This question is a special case of this one, which has no satisfying answer in my opinion.

I often encounter papers that have been published twice (or more), without any modification. An example is this one. First presented at the DLS'12 conference, and published in the proceedings. Then, the proceedings have been published in an issue of the ACM SIGPLAN. Notice that the paper is the same, same DOI, etc. However, there are two bibtex entries (one @inproceedings, one @article) which are similar, but not identical.

My question : in such cases, how should the paper be cited ? Using the @inproceedings entry or the @article ?

(Side question : Why is it the case that a single paper is published twice ?)

2 Answers 2


This may be a special feature of CS or not. We depend a lot on conferences for publishing work. The conferences usually publish proceedings. The first version is distributed at the conference to attendees and may have a limited run. It typically has the conference name on the cover. In the case cited, SIGPLAN, as is typical, will put a separate cover on the proceedings, turn it into a (regular) issue of the Notices and distribute that to all members whether they attend or not. It is really a service to the community, since not everyone can attend an important conference. It is possible that corrections are made between the two versions, but I don't think that is common.

If they are identical, you can cite either one with confidence that readers needing to follow the citation will find the paper.

  • 2
    Cite the one you used in your work - if the other has been updated then that may cause confusion...
    – Solar Mike
    Commented Mar 14, 2019 at 12:13

Manuscripts should (generally) never be published twice. (Exceptions exist, e.g., a famous academic's works might be compiled into a single volume.) A manuscript may be published at a conference (@inproceedings) and then subsequently be published in a journal (@article). However, journal versions are required to extend conference versions. In such cases, you have several options:

  • Cite both
  • Cite the version where the relevant result first appeared
  • Cite the conference version and mention the journal version in the reference

You must log in to answer this question.

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