Abstract
In August 2016, the American Chemical Society (ACS) announced its plans to develop a chemistry preprint service, and that it was seeking collaborators as well as input from all stakeholders to ensure that the service met the needs of the chemical community [1]. A year later, nearly to the day, on 14 August 2017, the ACS and its collaborative partners, the Royal Society of Chemistry (RSC) and the German Chemical Society (GDCh), launched the beta version of ChemRxiv (pronounced ‘chem-archive’), as a non-profit, free service for chemists around the globe. All three partners are supplying financial support. But why chemistry and why now, more than a quarter of a century after the first preprint server, arXiv, was launched to serve the fields of physics, mathematics, astronomy, and computer science? (Coincidentally, arXiv was also launched on August 14th) [2]. To answer these questions, I spoke with Dr. Darla Henderson, ACS’ Assistant Director of Open Access Programs, who brought me up-to-date on the status of ChemRxiv as it approached its first birthday.
Why Now?
Henderson noted that while throughout the years ACS considered the possibility of developing a preprint server, they did not believe that there was a technology that could affordably maintain a sustainable service. More importantly, there was no urgent request coming from the chemistry community, despite that fact the preprint servers had been around for some time in other disciplines. The reality is that the publication process in chemistry is relatively fast. But in 2016 the issue was raised again at an ACS editors’ meeting and they decided to look into it. They found that younger authors, especially those at the interface of the life sciences, informatics, and computational sciences (fields in which preprint servers had already been established) were starting to raise the issue with ACS. Also, ACS editors themselves were using preprint servers for the purposes of collaboration and to foster discussion in their own research. It was also around this time in 2016 that a group of thirty research funding agencies, in response to the Ebola and Zika outbreaks, were demanding that pre-publication sharing of data and results during public health emergencies become the global norm [3]. It seemed as though the perfect storm was brewing that would validate the creation of a preprint server in chemistry, hence the ACS call for input and collaborators later that year.
Why Preprints?
The obvious answer to the question of “Why preprints?” is that it speeds up the dissemination of scientific research. But the question remains: in a field such as chemistry in which the publication process is relatively fast, is a preprint server really needed? Henderson said that the publishing process is fast from the publisher’s perspective, not necessarily from that of a young researcher who is in the early days of their career. They have to find a journal that will accept their manuscript and this can take a lot of time [4]. She noted that there are other advantages as well. Preprints provide visible evidence of scholarly credit which is important when looking for employment or seeking grants. In addition, broader engagement prior to publishing actually can improve the final manuscript, increasing one’s chance of publication.
Her words echoed similar sentiments that were voiced at the 2018 Annual Conference of the National Federation of Advanced Information Services (NFAIS) that was held in Alexandria, VA in February 2018. Shirley Decker-Lucke, the Publishing Director of Elsevier’s SSRN, a pre-print server launched in 1994 primarily for research in the social sciences and economics (SSRN has just expanded into chemistry and biology) said that there are a lot of benefits for authors when they share their early stage research via preprints. Their research is quickly disseminated globally and this can lead to feedback from and collaboration with other researchers. It demonstrates their productivity and independence while showcasing their scholarly output and research accomplishments. It allows them to claim priority over their discoveries and provides a vehicle for the sharing of research results not suitable for traditional journal publishing [5]. Another speaker, Dr. Joris van Rossum, Director of Special Projects at Digital Science, said that preprints are assuming what had previously been a traditional role of publishers—that of registration, establishing an author’s precedence and ownership of an idea [6].
There is another perspective on preprints that must be considered beyond those of publishers, editors, and authors and that is the perspective of the funding community as noted earlier. A number of funders have come out in support of preprints as a form of Open Access to research. The Chan Zuckerberg Science Initiative, funders of research that is made openly-accessible, provides support to and collaborates with bioRxiv, the preprint service for the life sciences. The U.S. National Institutes of Health (NIH) believe that what they term “interim research products” (preprints and early registration of research protocols) can not only speed the dissemination of science, but also enhance its rigor. As of March 2017, NIH guidelines state that interim research products can now be cited anywhere research products (full articles) are cited as long as DOI’s are included to provide permanent access to the document [7]. Neil Thaker, who was the Special Assistant to the NIH Deputy Director for Extramural Research when he spoke at the NFAIS conference earlier this year, said that they surveyed authors, scientists, publishers, etc. when they were in the process of changing the guidelines and found that there was overwhelming support for the acceptance of preprints [8]. The major concern expressed was the lack of peer review of preprints and so NIH makes that clear in the guidelines to grant reviewers. He added that while NIH is neutral on whether or not reviewers should read references that are included in grant proposals, he found that 90 % of the reviewers do look at what is cited in a grant.
