Our changing scientific environment

Version imprimable
Vincent Poitout

JANUARY 9, 2018

In many respects, our working environment is rapidly changing and requires constant adaptation on our part. In particular, the scientific publishing scene is undergoing transformation, with the gradual disappearance of print publications, the growing practice of posting preprints, the near exponential increase of information available online and the growing number of fake scientific journals. To address these challenges, I propose some possible solutions based on my own analysis and personal experience. I invite you to provide your input by sending me your comments and suggestions (vincent.poitout@umontreal.ca), which I will summarize in a future Message from the Director.

Finding needles in the haystack

When I was a PhD student, we would systematically comb through the weekly Current Contents, and, using pre-filled-out postcards, ask the authors of a few articles of interest to send us copies by snail mail! Times have changed and we now have an inordinate volume of scientific literature at our disposal: 1.4 million articles are published every year by the 25,000 to 40,000 scientific journals1. We are all having to deal with this multiplication of sources of information and the increasing difficulty of reading them in a minimally thorough way.

Personally, I continue to receive a weekly e-mail with the list of articles published on PubMed based on keyword searches, and the tables of content of a dozen of relevant journals. But I am increasingly discovering articles through social media, Twitter in particular, thanks to tweets from my research colleagues, some of whom communicate regularly, systematically pointing out articles of interest in my field. The filter bubble phenomenon probably has a positive impact in this case, as it tends to personalize information (unbeknownst to us) based on previous searches. 

Separating the wheat from the chaff

The proliferation of fake scientific journals, also known as predatory journals, threatens the credibility of all our activities. The publishers of these journals are commercial enterprises that do not comply with the best peer-review rules and practices, have opaque publishing costs, and aggressively solicit articles. It is important to remember that publishing in a predatory scientific journal is a considered a violation of the CHUM’s Policy on Responsible Conduct of Research2. Unfortunately, the line between legitimate journals and predatory scientific journals is fuzzy. A recent study published in Nature3 concerning 2,000 articles published in suspected predatory journals shows that major academic institutions are not immune from these fraudulent activities.

We have to be vigilant in order to protect our scientific reputation and that of our institution. But how can we know what to trust? In the U.S., the National Institutes of Health (NIH) recently published a notice4 offering helpful recommendations and references to detect predatory journals, including the site “Think Check Submit”5. Personally, I stick with journals that have well-established reputations in my field and are published by credible publishers, ideally members of the Committee on Publication Ethics (COPE)6.

To preprint or not to preprint?

The practice of preprinting consists of submitting a publication to an open-access server (such as bioRxiv7, by Cold Spring Harbor Laboratory) before submitting it to a journal and having it peer-reviewed. This practice has been common in other fields, such as physics, for several decades, but has only recently emerged in the life sciences. Although far from enjoying unanimous support,7 it is gradually gaining legitimacy and large funding agencies such as the Medical Research Council in the United Kingdom, the National Institutes of Health (NIH) in the U.S. and the Howard Hughes Research Institute now encourage researchers to list preprints in their grant applications. Defenders of this practice see it as a way of accelerating the dissemination of research results and receiving peer comments, making it possible to improve articles before the final publication. Its detractors fear that their ideas will be stolen before they can publish their results and that the accumulation of non-peer-reviewed preprints on these servers makes it extremely complex to search for relevant, credible information. 

The subject is much too complex to summarize briefly here, and I suggest that you read the survival guide published in Science8 recently, which gives very useful information. Although I have not tried this practice out myself as a researcher, I see more advantages than drawbacks in it. In fact, one of my resolutions for 2018 is to preprint an article from my group! But I will be particularly interested in receiving your comments and opinions on this subject and those discussed earlier. Send me your e-mails!

Vincent Poitout
Director of the CRCHUM


1 https://scienceintelligence.wordpress.com/2012/01/23/how-many-science-journals
2 http://crchum.chumontreal.qc.ca/ressources-internes/conduite-responsable-en-recherche
3 Moher D, Shamseer L, Cobey KD, et al. Stop this waste of people, animals and money. Nature 2017;549(7670):23-25. doi: 10.1038/549023a. PubMed PMID: 28880300.
4 https://grants.nih.gov/grants/guide/notice-files/NOT-OD-18-011.html 
5 http://thinkchecksubmit.org
6 https://publicationethics.org
7 https://www.biorxiv.org
8 Kaiser J. The preprint dilemma. Science 2017;357(6358):1344-1349. doi: 10.1126/science.357.6358.1344. PubMed PMID: 28963238.