5
$\begingroup$

A lot of sites on Stack Exchange require - or at least strongly advise - users to include sources in their answers (and questions, when possible). Should we try to move in that direction? Requiring that answers provide references at this stage could restrict the flow a little, but it could help improve answer quality (not that we've got bad answers).

As a follow-up: What kinds of sources should we aim for? Peer-reviewed articles? Wikipedia?

$\endgroup$
6
  • 2
    $\begingroup$ This is a good question. The problem is some of the subject matter will be based on personal experience. How can someone reference something that thy learned over the course of 30-years worth of experience? Do we expect someone to find a token source just to link something when they knew the answer $\endgroup$
    – psubsee2003
    Commented Jan 21, 2015 at 0:28
  • $\begingroup$ @psubsee2003 I'd love an expert's firsthand account as an answer to a question of mine! We shouldn't require people to have "token sources". The issue is that we can't confirm if someone is truly an expert. I'm not saying that people will outright lie, but it could be hard to prove how much they know. $\endgroup$
    – HDE 226868
    Commented Jan 21, 2015 at 0:33
  • 3
    $\begingroup$ I would think that we will have questions where "In my experience, process X works better than Y for Z because reasons" is a valid answer. But there will be other questions about specific facts where sources must be provided, IMO. It's important for the engineers asking questions to verify any information they get, but it's also the responsibility of the engineers providing the information to give solid info. $\endgroup$
    – Matthew Read
    Commented Jan 21, 2015 at 0:53
  • 1
    $\begingroup$ The voting system might be the best way to decide this; if it is a question where references improve the quality of the answer, then the user that provides references will get the most up-votes and be rewarded. If the references aren't any good, or don't add to the answer, then users are unlikely to upvote and the arbitrary reference'r will be discouraged. $\endgroup$
    – thomasmichaelwallace
    Commented Jan 21, 2015 at 10:50
  • $\begingroup$ You raise a good question and I suspect our community will have to revisit this question as time passes and the breadth of the questions grow. $\endgroup$
    – user16
    Commented Jan 22, 2015 at 1:10
  • $\begingroup$ It's also a matter of culture. If we can establish a culture of material rich with sources, our attractiveness will surely be increased. $\endgroup$
    – NoDataDumpNoContribution
    Commented Feb 3, 2015 at 18:04

3 Answers 3

Reset to default
6
$\begingroup$

I don't think we need a special policy unique to our site. All other SE sites strongly encourage sources, we should be no different. They make for more complete, better answers, and will be voted up. Weak answers with no sources will be voted down to the bottom. The problem is solved by the very nature of the site.

Besides, if you blindly take advice from a stranger on the internet and things go wrong, you have no one to blame but yourself.

$\endgroup$
1
  • $\begingroup$ -1 for inviting whinging along the lines of "But the interwebs told me so!" Just kidding, it's a +1, and well said. Other sites with similar themes have done fine without having to have a special site policy. $\endgroup$
    – user16
    Commented Jan 22, 2015 at 1:10
2
$\begingroup$

I don't think that we should have a formal policy, but I think that good guidelines could be as follows:

  1. If the question is asking for a clear and unambiguous fact that could be found in a single source, then such a source should be listed.
  2. If the question relies on the combination of several sources, then those sources should be listed (with perhaps a brief note to how they were combined)
  3. If the question relies on a very large number of sources and/or professional experience/judgment then I think it should cite key sources and "professional experience."
  4. Regarding types of references:
    • Peer-reviewed sources would be best, of course! Unfortunately, most of those are locked behind paywalls so that will be difficult.
    • Some universities and other reputable sites have good information and I think that those would realistically make the best references.
    • I would only reference Wikipedia if there was no reputable reference and I had verified that the article was relevant and factually sound. (Can't do anything about future edits but I think that current-state is reasonable)
$\endgroup$
1
$\begingroup$

MOST answers should contain at least factual references (e.g. wikipedia). If they are advanced, they might contain more detailed, or nuanced, material from expert sources.

Questions should have references too, if they are detailed, or if they are about a specialized phenomenon.

The kinds of answers that don't need references are those that follow from engineering common sense: "If you drop an apple from a ten story building, it will fall with increasingly accelerative force until it hits the ground, where it will probably break up."

$\endgroup$

You must log in to answer this question.

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