Looking at the growth in the number of preprint servers since early 2016 it is quite possible that it is the widespread acceptance by the funding community that may have given new life to the use of preprint servers in scholarly communication. Just take a look at the growth since arXiv was launched in 1991 (see graph previous page).
There was a total of five preprint servers before the year 2000. Two more were added between 2000 and 2009 and two in 2013. But then growth accelerated with five preprint servers being launched in 2016, thirteen launched in 2016, and four already launched or planned for launch in 2018. Shirley Decker-Lucke contests that part of the growth is definitely a result of changes in funding policies. Funders have a positive view of preprints and encourage that they be included in grant applications or at the end of a grant report. But another driver of growth is that fact that preprints are increasingly seen as proof of progress and central to scholarly sharing practices, and that they are in alignment with recent sociocultural shifts in research: 1) the expectation of and a general cultural comfort with speed and ease over perfection; 2) a scrutiny of the peer review process and reproducibility concerns; and 3) a growing demand for free access to content.
She admitted that there are potential concerns around preprints, including the dissemination of poor quality and irreproducible data, but that this can be mitigated by basic quality control [5].
Truly, in 2016 the perfect storm had arrived!
ChemRxiv - Celebrating its First Birthday!
If you visit the ChemRxiv website it is still listed as “beta.” (as of 30 June 2018) Henderson said that they are in the process of establishing a Governing Board and once that has been done the site will no longer be “beta” and the formal vision and mission will be released. In essence, the mission (on which all three collaborators have agreed) is to operate as a preprint server for the global chemistry community in order to: promote research in the chemical sciences; increase the dissemination of chemical knowledge; and to promote scientific interests.
She added that they also want to integrate the service into the overall publishing process so that once a preprint has been posted, commented upon, etc., and the author has prepared a final manuscript, they can click a button and submit it to a publisher. All interested publishers can participate in this feature for a fee that will be used solely to support the service. This feature will be available in 2018.
Two metrics are being used to measure success: how many authors are submitting preprints and how many users are visiting the site. Henderson said that the first metric has met modest success. There have been 700 submissions that resulted in 400 postings and the postings have come from prominent institutions such as Harvard, MIT, and Imperial College. The delta between the two numbers is due to the fact that prior to actual posting the preprints are scrutinized for possible plagiarism, prior publishing, and content (no political opinions are accepted and the research must be new research in chemistry or a chemical-related discipline).
With regard to the second metric Henderson said that there were more than pleasantly surprised. As of when we spoke on June 12, 2018 there had been approximately 378,000 downloads/views. These numbers were totally unexpected!
She said that the process of posting is quite simple. One page of information needs to be completed, and then the author can click, drag, and upload supporting files. Because figshare (see: https://www.digital-science.com/products/figshare/) is supplying the basic technology, there are no limitations on file types. Authors can upload Word documents, Excel files, PDFs, videos, and original data sets and they are encouraged to do so. The submissions are reviewed as noted earlier and staff do an authentication of email addresses, ORCID IDs (required), metadata, etc. Information on an author’s institutional affiliation as well as any funding records are also gathered. A ChemRxiv staff member looks at the results of the preliminary review and makes a decision on whether or not to post the preprint. Henderson made it clear that they make no judgement on the accuracy of the science or its potential impact. Results can be positive or negative. The major requirement is that the research is new. When posted, all preprints are assigned a DOI.
She noted that the relationship with figshare is ongoing and that together they are looking at future enhancements. One new feature already implemented is the use of categories so that users of the site can filter their searches. The categories are based upon the high-level categorization used in Chemical Abstracts. They are also in the process of standardizing the information that is gathered in order to expedite the downstream publishing of manuscripts. Going forward, they are seeking input from users on what features and functionalities they would like to see added.
