Project Jupyter — LLMpedia

Project Jupyter
Cameron Oelsen · BSD · source
Name	Project Jupyter
Developer	NumFOCUS, Caltech, UC Berkeley, IBM, Microsoft
Released	2014
Programming language	Python, JavaScript, TypeScript
Operating system	Cross-platform
License	BSD, MIT

Contents

Project Jupyter is an open-source initiative for interactive computing that originated from the IPython project. It emphasizes notebook-style interfaces, reproducible research, and interactive data analysis used across institutions such as Massachusetts Institute of Technology, Harvard University, University of California, Berkeley, California Institute of Technology, and corporations like IBM, Microsoft Corporation, Google LLC, Intel Corporation, and NVIDIA. The project intersects with platforms and standards developed by communities around NumPy, SciPy, pandas, Matplotlib, and integrates with services from Amazon Web Services, Google Cloud Platform, Microsoft Azure, and research infrastructures at CERN and Lawrence Berkeley National Laboratory.

History

The initiative emerged from the evolution of the IPython project, influenced by work at University of California, Berkeley, collaborations with researchers at Massachusetts Institute of Technology, and contributions from engineers formerly at Enthought and Continuum Analytics (now Anaconda, Inc.). Early milestones include the split between interactive shells and notebook interfaces, shaped by events like the SciPy conferences, the founding of NumFOCUS as a fiscal sponsor, and adoption by academic groups at Princeton University, Stanford University, University of Washington, and research labs at Bell Labs. Significant releases and community governance were discussed at summits involving stakeholders from Mozilla Foundation, Mozilla, Wikimedia Foundation, and funders such as the Alfred P. Sloan Foundation and the Gordon and Betty Moore Foundation.

The ecosystem centers on a web-based frontend, a messaging protocol, and pluggable execution backends, developed by contributors affiliated with Mozilla Foundation, Microsoft Research, IBM Research, Google Research, and universities including Columbia University and Brown University. Core components include a browser-based interface influenced by Electron (software), a server application implemented in Python (programming language), and a kernel messaging protocol used by projects at Red Hat, Canonical (company), and Blue Yonder. Integrations extend to visualization libraries such as D3.js, Bokeh, Plotly, and compute systems like Hadoop, Apache Spark, Kubernetes, and Docker (software) supported by vendors like Red Hat and IBM.

The notebook document model uses a JSON-based representation, adopted by data scientists from institutions such as Columbia University, Yale University, New York University, and companies like Facebook and Twitter. File extensions and ancillary formats enable conversion workflows with tools from Pandoc, Sphinx (software), Sphinx, and static site generators used by projects at Mozilla and NVIDIA. Export targets include formats supported by LaTeX, PDF, HTML5, and presentation frameworks used at PyCon, SciPy, and JupyterCon community events.

The platform supports multiple language backends via kernels contributed by communities around Python (programming language), R (programming language), Julia (programming language), Scala (programming language), Haskell (programming language), Java (programming language), and others developed by teams at RStudio, The Julia Language, EPFL, Massachusetts Institute of Technology, and CERN. Language interoperability is facilitated by standards influenced by work at W3C, IETF, and open-source projects such as ZeroMQ, with developer contributions from Microsoft Research, IBM Research, and Google Research.

Adopters span academia, industry, and government labs including Stanford University, MIT, Harvard Medical School, National Institutes of Health, NASA, European Organization for Nuclear Research, Lawrence Berkeley National Laboratory, Facebook, Amazon (company), Twitter, Bloomberg L.P., and Goldman Sachs. Use cases include teaching at events like PyCon, research reproducibility promoted by journals such as Nature (journal), Science (journal), and Proceedings of the National Academy of Sciences of the United States of America, and workflows in industrial data science teams at Uber, Airbnb, Stripe, and Spotify.

The governance structure involves community-led working groups, steering committees, and fiscal sponsorship by NumFOCUS, with funding from philanthropic organizations including the Alfred P. Sloan Foundation, the Gordon and Betty Moore Foundation, and corporate sponsorship from IBM, Microsoft Corporation, Google LLC, and Amazon Web Services. Strategic decisions and roadmaps have been shaped in collaboration with representatives from University of California, Berkeley, Massachusetts Institute of Technology, Princeton University, ETH Zurich, and industry partners like Red Hat and Anaconda, Inc..

Critiques and debates have arisen regarding reproducibility concerns highlighted in reports by researchers at Stanford University, University of Toronto, and University of Washington, security issues discussed by teams at Google Project Zero and CERT Coordination Center, scalability challenges in deployments evaluated by Netflix and Dropbox, and governance transparency questioned in community forums involving members from NumFOCUS and academic institutions such as Harvard University and Princeton University. Discussions about licensing, commercial influence, and ecosystem fragmentation have involved stakeholders from Anaconda, Inc., Microsoft Corporation, IBM, Google LLC, and grassroots contributors from conferences like PyData and JupyterCon.