[gmx-users] Thesis developed software; which Open-source License?

[p.c.kroon at rug.nl]

Hi Henry,

Although I'm not a lawyer here's how I understand it. You may want to check it with your department/university/company's legal department.

I don't know anything about licensing data, but take a look at creative commons licenses if you want it to be open.

See also: https://opensource.org/licenses/
Relicensing (from BSD-3 to something else, for example) can be tedious, depending on how many people contributed to your code. Everyone who has a claim to the IP (your code) has to agree with the change. If you're the only contributor, or if all contributors are nearby, it can be pretty straightforward. It's also possible (I think with the same conditions) to have different licenses for different users. Either way, users of your code do not need to be notified. They can keep using your code under the BSD-3 license under which they obtained it.

BSD-3 and MIT are very simple and permissive licenses.
Apache2 is also a very permissive license, but there's a bit more legalese involved, reducing ambiguity. This is the license I chose for my FOSS projects after consulting with my uni's legal department.
(L)GPL and friends are so-called copy-left licenses. Whenever you use GPL licensed code in your project, your code also *must* be GPL licensed. This becomes a little fuzzy when you're linking against GPL libraries, or importing them in your python code. For this reason LGPL was made, which explicitly allows compiling against LGPL licensed libraries without making you also GPL license your own code.

As I see it, as long as you haven't looked at GPL licensed source code you can license your code however you want. I *think* scripts that just call gmx tools do not have to be GPL licensed. I personally consider parameter files to be data (and not code subject to the LGPL license). So using/adapting those should (I think) be ok.  I don’t know which "original GMX simulated solvated protein data" you mean, or how that is licensed.

As I said I know very little about the licensing of data, and I think that in general it's something to which not enough attention is being paid (at least up until now). Has anyone looked up the license for the RSCB PDB for example?

I hope this helps, and that it's not too factually incorrect 😊

Peter

-----Original Message-----
From: gromacs.org_gmx-users-bounces at maillist.sys.kth.se <gromacs.org_gmx-users-bounces at maillist.sys.kth.se> On Behalf Of Henry Wittler
Sent: Friday, February 21, 2020 10:22 AM
To: gmx-users at gromacs.org
Subject: [gmx-users] Thesis developed software; which Open-source License?

Greetings comrades. This question may be somewhat redundant, so dont expect reply to every aspect, may ask at RG also. However if anyone have more or less insight, please reply.

My thesis has calculation software that are built upon partly innovative and partly basic calculation/plotting/graphing software based upon "VMD/tcl” coding, GROMACS (GMX) and Python/MatplotlibScipy/Numpy (etc).
The solvated protein data (chap. 6 of thesis) are obtained from GMX simulations, moreover the analysis data by VMD/tcl etc.

I have uploaded so far just part of the code (for chap. 4.1) I intend to upload code from whole thesis to github link https://wittler-github.github.io/A_MD_Analysis_of_Insulin/
The following thesis link(downloadable) contains the calculated data along with the graphs.

http://hdl.handle.net/1959.9/568798  (
https://www.researchgate.net/project/A-Molecular-Dynamics-Analysis-of-Insulin)


I mainly want to share the novel GMX simulated data and tools I’ve developed from the above-mentioned softwares, that are described in my thesis. I do not expect to contribute to GMX directly from this repository, possibly via other created repository if so.

Currently I have applied the BSD-3 clause version, which I understand to be the most relevant and simple to use. Choice is not set-in stone it appears, however if changing I understand one need to notify everyone that has taken part of, or are using the code.
The BSD-3 do appears to be one of the most popular for open-source code and compatible between different softwares. I thought of using LGPL2 since GMX uses that, however BSD-3 does appear more simple to understand. The only innovative scripts including GMX are just linux bash/automatic scripts that make the simulation of replicas straightforward. Can I share these script under the BSD-3 license, even though technically they are modified scripts using standard gmx commands (not altering the original gmx software v.5.0.4)? Can I share some of the GMX code (parameter files etc) and some of the original GMX simulated solvated protein data (in addition to data calculated/graphed by matplotlib etc), at my github repository with no issues between BSD-3 license and LGPL2?

The other python, and VMD/tcl softwares I do not see any issue with. Are there any clash with these above-mentioned softwares to be aware of?

Any other insights anyone else has here about open-source licensing when distributing code?

*Kind regards,*

*Dr. Henry P.A. Wittler*

*Department of Chemistry and Physics, LIMS, La Trobe University, MelbourneWorking in Ludvika, Sweden*
*Skype: henry.wittler*
Researchgate & Linkedin
linkedin.com/in/henry-per-andreas-wittler-b03256191
<https://www.linkedin.com/in/henry-per-andreas-wittler-b03256191>
--
Gromacs Users mailing list

* Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-request at gromacs.org.