"Bike-shedding" is a phenomenon where there is excessive discussion of a topic because it seems easy, so everyone feels that they are competent to discuss it. We seem to have a couple of recent examples:

  1. How do I experimentally measure the surface area of a rock?

  2. Are we really moving at the speed of light in the time dimension?

In example 1, there is a good, reasonable question (I upvoted), and a couple of highly upvoted answers that state opposite points of view -- followed by another 20 answers.

In example 2, a question is asked that duplicates a previous question. There are currently 2 close votes, 6 answers, and 22 upvotes for an answer that IMO is very poor and would have been ignored if posted now as an answer to the original question. Buried among these answers to the duplicate question, but with only a few upvotes, is a nice answer from an expert in the field (author of a popularization on the subject) which contradicts the 22-times-upvoted answer and which would have been a nice answer to the original question.

Is there anything we can do to reduce this kind of thing? It's a very negative thing for people visiting the site if they have to sort through all of this to try to find a reliable answer.

The only thing that occurs to me right now as a solution would be that possibly activity on duplicate questions could be kept from spinning out of control if votes to close were somehow more prominently displayed -- but this would I guess require a change to the SE software.

We also have the mechanism where people with a lot of rep can single-handedly close a question as a duplicate. But this is not used much, both because people with this power exercise self-restraint and because it's highly tag-specific (e.g., I have the power for general-relativity, but not for special-relativity).

  • 7
    $\begingroup$ Didn’t both questions attract attention by being on the Hot Network Questions list? If it is possible for an SE site to opt out of “participation” in the HNQ, I suggest PSE do so. $\endgroup$
    – G. Smith
    Commented Nov 12, 2019 at 1:44
  • 12
    $\begingroup$ @G.Smith I disagree. I think it helps get people into reading about physics who might not always do so. Sure, we get some bad answers, but I think that outweighs the thousands of views and visitors to the site. Being against HNQ makes physics seem like a lofty subject only for the experts, which I feel like is what some people think already. $\endgroup$ Commented Nov 12, 2019 at 1:49
  • 9
    $\begingroup$ @AaronStevens IMO, HNQ completely ruins the experience of normal site users by broadcasting to non-users what are generally sub-par questions. i'd be content with PSE opting out of it, but i'd rather the whole thing just disappear in its entirety. $\endgroup$
    – Kyle Kanos
    Commented Nov 12, 2019 at 3:23
  • 1
    $\begingroup$ @KyleKanos Yeah, it's not ideal sometimes, but does it really completely ruin the experience when a sub-par question makes HNQ? I would think normal users would have enough experience to know that sometimes happens and to just move on. At worst it's a nuisance and at best it gets more people to view other questions and learn more physics. Plus not all HNQ questions are poor. $\endgroup$ Commented Nov 12, 2019 at 4:17
  • $\begingroup$ I am the asker of the first question. Quite honestly, I was watching for answers the whole time, and cannot quite pick one answer out of all as a correct answer, since each one has it's own advantages and disadvantages. I was considering creating a community wiki answer that tabulates all the existing answer methods into one single post, which is perhaps easier to access. Would that be preferable? $\endgroup$ Commented Nov 12, 2019 at 9:22
  • 4
    $\begingroup$ @PrittBalagopal Not being able to choose between multiple, equally-acceptable answers is part of why subjective questions and questions looking for lists of things are not always a good fit for the site format. If there is not a single, best possible answer then a lot of the mechanics of the site breakdown. $\endgroup$
    – tpg2114
    Commented Nov 12, 2019 at 10:40
  • 1
    $\begingroup$ @tpg2114 I understand now, looking at the answers. However, when I posted the question, I initially did believe I would get maybe one or two methods to solve my issue. I had no idea that this would occur. $\endgroup$ Commented Nov 12, 2019 at 10:42
  • 7
    $\begingroup$ I think (not entirely sure) that opting out of HNQ might actually be an option for the site. Someone could always start a meta post to see what the community thinks, and if there's support for the idea, the mods could take it up with the higher-ups and see what our options are. $\endgroup$
    – David Z
    Commented Nov 12, 2019 at 11:45
  • 2
    $\begingroup$ I disagree that #1 is a "good" question. It's an interesting question, but a good answer needs to be very specific to what the OP really wants to measure. Think of the difference if the "rock" is a lump of pumice or a smooth hard pebble from a river bed, for example. $\endgroup$
    – alephzero
    Commented Nov 12, 2019 at 14:50
  • $\begingroup$ @DavidZ Or is it better to just flag questions that should be removed from HNQ? As discussed here $\endgroup$ Commented Nov 12, 2019 at 15:58
  • 2
    $\begingroup$ @AaronStevens yes, it skews what non-site regulars perceive as "on topic" and "the type of questions we welcome/expect" for the site on the garbage end of things. so then these people think we have/want nothing but garbage and post more garbage (seen it happen a number of times). perhaps a "Trending on this site" option would be better, but even that probably would skew towards crap $\endgroup$
    – Kyle Kanos
    Commented Nov 13, 2019 at 1:10
  • $\begingroup$ @KyleKanos If there is a way to show that most of HNQ is garbage then I think that would be a sufficient case to get Physics SE off of it. $\endgroup$ Commented Nov 13, 2019 at 2:01
  • $\begingroup$ @AaronStevens I know it when I see it, but I cannot quantify it in any reasonable means. $\endgroup$
    – Kyle Kanos
    Commented Nov 13, 2019 at 2:24

2 Answers 2


I don't think there is much you can do beyond what you have discussed and what is already in place. You just have to trust the users looking at the question to be responsible to look for duplicates first (this one is hard to trust at times), give good answers, vote appropriately, and go through the close vote queues to agree on duplicate votes already cast. People are flawed, so this is not always ideal (as you have shown), but it is what we have.

Something that you did not mention that helps with HNQs garnering many answers is protecting questions (which has been done for the surface area question). This does not prevent many or bad answers from being posted still, but I think it helps.

Additionally, if you have found better answers to questions that are duplicates you could explicitly mention these in a comment on the new question. That at least brings further attention to them.


I think the situation you mention is best described by the idiom "If you can't beat it, lead it". People who would be able to find a duplicate, or downvote a poor answer, are confronted with a crowd of new users coming from the HNQ link, who typically have 101 rep and for this reason cannot downvote or VTC, even if they agree. All they can do is upvote.

This kind of voting asymmetry leads to the only possible behaviour that has a visible impact and is rewarded with reputation - write your own answer, no matter how poor or controversial it is. Personally I came to the conclusion that unless this voting asymmetry is somehow fixed (by either forbidding 101 rep users to vote at all, or letting them downvote), any other measure will be akin to symptomatic treatment at best.

To clarify: I didn't mean that "If you can't beat it, lead it" should be everyone's modus operandi. I'm saying that someone who takes time and effort to search for duplicates and downvote bad answers is penalized with rep loss and resentful comments, whereas someone who adds yet another answer to the lot (good or otherwise) is rewarded. Bad answers won't go anywhere until that changes.


You must log in to answer this question.