Use this tag for questions seeking a single specific paper or a short, non-open-ended list of references, like "What paper first discovered X?", "Where can I find the original derivation of X?", or "What is the canonical source for X?" etc. Also for requests for hard-to-find electronic copies of resources when the exact reference is known. NOT TO BE USED for generic requests for resources (use [resource-recommendation] for that).

This tag is for questions about identifying a single reference, or a short list of references, when the question contains enough detail to uniquely identify the desired source.

Most questions under this tag are asking for identification of a particularly distinctive paper, or papers. Good examples would include

  • Which paper originally reported the discovery of composite fermions?
  • Where can I find the original derivation of the the Schwarschild metric?
  • What is the canonical source for the LHC operating schedule?

In general, for questions, there is one correct answer, and all the references requested already exist at the time of the question being asked. A good answer to the question will give a complete list of the bibliographic information for the reference(s) being requested, or at least enough information to uniquely identify them. In some cases, the correct answer could change later, but then the new correct answer would invalidate the original correct answer. There shouldn't be an increase in the number of valid answers as time goes on. Contrast this with the tag, which is for questions that ask about a particular topic, not for a specific resource. The set of references on a topic will grow over time as people write new books and papers.

The tags and are mutually exclusive; they should never both be applied to the same question.

The other use of is finding an electronic copy of a reference which can already be uniquely identified, if the electronic copy is hard to locate. For example,

  • Is a copy of "Interaction with the absorber as the mechanism of radiation" by Feynman and Wheeler (Rev. Mod. Phys. 17: 157-181) available online?

These kinds of questions need to contain enough information to uniquely identify the reference.

We expect askers to have checked on arXiv and on the website of the publishing journal, and to have searched the web using the information provided in the question, before posting this type of question. (In particular, using and BASE can help uncover copies online.) If an electronic copy of the article is accessible through any of these sources, or through any other method the asker already knows about, it's not considered "hard to locate". Even if it costs money. (Even if it costs a lot of money. This is not a resource for getting free copies of paywalled articles.)