What would be a good (open source) license to use for reproducible research code? Based on my current experiences, I would go with the GNU General Public License (GPL). To my understanding, this license allows others to reuse and modify the code (which is the whole purpose), but any redistributions have to stay under the same license. Something I found out recently is that I (as the author) could in such a case still distribute it to some people under another license in parallel. So this would allow me for example to give a license for commercial use to a company.
Any other opinions on which license to use, or their (dis)-advantages?
This question might be considered from the applicative point of view. Maybe the type of license should be chosen according to the number of constraints it has against the number of potential uses the software may have.
For example, the lovely people at NIST who put some linear algebra software in the public domain were right. Everybody uses linear algebra, and accurate algorithms are not that trivial to code. There was also this story of Stallman first advocating for LGPL but changing his mind for GPL for now. IMO, LPGL would still be good for research, as our main problem here is to reproduce results, and thus to have access to the full source code. Sorry, Richard. But I agree GPL or BSD are better from an ethical point of view.
Moreover, the thin line between original and stolen code is not that clean and I would love to know when a code shall be considered different from another... Does that imply the very implementation of an idea is different, or that the (whole?) idea is different? What is the difference between an idea and its implementation? Sorry I did not study law!
I currently use Matlab for the code, and have scripts called figure_x.m to reproduce Figure x from the paper. While this does require a user to have Matlab, and is not a single command for all results, I consider Matlab to be quite standard for fellow researchers, and this allows me to make my code available with a minimum of additional efforts.