Extended definition
A preprint is the version of an academic manuscript deposited in an open repository before publication in a peer-reviewed journal — or alongside submission. The practice was born in physics with arXiv, created by Paul Ginsparg in 1991 at Los Alamos as an open repository for rapid pre-publication manuscript exchange (Ginsparg, 2011). Expansion to other fields was gradual: SSRN for social sciences (1994), RePEc for economics (1997), bioRxiv for biology (2013), medRxiv for medicine (2019), SciELO Preprints for Latin America (2018). Vale (2015) publicly advocated adoption in biology. Contemporary repositories assign a DOI to each deposited version, making preprints formally citable with URL persistence and Crossref integration. Later versions can replace the deposit (preserving history), reflecting revisions before or after peer review.
When it applies
Preprint is appropriate in any project that benefits from rapid dissemination — public health research during crises (massive use during COVID-19), areas with high competition for discovery priority, international collaborative projects needing a citable version before publication. It is also a central element in Open Science strategy and Plan S — green deposit with CC license can satisfy open access mandates without APCs. For early-career researchers, depositing a preprint establishes intellectual priority and visibility while manuscripts go through long review cycles (average 6-18 months in many fields).
When it does not apply
It does not apply in projects with contractual embargo clauses (industrial research, confidentiality contracts), nor in research involving patient data or sensitive intellectual property before formal protection. Some journals — minority, but existent — refuse manuscripts previously deposited as preprint (historical Ingelfinger rule, now in retreat). Checking target journal policy before depositing is reasonable conservative practice. It does not apply as a substitute for peer review: a preprint is not validated literature, and citing it in systematic reviews requires explicit caution about review status.
Applications by field
— Physics and mathematics: arXiv is the standard infrastructure; many articles are read preferentially as preprints, not as journal versions. — Biology and biomedical sciences: bioRxiv and medRxiv with growing adoption, especially in public health and genomics. — Computer science: arXiv covers most submissions to top-tier conferences; the community routinely reads preprints. — Social sciences: SSRN and SocArXiv with variable adoption; humanities have lower penetration.
Common pitfalls
The first pitfall is treating preprints as “publications” in CVs — in rigorous evaluation systems, only peer-reviewed work counts fully; preprints count as work in progress. The second is citing preprints without indicating review status: in systematic reviews, distinguishing preprints from validated literature is critical. The third is depositing a finished version and never updating it to the final post-review version — good practice is to deposit the published version (PMR — preferred manuscript record) with a distinct DOI. The fourth is confusing the preprint DOI with the published-version DOI — they are distinct Crossref entities. The fifth is depositing work with serious errors before minimal internal review: a preprint is not a private draft; it is gray literature, accessible and citable from the moment of deposit.