Read one or more of the absolute best (free) Python books[1] and do some hands-on.

If you can not choose, just start with "Python Programming for Data Science". No barrier for starting, its cc-by and requires no account to read.

[1] Check: https://nocomplexity.com/documents/pythonbook/bookreferences...

Too bad that the focus is open source. This is a common ailment seen in many security research papers!

A real concern should be Code-Level Backdoors in COTS software. Backdoors that even many resellers and paying customers are not aware off..

Nice deep insight. I love the personal story at the start!

Is this worth watching?

Full title: "Type Hints in Python Libraries and Frameworks: An Empirical Analysis of Adoption and Maintenance"

Conclusion: Type hints in Python libraries and frameworks primarily serve as API contracts rather than comprehensive descriptions of implementation details.

I come from C/C++, so for me using Python without type hints is the way forwards. But for some libraries it make sense. As long as type hints "do nothing"...but only for some tools.

" In this paper, we set out to fill this critical gap, focusing on real-world native code bugs in native Python applications (i.e., software written in Python and C as primary languages, a.k.a Python-C software). We start by collecting and labeling the first set of 1,039 automatically mined and then manually confirmed native-bug-fixing commits from 200 Python-C projects on GitHub. "

I have serious doubt about this study. It's does not meet crucial reproducible research criteria. And most real-world Python-C Application are not FOSS and available on a public repository. But doing research on public repository is easy.

This paper focuses solely on public repositories. This is harmful and fuels the inaccurate perception that open-source software is inherently buggy, a view still promoted by commercial vendors and adopted by many managers worldwide.

Nice improvement! Good reliable statistics for Python package use is a real challenge! So best is to use multiple indicators to get a rough indication!

"This change removes one source of potential confusion and misinterpretation, but does not mean download counts are an accurate measure of how many people use a package."

Some context and nuance: https://support.mozilla.org/en-US/kb/use-primary-password-pr...

Nice insight in cost involved with a huge free to use systems, Drupal!

I think it is always better to make a clear distinguish between FOSS and OSS. Just to avoid these kind of issues in discussions again and again!

Great to see this! Nice work.

Thank you! :)