freoh 11:58, 6 April 2023 (UTC)

I agree that we should not make new redirects to foreign wikis. Also, it is not possible to link d:Q17995793 to als:Quantencomputer, since that page is already linked to d:Q176555.

I did link some wikis (in european languages with which I am not very unconfortable with) to existing redirects, but beyond that I do not think there is much we can do from here. It is up to those wikis to create the pages and add sitelinks if they wish to. Jähmefyysikko (talk) 13:42, 6 April 2023 (UTC)

Why is Skepticism buried so deep?

Given that Quantum Computing is far from producing useful results (aside from deliberately far-fetched demos), why is the material on Skepticism buried so deep under engineering? I would think that most of the readers aren't as interested in the mathematical formalisms as in whether QC is living up to the hype. Igor Markov 00:39, 26 June 2023 (UTC)

Thanks Igor Markov for your recent additions to the article! I agree that we should make these limitations more prominent, but I think that we should avoid a dedicated top-level criticism section. I have just added an introductory paragraph to § Engineering that gives some better context, and I think that we could include more balance throughout the article. For example, as soon as we introduce quantum search, we could mention that the quadratic speedup is unlikely to be useful anytime soon. I will try to trim the mathematical formalisms in § Quantum information processing, but I think that it would be better to mention algorithmic problems in § Algorithms and engineering challenges in § Engineering rather than having a separate uselessness section.  — Freoh 22:48, 26 June 2023 (UTC)

@Freoh, this makes sense, except that it's more work :) If you can make such changes, all the power to you.

- Igor Markov 23:10, 26 June 2023 (UTC)

References

Lead too long

The lead on this article is too long. To be more readable to a wider audience, it should be shorter.

I caution anyone against putting a simple "lead too long" template on the article (Template:Lead too long), because it will show the message at the top of the article to all readers, whether or not they have the ability or desire to change the article. However, here on the talk page this message may be more signal and less noise. 96.227.223.203 (talk) 02:28, 31 August 2023 (UTC)

Expression "exponentially faster" etc makes no sense!

I have corrected the wrongful use of the word "exponentially" in a couple of places where it was simply meant as an equivalent to "a lot". I am aware of this trend where "exponentially" has become a buzzword. But.. The word "exponential" is a well defined mathematical term (see article Exponential growth). Seeing expressions such as "expeonentially faster" etc in cheap tv-movies is one thing that one may shake one's head at and shrug. But, not least for reasons of clarity, I don't think this kind of buzzwording should occur in a technical Wikipedia article.

- RP Nielsen (talk) 13:02, 2 November 2023 (UTC)

I support your change as an improvement. But "exponentially" does have a legit common meaning as "a very lot" in addition to it's mathematical meaning. Sincerely, North8000 (talk) 14:26, 2 November 2023 (UTC)

I reverted the change, because the use of "exponential" there was not "wrongful" or equivalent to "a lot". The whole point of the best know quantum algorithms is that they provide an exponential speedup over the bets known (and, as conjectured, the best possible) classical algorithms. Here "exponential speedup" means that as a function of input size (in bits) the best classical algororithms take an exponentially growing number of steps, while the best quantum algorithms only need a polynomial number of steps. Shor's algorithm is the best know example, but there are many more, see Quantum Algorithm Zoo.--Qcomp (talk) 00:11, 3 November 2023 (UTC)

Well in that case perhaps that needs to be clarified, because the way it is presently articulated is confusing and misleading (as one is prone to think that "exponential" is used in the popular, 'urban' sense mentioned before), so I would say that at the very least those parts of the article need to be re-articulated, and special expressions or special use tof, need to be clarified. - The whole point of encyclopedic articles is to explain the topic to non-experts; in other words that one don't have to already be an expert who knows all the inside esoteric phrases, to be able to read the article.

- RP Nielsen (talk) 13:06, 3 November 2023 (UTC)

Its a pity that language has been so contaminated that we can no longer use well-defined terminology, but it's unfortunately true that "exponential" is often used in an almost meaningless way nowadays. I've added a footnote to the first use of "exponentially faster", to indicate that (and in which way) here we mean indeed "exponentially faster":

Here and in the following „exponentially faster” has a precise complexity theoretical meaning regarding the potential quantum speedup enabled by quantum computers. Namely, that for certain problems the worst-case time complexity (number of elementary steps needed to solve the problem) of the quantum algorithm scales superpolynomially better than the best known classical algorithm. It thus implies the conjecture that the set of problems efficiently solvable on a quantum computer (BQP) is strictly larger than the set of problems efficiently solvable on a classical computer (BPP). See, for example, Nielsen and Chuang, Quantum Computation and Quantum Information, chapter on quantum computational complexity or Scott Aaronson. "How Much Structure Is Needed for Huge Quantum Speedups?". Edited transcript of a rapporteur talk delivered at the 28th Solvay Physics Conference in Brussels on May 21, 2022. arXiv:2209.06930.

