Project Jupyter Steering Council Receives 2017 ACM Software System Award

(L-R): Bloomberg Engineers Paul Ivanov & Jason Grout who also serve on the Project Jupyter Steering Council

The Association for Computing Machinery (ACM) recently awarded Project Jupyter the 2017 ACM Software System Award for its lasting influence on computing. As early members of the 15-member Project Jupyter Steering Council, two of Bloomberg’s San Francisco-based engineers, Jason Grout and Paul Ivanov, are among the recipients of this notable award, which was presented this past weekend at the ACM Awards banquet in San Francisco.

“Receiving the award is tremendously satisfying for all the time and effort put into Project Jupyter, but first and foremost, we recognize that this belongs to the entire Jupyter community,” said Grout. “Jupyter wouldn’t be what it is without a community of people developing it and organizations like Bloomberg supporting it. In a sense, this is a tremendous vote of confidence and affirmation of the architecture, design and principles we pursued in building Project Jupyter.”

Members of the Project Jupyter Steering Council at the ACM Awards Dinner on Saturday, June 23, 2018

Grout and Ivanov are primary contributors to Project Jupyter, both having worked on the open source system’s tools, such as IPython, the Jupyter Notebook, and ipywidgets, prior to joining Bloomberg. They have community-facing roles that serve as a bridge for Bloomberg resources and other engineers who are supporting the project. They have also been involved with Bloomberg’s partnership with Anaconda Inc. and Project Jupyter to develop the next-generation web interface, JupyterLab.

As members of the Steering Council, “we make sure the code that’s relevant to the project is reviewed and tested, and we’re the ones who are trying to grow our community base of users and contributors,” said Ivanov. “We do this by giving talks and lectures about the technical aspects of the project – how somebody can extend the code for their purpose. We make sure there’s a pipeline of new contributors and that the views and needs of our users are well-represented in our ongoing development efforts.”

The interactive computing tools developed as part of Project Jupyter have become standard in data analysis and have had a broad impact. There are more than two million Jupyter Notebooks on GitHub covering a wide range of use cases across many different domains. The Jupyter Notebook’s web-based environment allows researchers and programmers to do computational research with native support for code, math and data visualization.

“We were able to steward and build a community to make a tool that has wide usage throughout industry and academia – it’s used for teaching, development work, and data analysis,” said Ivanov.

This project not only develops open source software tools, but also provides standardized protocols for interactive data science and interactive computing and defines open formats for communicating results with others. “The real genius in Project Jupyter is that we’ve tried to design open protocols and formats that enable interactive computation,” said Grout.

With this foundation, people are able to share scientific results that can be replicated by others. Using open source software reaffirms the fundamental principles of scientific research of openness and reproducibility, while being able to modify and redistribute the tools of the trade. Standardized protocols allow people to work in a number of different front-ends, regardless of programming language, and talk to the back-ends that perform the actual computations.

Jupyter Notebooks are an invaluable tool to a company like Bloomberg that provides data, analytics and news with a focus on financial markets, but, at its core, is very much a technology company. They have been widely adopted within the financial world because of the availability of wide-ranging datasets and the easy sharing of the resulting calculations. This system is quickly becoming another medium that enables Bloomberg’s clients to consume and make data-driven decisions.

“Our clients are using it, we’re using it internally and we want to give back to the community as well,” said Grout. “We see the marriage between our data and advanced computational tooling as something our clients can use. Increasingly, our clients want to do more sophisticated analysis and they can use this for advanced computation in combination with Bloomberg’s data.”

Project Jupyter also provides an element of transparency and openness that’s important for anyone conducting data analysis. “It’s flexible and allows you to communicate about the data analysis you’ve performed, while also allowing others to extend the analysis in the future,” said Ivanov. “You can communicate what you did, so somebody else can redo that or even pick up from what you did and go in a different direction. That’s the flexibility of the Jupyter Notebooks.”

In general, supporting open source is a way for Bloomberg to help ensure that everyone can participate in the technological landscape. “In a way, Bloomberg is supporting the work of students and the ability of students of all backgrounds to gain access to computational resources, but also to participate along with other tech giants and academic institutional giants,” said Ivanov.

Members of the Project Jupyter Steering Council at Bloomberg Engineering's San Francisco office for a reception on Friday, June 22, 2018, in honor of their receiving the 2017 ACM Software System Award

The ACM Software System Award has been given annually to the pioneers that created the computing infrastructure we have today, like the internet. “Giving it to a major open source initiative like Project Jupyter, which is focused on building community and open standards, is a tremendous affirmation to how important open systems are,” said Grout. “It’s more than just open source – it’s an open community where people are encouraged to build systems on top of these protocols.”

These tools have a large impact, and it’s important that everyone continues to have an entry-point regardless of their ability, budget, background, etc. The Steering Council continues to work to open the system up to a wider audience, while still serving those who have given the project and its tools the prominence it has attained.

“Open source projects and their communities are egalitarian, where everyone can understand the source code, and you can change, modify and redistribute algorithms,” said Grout. “Results are open, as are the methods and tools. These principles are at the core of scientific progress in the world today.”