Notebooks as the Published Paper, How Ready are Scientific Communities?

Science today increasingly relies on software to conduct, analyze, and share research. However, the primary artifacts of scientific work, such as papers and PDFs, fail to adequately capture the dynamic and interactive nature of software-driven exploration. Over the past decade, several initiatives have emerged to bridge this gap. One notable example is EarthCube, which in its final three years called for geoscience submissions in the form of computational notebooks for its annual meeting. This experiment not only showcased the existence of a scientific community embracing the idea of a more interactive and reproducible format for scientific sharing but also led to further exploration: AGU, with support from the Sloan Foundation, launched an effort to prototype a complete pipeline for soliciting, reviewing, and accepting notebooks as primary publications.

This shift parallels the broader historical transition from physical papers to digital PDFs and XML-based publications. However, a critical question remains: Are research communities ready for this evolution? While geoscientists embraced the notebook format during EarthCube's final years, other disciplines have been more hesitant to move away from traditional static documents. The impediments include a range of technical gaps, lack of resources, as well as socio technical blockers. At the same time, many scientific communities express a desire for additional methods of scholarship, and are motivated to take part if others can lead the way.

To better understand the community's readiness, the NSF CSSI DeCODER project, an initiative that continues some of EarthCube's data and notebook-driven activities, conducted a survey targeting scientists across various fields. This talk will present the findings of that survey, offering insights into the scientific community's perspectives on notebook-driven publication and exploring potential next steps to foster broader adoption.