6
$\begingroup$

I recently included a few paper references in my answers (in the MCMC convergence question, for instance). Should we include footnotes with the full citation when we reference papers? That might be hard to enforce, but it would certainly be a good thing to do in my opinion.

| |
$\endgroup$
8
$\begingroup$

To provide a complete reference while minimizing the amount of work:

  • For papers that have a DOI, always link to http://dx.doi.org/thisisthedoi. Otherwise, link to the publisher's page for the paper, or some other source that provides a full citation and (if available) a link to the PDF.
  • Clicking on the link will give the reader a full citation, so providing the full citation (author, title, journal, etc.) in the answer should not be required.
  • Optionally, add a second link with the text “(PDF)”. This link should be to a free verison.
  • We shouldn't expect the answerer to dig up a link to a free PDF. Given a full citation, it is not too hard for the reader to Google for that himself if he doesn't have access to the publisher's version through his institution's library. It's ok to offer a link to a free PDF in a comment.
  • Incorporate the link into the answer informally, in whatever way makes sense.
  • If you want to include a full citation, feel free, but I don't think it's necessary to prescribe a format.

Examples of fine ways to incorporate links:

| |
$\endgroup$
  • $\begingroup$ I agree, the default should always be if a paper has a DOI, link to that. $\endgroup$ – csgillespie Jul 27 '10 at 14:50
2
$\begingroup$

I would prefer that the cite should ideally be a link to a free version (e.g., on SSRN, arxiv, on author website etc) and if a free version is not available then link to the publisher version.

| |
$\endgroup$
  • $\begingroup$ I can agree on that; I just got used too much to VPN. $\endgroup$ – user88 Jul 23 '10 at 17:30
  • 1
    $\begingroup$ I would always suggest to use the DOI instead of the direct link to the publisher. Of course free versions are (imho) always preferable. $\endgroup$ – Peter Smit Jul 23 '10 at 19:04
2
$\begingroup$

I don't think it should be enforced, but we can indeed encourage people to do it. When we start with some unified format, people will quickly pick that up. And high-rep users / moderators can always edit it in.

I would suggest the following guidelines:

  • Using the Parenthesis Name Year format as proposed in (Smit, P., 2010), immediate linked to either doi or free pdf
  • Putting the actual references in sup tags (hopefully small tags in future, vote it up!)
  • Including at least the doi link, and if available a link to a free pdf.

Smit, P. (2010) Including references in answers Meta Stats Stackexchange, doi, free link

| |
$\endgroup$
  • 2
    $\begingroup$ It's a lot of work... people pick things up, but (as in Metropolis method) rather those which make them work less. $\endgroup$ – user88 Jul 23 '10 at 17:28
1
$\begingroup$

Whatever we agree on should go into the FAQ. I support the use of DOI links where available. If not, I think we should encourage links to a standard open access repository such as ArXiv, RePEc or SSRN. As a last resort, link to a personal website.

| |
$\endgroup$
0
$\begingroup$

I vote for URLs to the publisher site. This will work in a majority of cases, and gives a fast access to the paper. I just hate copying references to Google.

| |
$\endgroup$

You must log in to answer this question.

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