We occasionally get "find me a PDF" requests, where a user knows exactly the paper she needs and (presumably because of a paywall) asks how to obtain an electronic copy of it. This has typically been considered off-topic, and I agree that a site built around how to get behind paywalls is not a good thing to have.

On the other hand, this question of mine, which asked the former Theoretical Physics site about how one might obtain any copies of a particularly hard-to-obtain paper, has been put on hold as off-topic for this reason. I argued in the comments, and I repeat it here, that this is a question that requires an good knowledge of the physics literature and that future questions like this should definitely be allowed. If I am desperate to find a given paper, have the exact reference, and have looked everywhere, then I would be glad to be able to voice it in a gathering of physicists. I feel banning questions like this would devalue this site.

However, if we decide to allow this, then we also need to define where to draw the line. If you (dis)agree on the general principle, (down)up-vote the main answer to this question, and voice your ideas for appropriate criteria!

Return to FAQ index


Edited the tag wiki to add

This is also for requests for where to find a resource when it has been demonstrated that that pdf has been hard to find.

Question reopened (unless anyone has any objections), since it follows this.

I think we need to follow a few caveats:

  • One per question. Don't dump your reading list and expect Physics.SE to gopher it for you.
  • The paper must demonstrably be hard to find (you need to include more details on that in the question -- "it's in journal X which is out of print" or some such1). In other words, asking for a pdf of the EPR paper is not OK

Finally, we need to work out what happens with paywalls. Should we allow requests that ask for de-paywalling of an obscure paper? Or is it OK to link to the paywal landing page and be done with it?

IMO we shouldn't allow requests that ask for de-paywalling, however a request that asks for an obscure artcle which eventually is found hiding behind a paywall is OK as long as the answers link to the paywall.

I think requests for a book/article in a nearby library should not be allowed as it's just too localized. There's a vary slim chance that someone else from that locality will be looking for the same resource.

1. Emilio's question doesn't demonstrate this thoroughly, but it's OK. The main point is that the paper should be hard to get, and there should be a bit explaining why.


Yes, we should allow some how-to-find-a-reference questions.

  • 2
    $\begingroup$ I think the issue at hand, though, is which ones? How should we (moderators and close voters) distinguish an on topic find-a-paper question from an off topic one? An answer that doesn't specify that is really not much help. $\endgroup$ – David Z Jul 2 '13 at 22:19
  • 2
    $\begingroup$ I sort of agree here. I have concerns similar to tpg, but overall I don't mind as long as they stay under control. Note: you may want to simultaneously draft a new tag excerpt in this answer. $\endgroup$ – Manishearth Jul 2 '13 at 22:37
  • $\begingroup$ @DavidZaslavsky I'm not sure we need or benefit from having a precisely defined policy about which reference request questions are allowed. (Although failure to google the phrase you are asking about should result in summary execution, IMHO.) The moderators have a certain amount of trust from the community, and should exercise their discretion. $\endgroup$ – user1504 Jul 13 '13 at 16:21
  • $\begingroup$ @user1504 that's just it, though, I don't think the community does trust us to exercise our discretion, based on how often we get called out for e.g. closing questions. The whole point of this question was to have a policy to point to when that happens to show that we're not abusing our power, just acting in accordance with what the community wants. $\endgroup$ – David Z Jul 13 '13 at 21:20
  • $\begingroup$ @DavidZaslavsky I suspect you'll find that having a policy is just an invitation for certain people to complain and play internet lawyer. This stuff is ultimately a judgement call. Anyways, count me as a vote that you guys are generally doing fine. $\endgroup$ – user1504 Jul 13 '13 at 22:37
  • $\begingroup$ @user1504 True, in some cases we can't win. :-( Thanks for the vote of confidence! $\endgroup$ – David Z Jul 13 '13 at 22:38
  • $\begingroup$ Comment to the answer (v2): What does the weasel word 'some' mean in practical terms? $\endgroup$ – Qmechanic May 31 '16 at 13:30
  • $\begingroup$ @Qmechanic I think that (i) Manishearth's (accepted) answer does a fine job of specifying well what is and is not on-topic, (ii) this answer is useful in showing that the community does feel that there needs to be at least some leeway for questions of this sort, and (iii) this issue has not really been a particular problem over the three years since this thread was active, so it's probably fine as is. $\endgroup$ – Emilio Pisanty May 31 '16 at 13:42

Let me just comment on a very limited, but possibly important aspect of this issue. I don't think we should mess with copyright, but articles that are behind the paywall often have copies on the Internet that are posted legally: a lot of journals allow the authors to post copies (sometimes slightly modified) of their articles at the author's site, at the author's institution's site, or in the arxiv. For example, there are a couple of my articles at my personal site, and they are posted there in compliance with the policies of the journals. I guess (requests for) links to such copies should be allowed at Physics.SE.

