What is the best way to handle a case where a question was asked a long time ago and received several incorrect answers but no correct ones? To be more concrete, this was the situation with this question: If photons have no mass, how can they have momentum? I've done the obvious things: (1) submitted a correct answer; (2) downvoted the incorrect answers; and (3) explained my downvotes in comments. However, the question is 9 months old, and one of the incorrect answers has been accepted and heavily upvoted. I would interpret this as a situation arising from the fact that physics.SE is relatively new, and is still in the process of attracting a critical mass of users. With a small number of users, someone can make an honest mistake in an answer and not have it corrected. The problem is that there are certain questions that are FAQs, and those FAQs got asked and answered early on. Now we're stuck with them. Students or other people without solid expertise of their own will come along and believe the incorrect answer, which has been upvoted and accepted and effectively frozen in place.

This is not about crackpottery, as discussed in this question: Answers they are a-changin' I'm talking about answers that are wrong, but not wrong in a crackpottish way. I'm sure the folks who gave these answers are competent at physics, but everyone makes mistakes sometimes, even people who are highly competent.

How have the older, more mature SE sites dealt with this?

The problem is that once a question has sat around for 6 or 12 months, nobody is likely to participate in it anymore. It's off everyone's radar, and if its answers are broken, they aren't going to get fixed.

For comparison, Wikipedia has mechanisms where you can try to attract people's interest in improving an article that needs work. For instance, you can nominate it as a good article, which initiates a review process; you can tag it as needing attention from an expert; you can nominate it for deletion; etc. SE, on the other hand, doesn't seem to have any mechanism (that I know of) for getting up on a soapbox and alerting people that a certain question is broken and needs fixing.

  • 1
    $\begingroup$ "The problem is that once a question has sat around for 6 or 12 months, nobody is likely to participate in it anymore." Then is it reasonable to put on hold as duplicate a new question on the same topic. Actually I doubt it takes that long to be off the radar. Both incorrect and duplicate answers are a problem of consolidation, which is not really addressed by SE. $\endgroup$
    – babou
    Sep 5, 2013 at 17:14

3 Answers 3


SE, on the other hand, doesn't seem to have any mechanism (that I know of) for getting up on a soapbox and alerting people that a certain question is broken and needs fixing.

Sure it does! You're using it right now. :)

Options are:

  • provide an alternative correct(er) answer

  • gently point out the problems with the incorrect answer in comments, perhaps also followed by a downvote -- to be removed once the problems are corrected, naturally -- to adjust the sort order

  • raise the issue here on meta, as you have

  • (optionally) edit the incorrect answer, if possible, to correct it; this will also naturally bump the question.

  • (optionally) flag the post for moderator attention, but that is unlikely to generate the votes you really need to correct the situation; I would only resort to mod flagging if the answer is dangerously wrong.

  • (possibly) Bring up the issue on the chat. With a little luck this puts in touch with some other experienced users, and it certainly provides you a forum to go back and forth until you figure out the exact nature of the error (and who has made it). Alas the physics is chat is often fairly dead.

  • $\begingroup$ Hmm...so is it really OK to use meta just for complaining about bad answers to a specific question? It doesn't seem ... meta. $\endgroup$
    – user4552
    Aug 9, 2011 at 0:10

Here is an example--- the answer claims that a black hole which is less dense than water will not float in water.

How can super massive black holes have a lower density than water?


As an answer to this question, I would suggest giving a little more attention to the answer that is accepted and upvoted. Although it may seem obviously wrong to you, other people still see merit in it that you may or may not have addressed, and I think this question in particular is a good example of this general type of conundrum.

In terms of how I use physics SE, I quite often find myself referencing and talking about other answers. This is more of a personal thing, and honestly I don't think this site is the best format to have a really major discussion and hashing out of the details of a problem from two competing perspectives.

In the case of this photon mass problem, as someone with a seemingly conflicting physical perspective than you, I'll try to articulate what I find to be missing.

There is a general notion in what you write that concepts of gravity can not be applied to a photon's relativistic mass. This is exactly the question I asked previously here:

Explain how (or if) a box full of photons would weigh more due to massless photons

Now, if this discussion was off-topic then I would understand why we should limit the discussion to special relativity. However, go back to the question.

As an explanation of why a large gravitational field, such as a black hole, can bend light, I have heard that light has momentum. This is given as a solution to the problem of only massive objects being affected by gravity. However, momentum is the product of mass and velocity, so, by this definition, massless photons can not have momentum.

The question obviously shows interest in the relativistic mass of photons in the context of gravitational attraction, gravitational lensing, and that sort of thing. Someone who wants to treat photons as gravitational attractors such as the questioner in this case, or myself, will be unsatisfied by limiting the problem to special relativity. We want to know specifically why using $m= h / \lambda c$, and applying it in all other classical senses will screw things up. I mean, I can take the above, say $F=dp/dt$, then plug and chug into Newtonian gravity all day.

I go into this detail to show why I wouldn't even downvote the main answer, meaning I don't see what is wrong about it, in addition to the fact it's useful to what I want to know. Hopefully I'm also helping you to identify what burden of proof you have to show why some concept shouldn't be applied.

  • $\begingroup$ The accepted answer is simply incorrect, for the two reasons I gave in my comment there. Eric Zaslow, in his first comment, identified one of those reasons. $\endgroup$
    – user4552
    Aug 9, 2011 at 15:17
  • $\begingroup$ Tbe answer is correct, but using a term "relativistic mass" which you happen not to like. Mass is a scaler, "relativistic mass" is the time component of a vector, and is the integral of the time-time component of the stress energy tensor. $\endgroup$
    – Ron Maimon
    Aug 11, 2011 at 7:51
  • $\begingroup$ @Ron Maimon: I've replied to your comment in the physics thread. $\endgroup$
    – user4552
    Aug 14, 2011 at 23:28

You must log in to answer this question.