Python Bytes Ep 51: Policing Analytics, PyPy, and IoT

Exploring Public Data with Python

In this episode, the hosts discuss using Jupyter Notebooks, Pandas, and NumPy to analyze public policing data. They emphasize the importance of data literacy and the ability to extract meaningful insights from public datasets while acknowledging the complexities of data cleaning and responsible analysis.

Boosting Performance with PyPy

The discussion shifts to performance optimization, highlighting a case study where code running on a quadcopter simulation was made 80 times faster by using PyPy and careful profiling. The hosts caution that PyPy is not always a silver bullet and may require algorithmic adjustments or specific library handling, such as using NumPyPy.

Sustainability in Open Source

The episode addresses the critical issue of what happens to open-source projects when a maintainer is no longer available. Key takeaways include:
* The concept of a Bus Factor to evaluate project risk.
* Potential integration of mechanisms in GitHub or PyPI for project ownership transfer.
* Encouraging community-owned development groups for libraries.

"There's a lot of them [open source projects] that don't have that many maintainers or sometimes just a handful or one. So really, how do you deal with that?"

IoT and New Resources

The hosts feature a solar-powered IoT sprinkler project built with MicroPython and Home Assistant, underscoring how accessible and low-level hardware programming has become. Additionally, they celebrate recently published Python books and discuss the release of Anaconda Distribution 5, which includes security-focused compiler improvements and the exciting JupyterLab preview.