What are issues to consider when making your code for an experimental paradigm, which was conceived by someone else, publicly available?

Question

General background

In the cognitive sciences, it is often necessary to program/script experimental paradigms to conduct your studies. Basically, these programs present stimuli and collect responses, such as in the Eriksen Flanker task. However, programming capacities are typically limited in my field because it is rarely taught during graduate education (at least in my country). Also, it seems there is a somewhat reluctant culture of sharing this software unless there is some sort of scientific cooperation.

Consequently, when designing a study the decision for one of several paradigms used to measure a given construct can, among other things, be motivated by availability of the software or the effort necessary to implement it. With respect to sound science (choosing the most appropriate rather than available paradigm, software bugs, etc.), reproducibility (incomplete descriptions of methods, lack of programming skills, etc.), speed of scientific development, and, more generally, open science this is less than optimal.

The situation at hand

For a current study, I have started to implement three paradigms from scratch for different reasons:

• One paradigm is publicly available from the authors but implemented in a commercial development environment for which I don't have a license (I, therefore, can't run it).
• Another paradigm, to my knowledge, is not available from the authors at all. I contacted the authors for a copy of the software but got no reply.
• The third paradigm is commercially available but is also closed source and only grants access to a limited set of the response data. For my analyses, I'd like to have access to richer data.

Given my prior general considerations, I think about releasing my implementations of the paradigms under an open source license, once they are finished. Now my question is:

What are issues to consider when I release my software to collect data in an experimental paradigm, which was conceived by another scientist, under an open source license?

Here are some of the things I'm wondering about:

• Are there legal issues, such as copyright infringement, to be considered for commercially or non-commercially available software? Note, I created all of the code and all images based on low resolution prints in publications. I'm merely copying the concept.
• Would such a release conflict with common etiquette in the case where the software is simply not available from the original authors (remember, I asked but got no reply)? Is it necessary to ask each author before releasing my code?
• Can I be held responsible morally if the software is used and turns out to be faulty? I'm less concerned about legal responsibility because the license I intend to use does not provide warranty. More importantly, I certainly intend for the software to work since I'll use it my self but you never know.
• If my study ends up being published and points to the availability of the software, can this benefit the citation count of the publication?

I'm sure there are more things to consider that I haven't thought of. But that's why I'm asking the question. ;)

Answer 1

Are there legal issues, such as copyright infringement, to be considered for commercially or non-commercially available software? Note, I created all of the code and all images based on low resolution prints in publications. I'm merely copying the concept.

Unless their method is patented, no. Software is covered by copyright which protects a given implementation, not the general idea. Just as you are allowed to write an own novel on a topic that is covered by existing novels already.

Would such a release conflict with common etiquette in the case where the software is simply not available from the original authors (remember, I asked but got no reply)?

No.

Is it necessary to ask each author before releasing my code?

You need to get permission from all copyright holders. Depending on the actual settings, that means: for each author who contributed to "your" code the acutal holder of the copyright (could be the author or the author's employer).

The authors of the "original" software are not involved at all.

Can I be held responsible morally if the software is used and turns out to be faulty? I'm less concerned about legal responsibility because the license I intend to use does not provide warranty. More importantly, I certainly intend for the software to work since I'll use it my self but you never know.

• moral responsibility: just like you will be held responsible for any other scientific publication.

• legal responsibility: while no liability clauses are common for FOSS licenses, completely excluding liability is not possible in all legislations (e.g. in Germany a general no liability at all clause is/can be invalid).

If my study ends up being published and points to the availability of the software, can this benefit the citation count of the publication?

Definitively, yes