SciPy 2024

From Spyder to DataLab: 15 years of scientific software crafting in Python
07-11, 11:25–11:55 (US/Pacific), Ballroom

This talk provides an overview of the evolution of scientific software in Python, with a focus on the speaker's journey from creating Spyder, the Scientific Python IDE, to developing DataLab, a platform for signal and image processing. The speaker will share insights into the challenges and opportunities encountered in developing and maintaining these projects, and discuss how they have contributed to the scientific Python ecosystem. The talk will also explore the evolving needs of both the scientific and industrial communities during this period, and why desktop applications remain relevant in the era of web-based tools.

The Early Days: Python(x,y), Spyder, WinPython

This part of the talk aims at explaining the context and motivations behind the creation of Python(x,y) and Spyder. It also highlights the challenges faced in developing and maintaining these projects, and the lessons learned from these experiences.
The speaker presents how Python(x,y) and Spyder were initially crafted to meet the needs of the scientific community, brick by brick, and how they evolved over time to adapt to the changing requirements of the scientific and industrial communities. Spyder - named initially as Pydee - started with a simple Qt-based variable explorer, as a complement to a Python interpreter, and evolved into a full-fledged IDE. Python(x,y) was replaced by WinPython, a more modular and flexible distribution, whereas Spyder became widely used thanks to a strong and active community of contributors.

Scientific Applications with Python and Qt

After creating the necessary tools for distributing and installing Python and its scientific libraries, and for developing and maintaining scientific code, it was time to create libraries providing the building blocks for scientific applications (guidata, guiqwt, ...).
This part of the talk focuses on the development of desktop applications for scientific and industrial purposes, and how Python and Qt have been used to create these tools. It begins with the motivation behind the development of these libraries, and then describes how they have been used over the years to create a variety of scientific applications.
The culminating point of this part is the introduction of the PlotPyStack toolkit, which is the foundation of DataLab, a platform for signal and image processing.

See also: DataLab website

Pierre Raybaut is a long-term advocate of Python in a scientific context, renowned as the creator of Spyder, the Scientific Python IDE, and other pivotal projects like Python(x,y) and WinPython. These tools have been instrumental in making Python a leading language for scientific computing.

Pierre's academic journey began with an engineer's degree from the Institut d’Optique Graduate School, specializing in laser physics. He further advanced his expertise by earning a PhD in optics and photonics from Université Paris-Saclay, where he developed software for simulating regenerative amplification in ultra-short pulse lasers.

Professionally, Pierre has held diverse and impactful roles. He served as a research engineer at THALES Avionics, a principal software developer at CEA (the French Alternative Energies and Atomic Energy Commission), and managed the Laser MégaJoule timing and fiducial system project. Eventually, he became the Head of a Research Laboratory at CEA before transitioning to Codra, an industrial software company, where he currently holds the position of Chief Technology Officer (CTO).

Beyond his work on Spyder, Pierre is deeply involved in the open-source software community. He has created tools such as guidata, PlotPy, PlotPyStack and DataLab, and has contributed to numerous other projects.