The number of references to a scientific publication is frequently used as an objective measure for the significance of the publication. This metric is far less precise than it may appear and in the short/medium-term it certainly fails to capture the most visionary and creative research. Consider, for example, that the publications of Richard Feynman are currently cited 6-10 times more frequently than during the peak of Feynman’s career. The respective comparison for the citations to publications by Albert Einstein is even more extreme. Still, citations are used as an influential metric, and thus the community standards of what is citable are very influential as well.

The community standards should be set by a community-wide discussion. Here, I will express my opinion in hope of stimulating discussion and soliciting more opinions. To be citable, a publication must (i) be permanent and traceable (e.g., it must have a DOI), and (ii) provide scientific support for what it is being cited. The second criterion is loaded and needs clarification. By scientific support, I mean data, reasoning and theoretical/computational results that are verifiable and refutable. Crucially, this assessment of “scientific support” must be made by the authors referring to the work. Most magazines and journals do not publish the names of the editors and the peer-reviewers, or even the contents of the peer-reviews. This type of hidden assessment and the anonymous people involved in it cannot possibly assume responsibility for the scientific merits of what is published. The scientific merits of a paper and the extent to which it provides scientific support must be evaluated by the authors referring to it.

These two criteria of what is citable apply equally to traditional papers having undergone open/hidden peer-review and to preprints uploaded on permanent servers guaranteeing timestamps and traceability. In fact, the data suggest that some communities have long recognized and adopted these standards. To my delight, I saw much enthusiasm among broader communities who have not yet adopted them:

I am optimistic that scientists will embrace their duties of independent critical assessment of the publications they refer to. I would love to hear your thoughts and ideas on what should be the criteria and the community practices for citing scientific publications.