0

I'm considering going back into school to get my graduate degree. A large part of my decision is the ability to Open Source all of my code. I'm really not familiar how the idea of IP reconciles itself with University as a concept, my experiences lie in corporate environments, where IP is a closely guarded secret. From what I've seen Open Source as an idea isn't yet a large part of the research process, though the process has increased in popularity considerably.

If I were to attempt to publish research to a journal where the research relies heavily on code, what challenges would I face if I made my code open source? Would there be any additional challenges if I shipped the examples from the paper in the source? What if I put the paper itself into the project as part of the source (IE releasing it under a copyleft license)?

2
  • What does it mean to "ship an example"? Nov 29, 2021 at 16:02
  • This depends very much on the policies of the university and may also depend on the source of funding for your graduate studies. Dec 1, 2021 at 0:14

2 Answers 2

4

It is standard for academics these days to release source code and other creative works as open source. That’s fully consistent with how academia works and how university IP policies are structured — such policies typically claim ownership for the university of patentable inventions, but leave the researcher as the owner of all copyrightable material.

In mathematics and several other disciplines, it is also standard these days to upload a preprint version of every article one writes to the arXiv or other paper repository, under a Creative Commons or similar license. So, even research papers themselves are largely open source and open access (although this is not the norm in all disciplines, and in the disciplines where it is the norm, it applies to preprints but not always to final versions of published papers). Not all researchers take advantage of the freedom to make their work available in such a way, but almost all have at least the option to.

In short, you will encounter no challenges and have absolutely nothing to worry about.

1

First, code produced as part of academic research is not automatically guaranteed to be open source. Universities, like companies, have technology transfer offices, and if you create a very influential piece of software (like one that could have patented parts, if you can do software patents) the university technology office may want you to keep it closed-source so the university can license it and the patent.

That said, I think it is very rare that universities will exert that control over software products (they usually do that over the other results like new materials, processes, etc.).

There is a growing trend to release code with papers (but not typically as part of the paper text itself) to aid in the reproducibility of the results. This is usually done by adding it to archival services (like Zenodo or Code Ocean), or through a different method chosen by the publisher (some are still zip attachments in the supplementary materials).

There is also a growing movement for "software papers," which are papers that describe and document a software tool that other researchers can use (such as a toolbox of analysis methods, simulation frameworks, etc.). This kind of paper can be more difficult to get into the large journals of a field unless they have a section more dedicated to experimental/research methods, but there are some software-specific journals like JOSS (Journal of Open Source Software) that publishes them.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .