Bourne, Polka, Vale, and Kiley (2017) define preprint:
"Typically, a preprint is a research article, editorial, review, etc. that is ready to be submitted to a journal for peer review or is under review. It could also be a commentary, a report of negative results, a large data set and its description, and more. Finally, it could also be a paper that has been peer reviewed and either is awaiting formal publication by a journal or was rejected, but the authors are willing to make the content public. In short, a preprint is a research output that has not completed a typical publication pipeline but is of value to the community and deserving of being easily discovered and accessed" (p. 1).
A preprint server is an online platform for archiving and distributing preprints.
While many journals do not consider the posting of a preprint as prior publication (Bourne, Polka, Vale, & Kiley, 2017), authors should confirm with the journal they hope to publish in whether posting an manuscript first in a preprint server is allowed.
The standard form of peer review, happening before an article or book is published.
Often paired with open peer review, post-publication peer review occurs after a work has been disseminated by a publisher. Subsequent revised versions based on the peer reviews may then supersede the first version (F1000Research, n.d.).
The standard form of peer review (single- or double-blind) in which reviewer identities are withheld from the author and not disclosed in the published work (Watson, 2015).
Open peer review does not have a standard definition, but can be considered to encompass a range of models including any of the following characteristics: "making reviewer and author identities open, publishing review reports[,] and enabling greater participation in the peer review process" (Ross-Hellauer, 2017, p.3).
Open peer review is a facet of open science.
F1000Research is one such publishing platform that uses open, post-publication peer review, and traditional publishers as well may have open peer review options:
Peer review serves as a gatekeeper for what scholarly information is out there. In some fields, such as medicine, this can be positive by limiting the public's misinterpretation of preliminary findings (Basken, 2018). In others, however, it may be oppressive and function more to maintain consensus (Fitzpatrick, 2011). Gatekeeping through peer review and the editorial process can also be unreliable and inconsistent (Ross-Hellauer, 2017). For example, the frequency of posts on Retraction Watch illustrates how peer review does not catch all faulty science, whether simple errors or research misconduct.
Articles expressing controversial views raise the questions of what views peer review should allow through, under what circumstances articles should be retracted, and whether rebuttal is a better response than retraction (Flaherty, 2017).
Peer review and the traditional publishing process doesn't happen quickly, meaning the published record may lag behind what is actually happening in a field. Long peer review times can create stress, frustration, and fatigue for researchers (Nguyen et al., 2015).
Institutions often use how many peer-reviewed publications academics have as outsourced credentialing of those academics (Fitzpatrick, 2011). This can overemphasize quantity of scholarship over quality, disincentivizing scholars from taking risks.
In the fall of 2017, a controversial "Viewpoint" essay, "The Case for Colonialism," was published in the journal Third World Quarterly. Though the article made it through the editorial process, including double-blind peer review, the essay provoked widespread outrage that such a defense of colonialism could make it through to publication. Members of the editorial board resigned, a petition called for its retraction, the author eventually asked for it to be removed--and though the publisher, Taylor & Francis, initially decided to keep the article up--death threats to the journal's editor led them to withdraw it.
This case raises the questions of what should make it through peer review, whether controversial views should be rebutted or retracted, and how peer review in a larger (and political) sense affects the information landscape.
The links below chronicle the debate.
The major open access publisher Public Library of Science (PLOS) will now automatically upload research papers submitted to its journals to the preprint server bioRxiv, after being screened to make sure they meet bioRxiv standards (McKenzie, 2018). Authors can opt out if they wish.
Ross-Hellauer's (2017) systematic review of open peer review makes it clear it is a model increasingly discussed, but still in need of further research to determine its effectiveness at addressing the problems of traditional peer review.
David Crotty (2010) of Oxford University Press notes that transitioning systems entirely to post-publication peer review could lead to near-overwhelming information overload, with volume up and average quality down.
Fitzpatrick (2011) argues for the importance of some kind of system to review the reviewers, to improve the integrity and value of peer review. Tennant et al. (2017) discuss several models based on non-academic sites such as Reddit and GitHub.
Walker and Rocha da Silva (2015) predict continued diversification of peer review models to meet the varying needs among different disciplines, publishers, authors, and readers.