Licenses

[EstherPlomp]

Based on the information you read in Data and Code Licences, please post your license preference for data and/or code below.

Don't forget to list both a license for the data and the code: the licenses used are different!

[EstherSpruit]

For my work I choose the standard CC By 4.0 licence.
I don't think I will produce code, but otherwise I think [GNU GPLv3] would be good.

[EstherPlomp]

Thanks for sharing! There seems a slight difference in your preference for data and code licenses: for example, CC-BY-SA and GPL are more compatible in their requirements, in that they require the reuser to use the same license. What made you go for GPL and not MIT or BSD/Apache for example?

[abermejillo]

For my data, I think the most appropriate license is: CC BY-NC-SA 4.0
I want what I do to be kept open source (SA) and not commercial.
For my code, I think GNU GPLv3 would work fine.

I am a bit confused as to what to do with the commercial use. Do I want other people to make profit from what I develop for everyone to use? For the code, all licences allow commercial use.

[EstherPlomp]

Both your licenses are indeed quite similar in what they allow reusers to do!

Regarding your question @abermejillo - There is indeed not really a restriction on commercial reuse in the code licenses, however, choosing the copyleft license that you choose may prohibit it nonetheless. This license ensures that any source codes needs to be made available - which companies may not like to do as that means that they would have to share any code that is 'contaminated' with the GNU GPLv3 code. Some companies will therefore avoid reusing software shared under such copyleft licenses.

That is indeed not a guarantee!

[demiteach]

For my data, I believe a CC BY-SA 4.0 would be most appropriate. off course, this would be first discussed with my supervisor as well. For data that includes PI, a different license would be chosen, including NC.
As I only plan on using code (ready: python and/or R) for data analysis, I think I do not really need a license for this; the code will be very general, and it does not contain anything sensitive

[EstherPlomp]

Thanks for sharing @demiteach!

You don't choose a license because the content would be sensitive or not - it is actually quite the opposite in the sense that licenses would only apply to content that is publicly available. A license ensures that others know what they can do with the data/code. Without it they can't do anything with it without having your explicit approval. This hugely hampers any reuse of your work!

If data you generated can be used for a patent application, always wait with sharing it until the patent applications are finalised. This will follow a similar process as any publications related to the patent applications.

And yes, always discuss these things with supervisors!

Taking this into account, what would be your preferred license for software?

[demiteach]

Thanks for your answer!

In that case, I think that Mozilla Public License 2.0 would work fine. If they directly use the same code, it would be nice to get referenced for it. Please tell me if I am wrong.

[EstherPlomp]

You're not wrong, and choosing this license may bring a bit more work from your side. TU Delft has pre-approved certain licenses -
MIT, BSD, Apache, GPL, AGPL, LGPL, EUPL, CC0, which can be used without further work. If you want to use a different license it is still allowed, but then you would need to get in touch with the innovation and impact center and may need to go to more formal procedures. So you may want to go for MIT/BSD instead of going through that, but this is entirely up to your preferences!

[irisforkink]

I believe that for my research CC BY-NC 4.0 is appropriate. I am a bit unsure if others can use/modify my research? My research is funded by a company and officially I think that, if the company wants this, the research 'is theirs', how does this impact my license?

[EstherPlomp]

@irisforkink CC BY-NC 4.0 does allow modification of the original material, yes!

If your research is funded by a company it means that they will have a major say in what can happen with the data/code and publications of your research. That means that if they do not want you to share the data, you will not be able to do so. If you are allowed to share the data, you will have to consult them to choose the license. Once the data is then published/shared, anyone can use it under the requirements of the license, including yourself!

I hope that clarifies it, if not, please let me know so that I can try again :)

Will you also share any scripts/software? What would be your preferred license for this?

[lkresik]

CC BY 4.0 for my data. And if any scripts will be generated they will most likely have The Unlicense, or one of the simplest licenses.

[EstherPlomp]

Thanks for sharing! Do note that the Unlicense is not one of the preapproved licenses of TU Delft (MIT, BSD, Apache, GPL, AGPL, LGPL, EUPL, CC0). So you may want to go for MIT/BSD instead! (It is possible to go for a different license but then you have to sign some forms and contact the innovation and impact center!)

[lkresik]

Thank you! I will keep that in mind.

[kmertiri]

If we would publish our code for the numerical calculations or simulations, I think a Free open source licence would be appropriate.

[EstherPlomp]

Wonderful! @kmertiri: Which open source license would you prefer and why?

[kmertiri]

I would prefer the Mozilla Public License, since it gives users the freedom to customize the code while still requiring the release of any modification.

[EstherPlomp]

Thanks for sharing! Do note that the Mozilla Public Licence is not one of the preapproved licenses of TU Delft (MIT, BSD, Apache, GPL, AGPL, LGPL, EUPL, CC0). It is possible to go for a different license but then you have to sign some forms and contact the innovation and impact center!

