Preprint manuscripts have been defined as “publication-ready research articles that have yet to undergo peer review and be formally published. They are free, openly accessible, and widely reusable under permissive copyright licenses. They accelerate research communication by putting dissemination under researcher control” (authorea.com/users/2/articles/169587-the-preprint-citation-bump). In some science fields, preprints of articles have become a standard form of pre-formal publication. In some cases, preprints are seen as a type of collaborative editing that allows a fellow re searcher to read and comment on papers. By doing so, the researcher can add more credibility or substance to those articles prior to formal journal submission. Recent research has implied that preprints “have a citation boost ranging from 83% to 269% and that preprints can in practice improve your manuscript and that can limit ‘scooping.’”
Back in 1991, preprints began to be treated as significant scholarly resources with the development of arXiv.org (pronounced “archive”). Begun to house electronic PDF versions of papers in the field of physics, it has grown to include the fields of astronomy, quantitative biology, computer science, mathematics, physics, quantitative finance, and statistics. Physicist Paul Ginsparg, the driving force behind the development of arXiv.org, notes that “this resource has been entirely scientist driven, and is flexible enough either to co-exist with the pre-existing publication system, or to help it evolve to something better optimized for researcher needs” (cs.cornell.edu/~ginsparg/physics/blurb/pg01unesco.html). It is widely credited as one of the major factors in the growth and development of the OA movement.
Although arXiv isn’t peer-reviewed, the system employs a group of moderators in each subject area who review, categorize, accept, or reject each submission. The system also has a somewhat controversial endorsement mechanism that works to prequalify authors included in the archive. Although the vast majority of preprints are eventually published, some authors—even highly regarded researchers such as Russian Grigori Perelman—have chosen to use the preprint system alone, skipping formal publication (arxiv.org/abs/math.DG/0211159).
A 2003 Nature Neuroscience article notes that “arXiv allows researchers to document their claims quickly, without waiting for journal publication, and it makes their findings freely available to anyone who may be interested” (doi:10.1038/nn0503-433; nature.com/neuro/journal/v6/n5/full/nn0503-433.html?foxtrotcallback=true).
According to Judy Luther’s article “The Stars Are Aligning for Preprints,” “The success of arXiv was due in part to the fact that the high energy physics community had been sharing their preprints via email prior to the launch of the server which solved the problem of clogged mail boxes. Researchers value the ability to preview papers and receive feedback on their work prior to formal publication. Analyses have shown that a high percentage of articles in arXiv subsequently appear in the Web of Science, confirming their publication in a formal journal with an impact factor” (scholarly kitchen.sspnet.org/2017/04/18/stars-aligning-preprints).
Open Science Framework’s OSF Preprints (osf.io/preprints), begun in December 2016, expands the arXiv concept into the humanities and social sciences. SocArXiv, PsyArXiv, and engrXiv form the base; however, law, education, business, architecture, and other fields are included. The OSF platform provides a solid structure that includes searching and sharing, as well as features that allow researchers to “design and manage their project workflow, data storage, DOIs and collaboration.”
Finding preprints of recent literature is still a work in progress. An additional concern to be addressed involves the lack of a system to register these various revised versions. “A surprising percent of submissions are revisions,” Luther notes. “SSRN indicates 40% and bioRxiv indicates 30%. Clearly identifying and preserving all versions may be an evolving topic for preprint repositories.” At present, however, Google Scholar is generally the only search option, and some DOI links are provided for various versions to make the search easier.
Since peer review (and further reflection) may change the very nature of research reports prior to formal publication, the lack of clear systems to connect these documents is a major unresolved issue.
DISCIPLINARY AND SUBJECT REPOSITORIES
A large number of repositories exist that take documents—preprints, post-prints, data, etc.—regardless of the organizational affiliation of the author. These allow for the development of stronger networks or communities of scholars and researchers. One of the first such disciplinary repositories was the University of Minnesota Libraries’ AgEcon (ageconsearch.umn.edu/?ln=en), begun in 1995 by two librarians hoping to provide better access to the grey literature in this field. The database has attracted support from the fields of “agribusiness, food security and supply, energy and natural resource economics, environmental economics, policy issues, international trade, and economic development” and includes “conference presentations, working papers, journal articles, government documents, and theses and dissertations” (age consearch.umn.edu/static/about.html).
Virtually every major research academic institution now maintains its own repository for the publications of faculty, students, and researchers of that institution. This includes conference presentations, theses and dissertations, class lectures, posters, data files, databases, post-prints of articles, and chapters of conference papers, as well as other information. Although these items may be included in Google searches, discovery across institutions remains difficult at best.
