I want this site to succeed. I want a site that attracts experts. Real experts. Not people like me, but scientists with 25 years of experience in a sub-field of atmospheric science.

I'm seeing some answers that appear to be based on speculation rather than on scientific findings, or at least where the answer does not make apparent where the knowledge is coming from. I appreciate peoples efforts nevertheless, but I'm not sure if it will attract experts. Some examples: ((1), (2), ((3)). I don't like to discourage anyone in such an early phase, but... shall we introduce a policy that answers should be properly referenced? This is a science Q&A, after all!

  • $\begingroup$ Good suggestion. Citations should certainly be strongly encouraged. There may be situations where citations are not possible though, and exceptions should be made. Extraordinary claims should require at least some citation. $\endgroup$
    – naught101
    Apr 15, 2014 at 23:02
  • 1
    $\begingroup$ @naught101 Not only extraordinary claims, but also widely held but challengeable beliefs. Those are more at risk to seep through than extraordinary claims. $\endgroup$
    – gerrit Mod
    Apr 15, 2014 at 23:08
  • 1
    $\begingroup$ Citations should certainly be encouraged, but do note that no other site on the network besides Skeptics strictly requires citations in answers. $\endgroup$
    – senshin
    Apr 16, 2014 at 0:50
  • $\begingroup$ Is there a syntax that can be used to cite properly, e.g. embedding bibtex in an answer? $\endgroup$ Apr 16, 2014 at 12:55
  • $\begingroup$ Looks like there isn't a syntax for citations, even on skeptics. I feel THIS IS OUTRAGE! How are we supposed to have an "experts" discussion site without the ability to cite? This needs to be fixed before the site goes public. $\endgroup$ Apr 16, 2014 at 13:05
  • $\begingroup$ While I'm definitely all in favour of citations, It should be borne in mind that most scientific papers are inaccessible (paywalled) for anyone outside of an academic institution with institutional access. So citations are only as good as the availability of a PDF. $\endgroup$
    – Ben Brooks
    Apr 21, 2014 at 18:17
  • 1
    $\begingroup$ @BenBrooks Although an open access citation is advantageous, I think an authoritative citation to a closed access source (for-pay paper, book, etc.) is still much better than none, even if many people can only judge its contents from the abstract (which is often enough to get the main idea). $\endgroup$
    – gerrit Mod
    Apr 21, 2014 at 18:39
  • $\begingroup$ Ben Brooks, most people can use a public library to gain access. It is unfortunate that not all information is free, but that doesn't mean we can pretend it's not information. $\endgroup$ May 16, 2015 at 18:50

3 Answers 3


I believe it is impossible to do so for every post. Some of the questions posed are basic textbook level and a reference in that case would be to one of many text-books. It is clear as in all science that citing your sources is useful and necessary but clearly there is a fine line when it should be "common knowledge" and in such cases one can ask common to whom?

I suggest any response that lacks references, and where we deem such would be useful should be asked through a comment to add such references (rather than a quick down-vote) that way the answers can be built constructively. Also remember that answers are not set in stone. It is possible to add to the answers by editing and thus improve them. the ultimate goal is after all to have good answers to al questions.

  • $\begingroup$ I think that Wikipedia will usually contain such basic textbook-level information, and is a fine reference for basic questions (although the questioner should at least check over the citations on wikipedia). $\endgroup$
    – naught101
    Apr 16, 2014 at 7:52
  • $\begingroup$ I agree. I would like to see citations for numerical quantities, where given, and for anything that may be controversial - and as you suggest, this is a perfectly reasonable thing to ask for in comments. I would not want to discourage experts from answering things that they consider simple, because they would have to go looking for a reference to cite. $\endgroup$ Apr 16, 2014 at 12:19

No, I am not in favor of this. If I'm looking for citations, I don't come to stack exchange. I have 3 arguments against required citations:

  1. A lot of the reason this site doesn't attract more users (experts or otherwise) is because of the rigidity we've developed with what's acceptable for a Q or A, which trickles down to total site activity. By discouraging activity, we are reducing usage. But, if we voted/edited more often (both up and down), instead of restricting posts, we would get more users that want to join in. By making the rules more strict we will attract less users, votes will go down, and our posts will be graveyards.
  2. An expert can explain an answer to many Earth Science questions without a reference. If you want to give an expert a homework assignment, and tell them they have to go look up references for all the ideas they share, you will not attract more experts.
  3. There are many forums/pathways for experts to share information with other experts. Stack Exchange is for both experts and enthusiasts. We shouldn't make it harder for enthusiasts to join in.

While I agree that we don't want an opinion-based site, I think it's perfectly acceptable for users to post their best attempt at an answer whether they are an expert or not. If it's wrong, let's downvote it. If it's right, we should upvote it regardless of citations. However, if we make citations required and just vote even less on what little traffic we do have... we will go from a site with a few active users and some passersby to a site with very little activity.

Finally, let's not forget the power of the edit feature. If you think a post is lacking a reference, go put one in!

  • $\begingroup$ I agree with you. This post was motivated by Walsby. Without a spam issue, I think it is not necessary. If it is possible to quote better, but some students can't read the papers because they don't have a license. $\endgroup$
    – user20559
    Sep 20, 2021 at 2:17
  • 1
    $\begingroup$ @Universal_learner This is the more recent discussion we had after the "Walsby case": earthscience.meta.stackexchange.com/q/1818/18081 $\endgroup$ Sep 20, 2021 at 15:01

Yes, we should have citations on every answer. There are plenty of speculators and chancers in this family of subjects out there, and we'll do well to deter them.

And no, Wikipedia is not a reliable source. Any fool can edit Wikipedia to say what they want, when they want; and content disputes there sometimes get resolved by a loud obsessed fools over-ruling a few quietly-spoken experts who've got better things to do than spend hours arguing on talk pages and arb pages.

If it's a common text-book fact, then finding a citation should be easy.

And where possible please use DOIs when citing (rather than the publisher's deep linked page), via one of the long-standing reputable DOI linkers, such as dx.doi.org

  • 2
    $\begingroup$ -1: we're often compared to Wikipedia. I'm not saying that it's perfect, but, like SE, it usually works out. Also, most WP info is sourced, so you can verify that information. If an expert here sees something wrong, they can comment. Also, it it's a common knowledge thing, why should you have to spend ten minutes digging up a source for a one minute answer? $\endgroup$ Apr 16, 2014 at 21:15
  • $\begingroup$ @AnnonomusPerson and if people were to cite wikipedia, do you think they should verify that information first, or just use whatever's there? $\endgroup$
    – 410 gone
    Apr 17, 2014 at 7:30
  • $\begingroup$ They should verify the source by using the included references if they don't know for sure, or just Google that fact. If it's a textbook level fact that they know, and want a little extra explination, you can quote a small portion of text from Wikipedia, as long as they know the information is correct themselves. $\endgroup$ Apr 17, 2014 at 21:35

You must log in to answer this question.

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