loading page

Jupyter Book-based Supplemental Material: a FAIR Practice to Connect  Research Articles with Scientific Data  
  • +7
  • Whyjay Zheng,
  • Fernando Perez,
  • Chris Holdgraf,
  • Tasha Snow,
  • Shane Grigsby,
  • Facundo Sapienza,
  • Erik Sundell,
  • Matthew Siegfried,
  • Jonathan Taylor,
  • Executable Books Community
Whyjay Zheng
University of California, Berkeley

Corresponding Author:[email protected]

Author Profile
Fernando Perez
University of California, Berkeley
Author Profile
Chris Holdgraf
Author Profile
Tasha Snow
Colorado School of Mines
Author Profile
Shane Grigsby
NASA Goddard Space Flight Center
Author Profile
Facundo Sapienza
University of California, Berkeley
Author Profile
Erik Sundell
Sundell Open Source Consulting
Matthew Siegfried
Colorado School of Mines
Jonathan Taylor
Stanford University
Executable Books Community


Supplemental material (SM; also known as supplementary information) comes with its associated research article and provides study details such as metadata, additional figures and text, multimedia, and code. Well-designed SM helps readers fully understand the underlying scientific analysis, reproduce the work, and even reuse the workflows for exploratory ideas. Thus, the concept of FAIR (Findable, Accessible, Interoperable, and Reusable), which is originally designed for data sharing guidelines, also matches these core qualities for SM.
We evaluate different SM-preparation practices that are commonly found in Earth Science journal articles. These practices are classified into five tiers based on the FAIR principles and the narrative structure. We show that Jupyter Book-based SM belongs to the top tier and outperforms the other practices, despite being not as popular as the other SM-preparation practices as of 2022.
We identify the advantages of the Jupyter Book-based SM as follows. Jupyter Book uses a narrative structure to combine different elements of SM into a single scholarly object, increasing readability. Jupyter Book's direct support of HTML publishing allows users to web host the SM using services such as Github Pages, improving the web indexing ranks and resulting in higher exposure of both the research article and the SM. The entire SM is also eligible to be archived in a data repository and receive a Digital Object Identifier (DOI) that can be used for citations. In addition, Jupyter Book-based SM lowers the threshold of reproducing and reusing the work by accessing an interactive cloud computing service (e.g., MyBinder.org) with all data and code imported if the content is available on a code-hosting platform (e.g., Github).
These features summarize the core values of SM from the perspective of open science. We encourage researchers to use these good practices and urge journal publishers to be open to receiving such supplements for maximum effectiveness.
20 Dec 2023Submitted to ESS Open Archive
27 Dec 2023Published in ESS Open Archive