As noted by the Johns Hopkins University JScholarship repository (jscholarship.library.jhu.edu), these institutional repositories offer their communities “fast worldwide dis semination of your work, indexing on Google, Google Scholar, and other specialty academic search engines, increased visibility for research and teaching activities, full-text search capabilities, permanent unbreakable URLs and safe long-term archiving” (guides.library.jhu.edu/c.php?g=202599&p=1335149). Most of these are maintained by their institution’s libraries which see these “important tools for preserving an organization’s legacy … [facilitating] digital preservation and scholarly communication” as strategic to their institutional missions (repository.si.edu).
Recently, articles such as “Institutional Repositories as Infrastructures for Long-Term Preservation” have questioned the “sustainability of research publications and data” in institutional repositories (informationr.net/ir/22-2/paper757.html). The Coalition for Networked Information (CNI) just released a report which urges less of a focus on individual, library-centered repositories and instead on making these more institutional in operation and function, bringing more attention to standardizing systems and interoperability as we move to the future (cni.org/wp-content/uploads/2017/05/CNI-rethinking-irs-exec-rndtbl.report.S17.v1.pdf).
How these systems will evolve is still anyone’s guess. Other platforms, such a Mendeley (mendeley.com), the free refer ence manager and academic social networking system, and ORCID (orcid.org) are challenging these services as methods to connect researchers individually or in communities to their research work. If you have a known item—dissertation, paper, technical report, etc.—the easiest way to locate it is a title-in-quotes search in Google. It may not work, but this is still the best route to finding content quickly and avoiding the maze of individual repository search engines.
THE PRIVATE SECTOR ISN’T SNOOZING
SRRN, the acclaimed Social Science Research Network (ssrn.com/en), is one of the leading preprint services. Established more than 25 years ago, SSRN is “devoted to the rapid dissemination of scholarly research, beginning the social sciences and humanities and now including other fields to promote scientific discovery by circulating preprints of research and allowing others to comment at this prepublication stage with the goal of enhancing the quality of publications and creating informal networks of researchers.” Its success didn’t go unnoticed.
In May 2016, SRRN agreed to be acquired by Elsevier, which had earlier acquired Mendeley. In a press release, SRRN chairman Michael C. Jensen notes that, though initially concerned ab out the offer, “in evaluating our future in the evolving landscape, I came to believe that SSRN would benefit from being more interconnected and with the resources available from a larger organization. For example, there is scale in systems administration and security, and SSRN can provide more value to users with to more data and resources” (ssrn.com/en/index.cfm/ssrn-joins-mendeley-elsevier). The reaction among many academics was loud and negative. Would these efforts to circumvent the control of commercial publishers be upended?
In August 2017, Elsevier followed up by acquiring bepress (bepress.com), which was founded by academics in order to create an OA platform for journal publication that would topple the dominance of commercial publishers and bring publishing back to the control of the academy itself. bepress is working to assure its member organizations and authors that this was “part of a long-term move into the technology and services space for Elsevier. They have invested in bepress’s current model and have no interest in disrupting that.” Further, bepress managers quoted the company as promising that it “fully supports open access and the projects that the bepress community has already made so successful, for example through Digital Commons. We want to do all we can to continue to support institutions to showcase their research output and promote their value” (bepress.com/early-answers-questions-elsevier-transition).
“In a move entirely consistent with its strategy to pivot beyond content licensing to preprints, analytics, workflow, and decision-support,” Roger Schonfeld writes in a Scholarly Kitchen post, “Elsevier is now a major if not the foremost single player in the institutional repository landscape. If successful, and there are some risks, this acquisition will position Elsevier as an increasingly dominant player in preprints, continuing its march to adopt and coopt open access” (scholarlykitchen.sspnet.org/2017/08/02/elsevier-acquires-bepress).
In its press release, Elsevier also notes that “showcasing research is critical as competition increases among institutions to secure funding and attract faculty and students. By joining Elsevier, bepress will be better able to address institutions’ promotional needs, such as attracting students, faculty and grants, and preserving research data and outputs. Elsevier’s suite of research products, such as Scopus, Pure, SSRN and SciVal will enhance the breadth and quality of the reach, promotion and impact services bepress delivers to its customers” (elsevier.com/about/press-releases/corporate/elsevier-acquires-bepress,-a-leading-service-provider-used-by-academic-institutions-to-showcase-their-research).
MAKING RESEARCH AVAILABLE TO EVERYONE
Commercial publishers can hardly be expected to ignore this growing area—and sea change in academic publishing. Will they begin to adopt new, more balanced collaborative partnerships with academe and become more responsive to fiscal realities and new opportunities being created by technology? This time around, the academy has a better bargaining base and the experience to know that it has the ability to do so.
Today, funding agencies, universities, and other organizations are requiring transparency and OA. Researchers have clearly seen the value of greater visibility of their research and the impact on their reputations and future opportunities. We are far from having perfect access, but we have never had so much information available across the disciplines—if only we can find it.