24
$\begingroup$

The Stack Overflow faq says:

We feel the best Stack Overflow questions have a bit of source code

I think the Physics Stack Exchange faq should say:

We feel the best Physics Stack Exchange answers include external citations.

I have seen more than one completely incorrect answer accepted on Physics Stack Exchange, which I have not seen on Stack Overflow. I suspect this is because unlike coding, it's hard to immediately verify physics claims. For this reason, I think external citations to trusted sources are especially important for physics answers.

I would also like to see more answers clearly indicate their logical foundation. Is your answer...

  1. Original research?
  2. Predicted by peer-reviewed theory, but unverified?
  3. Indirectly experimentally verified?
  4. Directly experimentally verified?

Don't make us guess!

I have seen many answers here based on rock-solid, universally trusted theory. This is great! However, if experimental verification has not actually been cited, this should be clear. Experiment frequently lags behind theory, but has been known to expose flaws and loopholes in theory, and very occasionally, invalidates theory entirely.

An example from my personal experience: Can I levitate an object without using an electromagnet?

For months, the accepted answer was basically "No, because of Earnshaw's theorem". This was an excellent answer; Earnshaw's theorem is rock-solid, universally accepted theory. End of story? No, because this answer was also good enough to include an external citation. As the external citation made clear, there are loopholes in Earnshaw's theorem that make the correct answer 'yes', not 'no'. The correct answer is supported by a picture of a levitating object - direct experimental verification.

$\endgroup$
6
  • 1
    $\begingroup$ Doesn't this limit questions and answers to people at institutions with a subscription to the various journals? $\endgroup$
    – Martin Beckett
    Mar 21, 2011 at 17:11
  • 1
    $\begingroup$ Excellent question, Martin. I should be clear that by 'citations' I don't mean just peer-reviewed journals. Wikipedia, Hyperphysics, and arXiv are excellent citations for many answers. When a journal article is appropriate, there may be free versions available. An example from my personal experience: I felt this question was best answered by an AJP article. I linked to it, and also to a free version of the same paper I found with Google Scholar. $\endgroup$
    – Andrew
    Mar 21, 2011 at 17:25
  • $\begingroup$ There is also a discussion about the level of questions in another thread. It would be a shame if answers in phsyics.so were limited to - "see App J vol/page/etc". $\endgroup$
    – Martin Beckett
    Mar 21, 2011 at 17:40
  • $\begingroup$ I agree completely. $\endgroup$
    – Andrew
    Mar 21, 2011 at 17:58
  • 1
    $\begingroup$ Suggestion to the question (v4): Replace the word citation with the word reference. Usually references and citations of a given post are two different things. A given post can include several references to outside articles. Conversely, outside articles can cite a given post. They become citations of the given post, but are obviously for chronological reasons not mentioned in the post. $\endgroup$
    – Qmechanic Mod
    Apr 15, 2013 at 18:13
  • $\begingroup$ What about stuff that you're 100% sure of (like really, 100%) but you don't remember where you got it from? You can't make Physics.SE a journal (for which this principle works) or wikipedia (which stupidly incorporated this principle). $\endgroup$
    – Abhimanyu Pallavi Sudhir
    Jul 28, 2013 at 5:11

3 Answers 3

Reset to default
13
$\begingroup$

Sure -- this should be de rigeur on our harder science sites.

Even on sites where the topic is much more subjective (SciFi, etc) we ask that opinions be backed up with facts and references.

http://blog.stackoverflow.com/2010/09/good-subjective-bad-subjective/

Great subjective questions insist that opinion be backed up with facts and references.

Opinion isn’t all bad, so long as it’s backed up with something other than “because I’m an expert”, or “because I said so”, or “just because”. Use your specific experiences to back up your opinions, as above, or point to some research you’ve done on the web or elsewhere that provides evidence to support your claims. We like you. We want to believe you. But like wikipedia itself, {{citation needed}}. And good subjective questions make this clear from the outset: back it up!

$\endgroup$
8
$\begingroup$

I agree with this approach quite strongly: i think that every answer that can be backed by further "evidence" (as in references ranging from Wikipedia to scholarly published articles) should include those — in fact, i wish people would recognize earlier and earlier how important this modus operandi really is: you build your work on top of previous work — ergo, the relevance of the use of citations, references, etc.

So, given Physics.SE's volume of answers, the rate of answers that have citations (of any sort) seems to me to be relatively low.

And, in all fairness, i think this is part of the reason of why the overall quality of answers is a bit below what some of us had expected. :-(

$\endgroup$
4
$\begingroup$

Well, I sort of agree with the principle, and would endorse fully it if we had higher level of questions, which would make this site generally more interesting. But, for questions that either ask for an overview of textbook material, or sorting out some fairly straightforward confusion of the OP (or a majority of the existing answers, as the case may be), this etiquette is probably not necessary. From my experience, you rarely need to use or cite any specific research papers or experimental results from, say the last couple of decades. I agree that for higher level questions which call for information like that a good answer should be clear on the source of the information, and the level of confidence one has in the answer. I really hope that at some stage a larger fraction of the questions would be at that level, where the worry you express is justified.

$\endgroup$
2
  • 6
    $\begingroup$ Right. I'm not going to start citing Newton or Maxwell to back up answers to most the questions I take on. $\endgroup$
    – Mark Eichenlaub
    Mar 21, 2011 at 6:41
  • 1
    $\begingroup$ For textbook-level questions, a link to Wikipedia or Hyperphysics might be helpful, and save writing too. I agree that some of the most basic answers boil down to "the following assumption in your question is wrong", and little more. $\endgroup$
    – Andrew
    Mar 21, 2011 at 12:24

You must log in to answer this question.

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