This is a guest post by Nicole Contaxis.
Example submissions for the Astrophysics Source Code Library: http://ascl.net/code/all
On April 12th2016, Alice Allen, editor of the Astrophysics Source Code Library, came to the National Library of Medicine to speak with National Digital Stewardship Residency participants, mentors and visitors about the importance of software as a research object and about why the ASCL is a necessary and effective resource for the astronomy and astrophysics academic communities.
Astrophysicists and astronomers frequently write their own code to do their research, and this code helps them interpret and manipulate large data sets. These codes, as an integral part of the research process, are important to share for two reasons: (1) they increase the efficiency of work by allowing code to be re-used and (2) it helps ensure the transparency of scientific research.
Yet, difficulties persist when it comes to encouraging researchers to share source code, regardless of the benefits. Allen talked about how researchers are reluctant to share code that may be “messy” and that creating this source code library requires community engagement and change management. She spoke about studying the impact of non-traditional scholarly outputs, like code, and the issues of scholarly publishing. Allen showed how ASCL has helped allow journal authors cite code, which had been a far more difficult procedure earlier. The ASCL assigns unique identifiers, called ASCL IDs, so that future academics can cite that code, even if that code is not featured in a journal article. Every major astronomy journal accepts ASCL IDs in citations.
Image from the Visible Human Project at the National Library of Medicine.
The discussion turned to the difficulties of grant-based funding. The ASCL is basically unfunded, and all labor, including Allen’s, is voluntary. While talking about other code libraries that have lost funding and closed, Allen talked about how grant-funding, which runs on two- to five-year cycles, does not provide enough time to fully engage a community with a resource, regardless of how well that resource is designed, implemented and managed. Funding, as a universal source of concern, was a common point of interest, even for attendees without experience working with software or code.
The session included a tour of Visual Human Project, which is an NLM project that collects extensive data on a male and female cadaver, allowing artists and researchers to visualize that data in new and exciting ways.