Friday, July 21, 2017By:
If this week has taught me anything, it’s that the term “open access” is not nearly as clear-cut as I thought.
Before I jump into the meat of this post, let me quickly throw in the fact that I not only got a fourth article published this week, the people at Physics Today liked the article I wrote with Alexis on Horizon 2020 so much they posted an abbreviated version on their website!
Back to business, though. This Thursday I attended a kickoff meeting for a new 18-month long National Academies study on moving toward an open science enterprise. As stated on their project webpage,
Open science is defined, for the purposes of this study, as public access (i.e., no charge for access beyond the cost of an internet connection) to scholarly articles resulting from research projects, the data that support the results contained in those articles, computer code, algorithms, and other digital products of publicly funded scientific research, so that the products of this research are findable, accessible, interoperable, and reusable (FAIR), with limited exceptions for privacy, proprietary business claims, and national security.
An ad hoc committee under the Board on Research Data and Information (BRDI) will conduct the study.
During the Thursday meeting, I expected there to be a lot of publishers protesting the idea. While I’m no expert, I’ve been aware of the recent debate raging in the science publishing world over access rights to academic articles. It used to be that the results of a study were kept quiet until the official release of an article, and commonly that article was only available to those with subscriptions to the publishing journal. However, a shift has been taking place over the years toward a much more open format, where preprints are common (a printing of a paper, often in a preliminary form, before its final publication in a journal) and less articles require a subscription to read. Yet some journals have been fighting back against this paradigm since it loses them money, and will refuse to publish an article if the author chooses to preprint it first.
I didn’t see any of this at the meeting. In fact, one of the speakers, Heather Joseph, the Executive Director of the Scholarly Publishing and Academic Resources Coalition, encouraged the use of open access, which she defined as a subset of open science that focuses specifically on access to articles that report on outcomes of research. What she argued against, however, was the immediate accessibility of data.
As she explained, making articles immediately available wasn’t so much the issue as figuring out how to make all the data behind the article accessible. By making articles immediately accessible, it can save other researchers time and energy by letting them know straightway if a certain research path appears unfruitful. However, keeping the actual data the author used to come to their conclusions under closer surveillance protects from possible copyright infringements.
Yet this comes with its own costs. Without access to the exact data and methodology used to conduct a study, other scientists are unable to conduct reproducibility experiments to see if they can achieve the same results. That’s why many of the meeting’s other speakers stressed the need to make this information available as well, though admitted creating a repository to hold all that data would be a tremendous undertaking. Marcia McNutt, the president of the National Academies of Science and editor-in-chief of Science journal, said Science does not create a repository of all the published articles’ data but requires each author to be able to produce that data for interested parties when asked.
Not only are there various ways to define what “open access” means, the levels at which different scientific disciplines struggle with this problem vary widely as well. While I look forward greatly to seeing the conclusions this committee draws in their final report, I do not envy them the work it will take to create it.