Jupyter Notebooks Facilitating Productivity, Sustainability, and Accessibility of Data Science

2017-02-26T19:36:00Z (GMT) by Min Ragan-Kelley Project Jupter
Poster presented at SIAM CSE17 PP108 Minisymposterium: Software Productivity and Sustainability for CSE and Data Science<div><br></div><div>Jupyter notebooks provide a document-centric web-based environment for data science and education. The web-based platform allows Jupyter to be used as a local desktop application or portal to large scale computing resources. The document aspect facilitates recording, sharing, and reproducing analyses. The integration of notebooks into existing CSE research and education workflows can improve reproducible workflows and communication and collaboration. The interactive nature of the notebook environment enables productive exploration of libraries and analyses, and being based on web technologies allows the same productivity to extend even to large-scale computing resources, which often pose a challenge to productivity and accessibility. The document nature of Jupyter notebooks facilitates preserving and sharing computations and results. Being a free, public, open source project enables equal access to all students and researchers, and the web environment is more familiar and accessible to a much wider set of current and prospective researchers than traditional terminal environments. Due to Jupyter's language agnostic protocol, with support from dozens of programming languages, notebooks can be used in many scientific and mathematical domains, which can have diverse language preferences.</div>