Licensing may seem like an obscure legal detail, but it plays a critical role in scientific software development. In “The Whys and Hows of Licensing Scientific Code”, Jake VanderPlas breaks down why picking the right license is key to sharing and advancing research.
Summary of the Article
VanderPlas emphasizes three key takeaways:
Always license your code. Without a license, code is effectively closed, limiting its reuse. If you don’t, it’s basically off-limits for anyone else to use.
Use a GPL-compatible license.This makes it easier for your code to work with other open-source projects.
Prefer permissive licenses like BSD or MIT. These licenses lower barriers for collaboration, they are the most flexible and let people from both academia and industry collaborate freely.
Licensing is crucial for scientific reproducibility and collaboration. Even if you post your code publicly, without a license, it’s still “all rights reserved,” meaning others can’t legally use it. VanderPlas recommends permissive licenses because they encourage more people to adopt and improve the code. On the other hand, copyleft licenses (like GPL) keep the code open but might scare off companies from getting involved.
One thing I hadn’t thought about prior to reading VanderPlas’s article is the fact that licensing decisions can affect how long a project will last. A good choice of license not only encourages others to use and further develop your code but also ensures your work remains useful and accessible over time. For instance, by choosing a permissive license, such as MIT or BSD, researchers open up their software to being included in both academic and commercial projects, thus enabling innovation in many different fields.
Conversely, if code is not licensed correctly, it risks becoming abandoned since others might be wary of working with it due to legal ambiguities. Licensing is not just a legal step; it’s an investment in the future of scientific progress.
Personal Reflection
While reading this article, I found VanderPlas’s insights particularly relevant and important. I appreciate how licensing can help bridge the gap between innovation and real-world impact. The idea of using BSD or MIT licenses makes sense because they’re simple and open the door for more people to get involved.
This also made me think about how intentional we have to be with our work. Just like we carefully document research methods, licensing makes it clear how others can use and improve our code and/or tools. It’s a good reminder that open science isn’t just about sharing, it’s about having solid guidelines that make collaboration easier and that push science forward.
Citation
VanderPlas, J. (2014, March 10). The Whys and Hows of Licensing Scientific Code. Pythonic Perambulations.
Link of the article: https://www.astrobetter.com/blog/2014/03/10/the-whys-and-hows-of-licensing-scientific-code/
Maria Delia 