Amend project license to require citation if used in scientific publication

[philipstarkey]

Original report (archived issue) by Chris Billington (Bitbucket: cbillington, GitHub: chrisjbillington).

---

It is fair to require users of the labscript suite to cite the paper if they publish experimental papers in which the experiments used the labscript suite. This would be in addition to the requirements of BSD 2 clause, but not in violation of the existing license and therefore not requiring permission from copyright holders.

We should suggest users cite our paper in their acknowledgements section so that is is unobtrusive to the flow of their paper whilst still giving us a cite.

This may require discussion but I'm putting this here to remind myself to initiate said discussion.

The license would need to be agreed to in the installer. Unlike typical end-user license agreements, it would actually fit on one screen so this is not so bad.

[philipstarkey]

Original comment by Philip Starkey (Bitbucket: pstarkey, GitHub: philipstarkey).

---

I disagree with this approach and think it's likely to put people off using the software. Mandating citations or acknowledgements seems overly prescriptive, especially given the fact that some journals (particularly high ranking ones) have quite restrictive word/citation counts. Given that I don't even think our own group has followed such a policy, I really doubt this is a good idea.

[philipstarkey]

Original comment by Chris Billington (Bitbucket: cbillington, GitHub: chrisjbillington).

---

Hm, well although there are other examples of scientific software doing this, I suppose science does not do this in general, i.e. you're not legally required to cite Zeeman every time you make a MOT - it's up to you to make a judgement call whether your work is directly related enough to cite whatever you're using.

It would be nice if we could ensure in some way that we get cites at least when methods are being described, it's often the case that groups will have a paper that describes their methods and then they self-cite that paper in all subsequent papers on the same apparatus. We would want to be cited in the initial paper but not necessarily the subsequent ones.

Maybe a friendly request upon installation and in 'about' screens would be useful - along with, a copy-paste bibtex entry in the 'about' screens. So long as it is in people's minds that citation should be treated in the same way you would if you used a scientific technique someone developed, then I'm happy with it not being made an actual legal requirement.

We could add something along these lines that is not legally enforceable to the license agreement still - that citation is reasonably expected when describing other implementation details of an experiment. But if it's not legally enforceable then maybe it doesn't belong in the license. So I'm leaning toward a message in the install script, about screens, and READMEs/sphinx docs.

[philipstarkey]

Original comment by Jan Werkmann (Bitbucket: PhyNerd, GitHub: PhyNerd).

---

In our group we agree that it is fair to require citation in papers that deal with the experimental methods. After all labscript is a important part of the experimental setup. But we also think that all other experimental papers, unless they (heavily) relied on some labscript feature, should not be required to cite the labscript paper. We agree with @philipstarkey that this requirement seems off putting. Keeping the adaption hurdle as low as possible is in the interest of the labscript community as a whole. The more labs use labscript the quicker bugs will be found and the faster the code will be improved.

[philipstarkey]

Original comment by Chris Billington (Bitbucket: cbillington, GitHub: chrisjbillington).

---

Well, I'm convinced. Closing in favour of the 'ask nicely' approach.

[philipstarkey]

Original comment by Chris Billington (Bitbucket: cbillington, GitHub: chrisjbillington).

---

• changed state from "new" to "closed"