Henderson noted that the server does not have a “comments” feature, but this is under evaluation. BioRxiv does have this feature, but they have said that it is a time-consuming, manual effort to monitor. Most authors and users seem to prefer to communicate via phone and/or email, so ChemRxiv has author email addresses hot-linked in the system for ease of use. She noted that some of the preprints have already found their way into journals. An agreement is in place with Crossref that when a publisher deposits data that matches data in ChemRxiv, Crossref will notify them so that ChemRxiv can add a link from the preprint to the Version of Record.
Next Steps
Henderson said that they are still open to other partners joining the collaboration. They continue to market the service, although she admitted it is mostly word-of mouth. She also noted that in recent months submissions have become more international so she is hoping to work with global organizations in getting the word out.
She noted that word is getting out and that ChemRxiv was in the news earlier this year when an article appeared in Wired magazine. Apparently, last August an organic chemist had spoken at a conference discussing his radical research results. This was done prior to publication and he quickly realized based on audience reaction that he had to establish his ownership of the work. A colleague at the Scripps Institute of Research suggested that he post a preprint on ChemRxiv and in less than twenty four hours later the paper was online and being discussed. When the Wired article was released on 8 January 2018 there had already been 8 600 views of the preprint [9].
The article is a must-read because it also discusses how the co-existence of preprints and traditional journal articles in the scholarly communication process can result in complications. The article is well-written for both scientists and non-scientists alike, and clearly lays out the issues surrounding preprints versus traditional publishing. But preprints are here to stay. Today’s world of scholarly communication strives for quick dissemination of research results in an environment of openness and sharing, and preprints fill the need. Equally important, research funders agree.
In closing, I leave you with a quote from the Wired article mentioned earlier: “Preprint has already transformed scientific publishing; now chemistry preprint has a helluva trip ahead of it.”[9]
Über den Autor / die Autorin
Bonnie Lawlor <chescot@aol.com> is Chair of the IUPAC Committee on Publications and Cheminformatics Data Standards (CPCDS).
References:
1. “American Chemical Society Announces Intention to Establish “ChemRxiv” Preprint Server to Promote Early Research Sharing,” Press Release, American Chemical Society, 10 August 2016, https://www.acs.org/content/acs/en/pressroom/newsreleases/2016/august/acs-announces-intention-to-establish-chemrxiv-preprint-server-to-promote-early-research-sharing.html (accessed 30 June 2018).Search in Google Scholar
2. Wikipedia, https://en.wikipedia.org/wiki/ArXiv (accessed 30 June 2018).Search in Google Scholar
3. Wellcome Trust. Sharing data during Zika and other global health emergencies; 10 February 2016. https://blog.wellcome.ac.uk/2016/02/10/sharing-data-during-zika-and-other-global-health-emergencies/Search in Google Scholar
4. Martin, B., “Learning to Love Rejection,” Inside Higher Ed, 8 July 2013, https://www.insidehighered.com/advice/2009/04/27/when-journal-says-no (accessed 30 June 2018).Search in Google Scholar
5. Shirley Decker-Lucke,”Transforming the Stage at which Scholars Share Information about their Research: How Early Can You Go?” presentation at the 60th Annual Conference of NFAIS <https://nfais.memberclicks.net/2018-conference-program> and access to slides therein (accessed on 30 June 2018).Search in Google Scholar
6. Joris van Rossum, “Academic publishing, the blockchain, and shifting roles in a rapidly changing world” presentation at the 60th Annual Conference of NFAIS <https://nfais.memberclicks.net/2018-conference-program> and access to slides therein (accessed on 30 June 2018).Search in Google Scholar
7. Reporting Preprints and Other Interim Research Products, U.S. National Institutes of Health, Notice Number: NOT-OD-17-050, March 24, 2017, https://grants.nih.gov/grants/guide/notice-files/NOT-OD-17-050.html, (accessed on 30 June 2018).Search in Google Scholar
8. Neil Thakur, “Using open science to speed dissemination, reduce burden and measure impact” presentation at the 60th Annual Conference of NFAIS <https://nfais.memberclicks.net/2018-conference-program> and access to slides therein (accessed on 30 June 2018).Search in Google Scholar
9. Simon, M., “Salvia Leads Chemists on a Psychedelic Existential Journey,” Wired, 8 January 2018, https://www.wired.com/story/salvia-leads-chemists-on-a-psychedelic-existential-journey/ (accessed 30 June 2018).Search in Google Scholar
©2018 IUPAC & De Gruyter. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. For more information, please visit: http://creativecommons.org/licenses/by-nc-nd/4.0/