From CSE to SSP: Evolving Perspectives on Data Management
The Society for Scholarly Publishing (SSP) annual meeting is just getting ready to kick off in Arlington, VA today. I’m going to be a bit slammed over the next couple of days. Somehow, they’re trusting me to moderate three sessions, including a pre-meeting seminar and one round table session on which I’m also a speaker. I hadn’t intended to do both roles for the round table, but the original moderator had to pull out. Of course, this does mean that I’ll be asking myself questions, so we’ll have to see how successful I am at not making that sound awkward.
With all this preparation, I’ve barely had time to stop and think about the things that I learned from the Council of Science Editors (CSE) conference in Philadelphia last week. In many ways, CSE is an interesting counterpart to SSP. It’s quite a bit smaller and (at least for this, my first year there) was attended by a lot of science editors, managing editors, production people and the occasional technologist, looking to understand more about the ways in which information technology might solve problems within the publishing workflow. I found the whole event immensely useful and learned a lot about how the people who do most of the actual work in scholarly publishing are thinking about issues like production workflows, metrics and user experience.
Aside from our very own Amy Brand winning the award for outstanding achievement this year, the highlight of the conference for me was Sayeed Choudhury’s keynote talk titled, The Research Data Revolution.
Choudhury is the Associate Dean for Data Management at the Sheridan Libraries and a frequent public speaker on the subject of data, so you may have heard him speak before, but for those who haven’t heard this talk, Choudrey’s central thesis is that data represents a new frontier in scholarly communication. Research articles represent the final report on a body of hypothesis-driven research, and as such are central to the scholarly record in countless ways. Research data, on the other hand are the building blocks of those articles and currently, the archiving and preservation of those blocks is an unsolved problem. As Choudrey pointed out, most researchers struggle to find or re-analyse even their own data five years after it was taken.
Some people argue that data publishing suffers from a fatal obstacle that the creation of data will always be less valued than the interpretation of it. While there is some truth to that, particularly in biology, it’s less true in fields like physics, where researchers can make very successful careers as experimentalists coming up with new ways of testing the implications of the theoreticians’ ideas. I personally believe that giving instrumentalists, statisticians and informaticians a little more credit in fields like biology, would be good for those disciplines and prevent the loss of many talented individuals who currently feel undervalued. Incidentally, Amy Brand spoke at CSE about this very topic when she spoke about the CRediT project later in the program.
One of the most interesting observations that Choudhury made, at least from my perspective, was to acknowledge the relationship between librarians and scientists is not as strong as it is between librarians and humanists. The reason for this, he suggested, is that librarians often know the relevant material in their holdings better than the humanists who come in to ask them about it. Often, a humanist might come into the library asking for one resource and the librarian will be able to show them something even better.
In science, this is a relationship that is rarely present. I agree with Choudhury that this is partly due to the fact that many librarians have backgrounds in the humanities and are therefore better positioned to understand a humanist’s needs. However, I think that it’s also true that in the humanities, the object of study is very often contained inside the library, in special collections. In the sciences, the object of study is either in the lab, or it may be outside of the university altogether (for example, it might be a distant galaxy). This casts the role of the librarian in data management in an interesting light. As Choudhury puts it, perhaps data represents a potential new form of special collection, one which will enable librarians to strengthen their relationship with science faculty.
There was much more going on at the CSE conference last week than I summarized here. There were a series of wonderfully forward-facing, idea-driven conversations covering everything from the true purpose of altmetrics to the importance of information interchange standards and how to improve typesetting meta-data to make content future proof and convertible. I’m looking forward to exploring these ideas more this week at SSP, particularly around data, credit, and the role of both libraries and publishers in solving the data archiving and preservation problem. If you’ll excuse the shameless plug, that pre-meeting seminar that I mentioned at the top of this piece is on the topic of the publisher’s role in data management and takes place on Wednesday at 8:30am EDT.