[RiteshDas2000]

As a theorist, I am not producing "Data" in the strictest sense. My data would be the results and analysis of my simulations and calculations. Similar data is most often licensed under the CC BY 4.0. I agree and believe my data is best utilized by other academics under the CC BY 4.0's open source nature while still doing justice to myself as the author.
For similar reasons, for my code, I would choose the GNU AGPLv3. The main reason would be to give users who interact via network to receive my code, as that is the main evidence and product behind my work.

[EstherPlomp]

Sounds great! Thanks for sharing!

I am wondering why you would prefer CC-BY above CC-BY-SA (which is more similar to the software license you prefer in a sense, as it requires redistribution under the same license). Why this difference between data and code?

[RiteshDas2000]

Hi @EstherPlomp. You are right that the nature of the licenses has to do with the nature of the data and the code. My "data" is, of course, the result of theoretical simulations that can be accommodated into a variety of models. This results in a vast number of potential applications, including commercial ones. Thus, I do not want to limit the applicability of my results. My code on the other hand, is just a means to run the simulations, barely considered copyrightable. The main advantage of sharing the code is to make others' lives easier. As such, I believe keeping the code fully open source would not serve to limit its applicability in any way, and will in fact, expand it.

[anish8194]

The majority of my work is going to be in producing simulations hence a CC by 4.0 license would be appropriate since as per the EU funding agreement we will have to make the code publicly available. It might be good to get cited if someone uses my code directly or builds on my work.

[EstherPlomp]

Indeed, EU funding places more stricter requirements on sharing! CC-BY 4.0 is in line with these requirements, and works well for data. This license is generally not used for code however, so you may want to check which pre-approved TU Delft license you would prefer (MIT, BSD, Apache, GPL, AGPL, LGPL, EUPL, CC0).

[anish8194]

Thanks for the clarification. Yes, I think MIT license will be the best since most in my field of research with code availability use MIT license so I will also stick with it for the codes.

[Nico7575]

Since my code involves the design of a nuclear reactor, the code will not be open source. In this case, I do not think I can choose a license, do I?

[EstherPlomp]

If it is indeed to sensitive to share you will not be able to share the code indeed. Will you be able to share any data, and if is, what license would you prefer here?

[Nico7575]

Thank you @EstherPlomp for the answer. I will be able to share the data from simulations analysis (so performance and behavior of the system), because these will be the subject of my publications. For those I think the CC BY-NC-SA 4.0 license would then work, so that I get cited if someone is using my work, other people in the research group can build up on it and it cannot be used for commercial purposes by others. To be sure, I will ask my Supervisors about this, so that I can be certain about what to put in the final RDM plan.

[mmsavic]

Based on my work, I think that CC BY-SA 4.0 licence would be the most appropriate. I'm not completely sure about using the data and/or code for commercial purposes, as that would need to be discussed with my supervisor.

[EstherPlomp]

Indeed, always discuss your license preferences with your supervisors or other collaborators! Will you also share any scripts/code and if so, what license would you prefer for this?

[linghangdelft]

I would use CC BY 4.0 license for my data, it can be used for commercial use and modification as long as it's attributed to me. For software, I would use MIT license because it's approved by TU Delft and relatively explicit. But in general, my research won't cover coding or having software as research output.

[EstherPlomp]

Sounds good, thanks for sharing!

[BhaveshSChavan]

For my research work, which is based on consortium and only companies involved can use my work or data. In such case I would use CC BY-NC 4.0 license.

[EstherPlomp]

Sounds good, thanks for sharing! 👍

[LonghengZhang]

I think CC BY-NC-SA 4.0 is the license that is suitable for our experimental data. We only write simple code, do we still need license for software?

[EstherPlomp]

Thanks for sharing! Even if you only share a simple script/code you would need a license - otherwise no one would be allowed to reuse it when you share it: so please indeed still share your software license preference!

[LonghengZhang]

Hi, I just discuss with other members in my group! We use MIT license for the code because we want to make it open.

[EstherPlomp]

Nice! Glad that you had a discussion about this and thanks for also sharing it here!

[rjhvanderkolk]

Computer says : CC BY 4.0
Creative Commons Attribution 4.0 International
This license requires that reusers give credit to the creator. It allows reusers to distribute, remix, adapt, and build upon the material in any medium or format, even for commercial purposes.

[EstherPlomp]

Thanks for sharing @rjhvanderkolk! Do you also have a preference for a software license?

[oonimabisht]

For my experimental work, I think that CC BY-SA 4.0 licence would be the most appropriate. I do not know what software license should I go for, since we use our codes only to plot our experimental data. Do we need software license for that as well?

[EstherPlomp]

Yes, because that is the method you used to generate the plot, so ideally you would share that together, but with a different license for the software. What license would you prefer for software?

[DaphneCette]

I think that the CC BY-SA 4.0 license should be appropriate for both my experimental and modelling work.

[EstherPlomp]

For modelling/code the CC-BY or CC BY-SA license is generally not used - so do have another look at the software licenses to decide what license you would prefer for sharing any software/analysis scripts of the research! @DaphneCette