Project Jupyter: collaboration is the name of the game
Data science relies mostly on collaboration. Data scientists work together to clean data, create models and to do statistical analysis. Then, after all that work, the day comes when they need to decide how to show their findings to the rest of the world.
One of the most interesting platforms nowadays to do this is through Project Jupyter (formerly IPython), a set of open-source software tools for collaborative and exploratory computing.
History and Functions
IPython began its life back in 2001 as an environment for interactive scientific computing and data analysis focused on Python. It eventually evolved into one of the main elements of the collaboratively developed ecosystem of open source science tools in Python. In recent years, IPython has evolved into Project Jupyter. Why the name change? Other languages are important for data scientists, so IPython should and can run other languages, such as R, Julia, Scala, Haskell, Bash, among others. So, IPython is and will be known as one of the many kernel options of the Jupyter Notebook.
Among the many tools that Project Jupyter offers, the most innovative is a web-based notebook environment called Jupyter Notebook. With it users can create data- and code-driven narratives that combine live code, equations, markdown text, visualizations, interactive dashboards and other media. These documents, which provide a complete record of computation, can be shared online, converted to other formats and be used as a tool for research and education. They can also be published online thanks to Jupyter Notebook Viewer, a free service that allows anyone on the web to view a notebook as a regular web page. GitHub also now offers a render for *.ipynb files, which are the standard format for Jupyter Notebook files.
The Jupyter project now also has a new multiuser server, called JupyterHub, which allows multiple users to authenticate and launch notebook servers, which is designed to be deployed within organizations and on private or public clouds.
The Future is in Collaboration
Jupyter is the standard tool among data science centric academics and scientists who have to deal with large datasets in multiple formats. Such is the support for this tool that it’s getting economic support via grants to broaden its capabilities to reach even wider audiences
On the first days of July 2015, it was announced that Project Jupyter will receive $6 million in grants over three years from three foundations: the Alfred P. Sloan Foundation and the Gordon and Betty Moore Foundation, which are each providing $1.5 million to UC Berkeley, and the Helmsley Charitable Trust is providing $3 million to California Polytechnic University, San Luis Obispo.
With funding over the next three years, the researchers will expand and improve upon capabilities of the Jupyter Notebook.
So far, the creators of Jupyter estimate that more than 1 million people in fields ranging from astronomy to finance currently use it. Applications include the analysis of massive gene-sequencing datasets, processing images from the Hubble Space Telescope and developing models of financial markets. The long-term goal is that Project Jupyter serves not only the academic and scientific communities, but also professionals that work in the education, industry and journalism fields.
Why re-invent the wheel? Jupyter Notebook and JupyterHub are fabulous pieces of software and we feel lucky and honored to be able to collaborate on making the software better for the community.
Would you like a demo for 3Blades? Enroll for one by clicking on the button below!