Qcomp (talk) 14:46, 3 November 2023 (UTC)

Thank you. Though I wonder if the point could not have been made in a much simpler and clearer way, by a formulation along the lines of how you expressed it in your previous reply, when you said:

Here "exponential speedup" means that as a function of input size (in bits) the best classical algororithms take an exponentially growing* number of steps, while the best quantum algorithms only need a polynomial number of steps.

*(and - as a sugestion- possibly with "exponentially growing" linking to the exponential growth article, just to make sure everyone is on board)

Apart from being simpler and much more digestible, this formulation also shows directly where it is the link to exponential growth enters the picture. - Of course it doesn't give you the opportunity to flex the muscle of how much you know about it, but on the upside it's easy and straight foreward for anyone to see what it means.

- RP Nielsen (talk) 07:10, 4 November 2023 (UTC)

I am glad that @Qcomp caught this issue quickly, but @RP Nielsen was not that far off in that the wording didn't make sense to non-experts. I added a parenthetical remark to point out that "exponentially faster" is with respect to input size scaling. Qq8 (talk) 10:31, 5 November 2023 (UTC)

Thanks again. But I'm afraid the parenthesis doesn't really clarify much about the legitimacy of the expression. And @Qcomp's footnote remains an unnecessarily extensive half-a-page of what, to the non-expert, appear to be more or less inconprehensible giberish. And I apologize in case if this my assessment might hit a sore nerve, but..

Keep in mind that IT'S A FOOTNOTE, the whole, and sole, point of which was to avoid confusion with the popular catchphrase. - It doesn't have to go in depth with complexity theory, or account for implications (all that is better done the article itself rather than in the footnote). - It's ONLY purpose was to clarify that the expression "exponentially faster" is an inside term, and (in a reasonably conceivable way) explain to the non-expert(!) what is meant by it, so as to avoid confusion with the popular catchphrase.

@Qcomp already did that excellently in his earlier reply to me, and so my suggestion was that THAT wording be used IN STEAD of the, as I said, unnecessarily extensive, and rather esoteric, half-a-page of insider giberish (I apologize again).

Thus, I suggest, for example, the following wording in stead:

Here and in the following „exponentially faster” has a precise complexity theoretical meaning. Namely, that as a function of input size (in bits) the best classical algororithms take an exponentially growing number of steps, while the best quantum algorithms only increase by a polynomial number of steps.

So I suggest this, or a similarly clear and to-the-point wording, to replace the footnote, althought I will abstain from making that edit myself, from fear of offening some if I did. So I'll leave that up to you guys. But that is how I would suggest the footnote be worded, in stead of what it is at present. But I leave it up to you guys to do that edit, if and in case you decide that what I have argued here makes any sense. RP Nielsen (talk) 19:19, 5 November 2023 (UTC)

On rereading the footnote I agree that it was unnecessary lengthy and detailed (an attempt to support that there's indeed an exponential speedup, and to be as precise as possible what speedup is claimed - I don't accept the qualification as "gibberish", though, especially since all technical terms were properly linked ;-), and I agree that most of that would rather belong in the complexity section or the quantum speedup article and I removed it from the footnote. --Qcomp (talk) 22:03, 5 November 2023 (UTC)

Just a sidebar...it's quite common and legit for there to be a common meaning of a term which is different than it's technical meaning. The latter does not make the former illegitimate. North8000 (talk) 14:27, 5 November 2023 (UTC)

Even with the footnote, I still disagree with the expression "exponentially faster". I think it also is not neutral due to how ambiguous it is. By ambiguous, I mean that from a reader's perspective, I would have thought you are claiming that there is an exponential speedup for ALL problems. For some problems however, this is definitely not the case. Erdabravest (talk) 00:39, 6 March 2024 (UTC)

Quantum Computing Overview

Hey everyone,

Would it be helpful to place more emphasis on why one would use quantum computing from the get go rather than starting from a hardware perspective?

The current beginning talks about the use of quantum computing in terms of quantum devices being operated but I would argue that the entire existence of quantum computing though is due to using quantum mechanics to better solve scalability problems that classical algorithms cannot manage.

From what I understand , if I were a reader, I would also risk "putting the word quantum in front of everything." Would it be fair to say this?

I am not sure if people agree with me on this. Thoughts? Erdabravest (talk) 00:56, 6 March 2024 (UTC)