  • 1
    $\begingroup$ I should (much to my chagrin) disagree with this. Such copies are almost always found within a google search of the article's title. Encouraging requests for such links could flood this site in questions that are either trivial (e.g. giyf trivial) or off-topic (e.g. no legal free copy exists). $\endgroup$ – Emilio Pisanty Jul 13 '13 at 12:50

I think "how do I find this paper," no matter the subject, probably has very similar answers. And most likely, it would be "Go to a university library and talk to the librarian."

I know in the US at most institutes, one could get any paper desired through the library. If they don't have it electronically or in print, through the Inter-Library Loan system, they'll get a copy from another library that does.

So I would argue that even if the paper is about physics, from a physics journal, etc., this is really a question better suited elsewhere. I would think something like that would be at home on academia.SE, provided it's more general than "How do I get the paper titled X" as that's too localized on any site really.

  • 1
    $\begingroup$ This is not necessarily the case even in the US. Papers from Soviet-era Russia, for example, can be hard to obtain and I imagine they're particularly so in the US. $\endgroup$ – Emilio Pisanty Jul 2 '13 at 17:55
  • 1
    $\begingroup$ I don't think that changes the point I was making -- Physics.SE isn't the place for such questions in my mind. It's more generic than that and is better suited elsewhere. $\endgroup$ – tpg2114 Jul 2 '13 at 18:07
  • 1
    $\begingroup$ Nope, reference requests are allowed according to this meta questions. I am not aware that the community of physicists has agreed on overthrowing this... $\endgroup$ – Dilaton Jul 2 '13 at 21:31
  • 2
    $\begingroup$ It's not a reference-request. $\endgroup$ – David Z Jul 2 '13 at 22:15

This answer contains the discussion in the comments to the question that sparked this discussion, which was originally posted at the (now defunct) Theoretical Physics site. I have moved them here because they belong to this discussion more than they do to that question. Comment vote counts, at the time of transfer, are in brackets.

Requests to find a PDF or print copy of a paper when the exact citation information is already known are off topic. – David Z♦ Jul 1 at 20:11. (1)

...apparently custom close reasons get posted as comments. Anyway, we've generally held "find me a PDF" and similar requests to be off topic, but that could be opened up for discussion, if people think questions like this should be on topic. – David Z♦ Jul 1 at 20:15

@DavidZaslavsky I have to disagree, strongly, for this case. This paper is outstandingly hard to find in either print or electronic copies and it was only through this question that I found one (unlike many people that cite it (!), I suspect). While I would generally think "find me a pdf" questions to be off-topic, some such requests can, would and do require an in-depth knowledge of the physics literature, and there are very precious few places on the internet to find physicists that can answer them. – Emilio Pisanty Jul 2 at 0:12. (4)

While it is not usually used as a criterion on this site, I often find MathOverflow's criterion "would I ask this at a seminar?" to also work well here. In this case, my answer would be an emphatic yes: if I really need a paper, have looked for ages and can't find it, and really need a physicist to point out alternatives, a physics seminar is about the first place I'd ask. (I did.) – Emilio Pisanty Jul 2 at 0:18. (3)

If this site disallows questions like that then it risks failing people who put in effort to write answers when they have a question, and we really should be moving in the opposite direction to that, fostering high-quality questions and keeping our good users. – Emilio Pisanty Jul 2 at 0:22. (3)

@Emilio (3 comments up) yeah, I was thinking along the same lines myself. If we want to allow some questions of this sort, we (the community) can work out a guideline for what is allowed and what isn't. But as of now, that guideline, which would presumably allow this particular question, isn't in place. – David Z♦ Jul 2 at 0:54. (1)

I have asked the relevant question on meta. – Emilio Pisanty Jul 2 at 14:42. (2)


You must log in to answer this question.

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