Contentions about shouldacite?

[katrinleinweber]

Noticing this meeting note, I am curios what the exact contentions were?

force11-sciwg/meetings/20190116-Guidance.md

Lines 83 to 86 in f11835b

	* Existing guidance whether a software shold be cited: http://mr-c.github.io/shouldacite/index.html
	* This list proved contentious

To those who raised them, please consider posting criticism & suggestions to https://github.com/mr-c/shouldacite/issues (if not done already) so that they can be concretely improved there.

I find that checklist very practical and am considering to use it for teaching best practices around scientific software projects.

[danielskatz]

For what it's worth, I disagree with some of this, for example, I agree with mr-c/shouldacite#2
This guidance is different than the software citation principles, as well.
And finally, it's not clear that this is really active.

[katrinleinweber]

Well, there was a chance to elicit activity ;-) Also, it's obviously a bit different from the CSPs because it was written a few months before and intended as a handy checklist.

[mr-c]

Hello!

Shoudacite is not a funded project, true.

Making it available for community adoption would be my vote, but I'd have to check with my co-creators.

However the license shouldn't get in the way of adapting it for differing principles. As for mr-c/shouldacite#2 I've issued a PR to clarify the subject of the guide (software that is used, not for a piece of "related" software)

[mr-c]

@katrinleinweber shouldacite is not purely nor primarily my product, can you rename this issue to remove @mr-c's ?

[ChrisJefferson]

I commented on "shouldacite". I really don't think it should be used for "best practices", as Q1 (do the authors want citing) and Q4 (do the authors need citing for money), are both really terrible reasons to decide if you should cite something.

The list should be (in my opinion) just Q2 and Q3.

[mr-c]

As a reminder, the context of the "shouldacite" guide is software one has used in ones research. If you used it, there is little reason not to cite it, except if it was a piece of commodity software that is freely interchangeable for another. The "shouldacite" website is a guideline to help authors get to "yes" except for this one case (and to help them figure out how to cite the software).

If you are using software in your research that asks you to cite it, and you disagree with that request, then you should probably use another piece of software. I don't see Microsoft Excel requesting software citations anytime soon.

Likewise if the software you use is from the academic world with an explicit mention of public funding, then even lacking a specific request to cite it, you should do so. Again, if you disagree, then you should probably use another software package.

In either case, if you can't find a replacement software package, then it is likely that this software is critical to your research. Therefore it needs a citation :-)

[ChrisJefferson]

Citations are not bonuses to give out to people who really want one.

If a program deserves a citation, because it was important to your project, it gets one. If it doesn't, then it doesn't. It does not matter if the authors of the software want citing, hate citing, are a single researcher, a huge team, Microsoft or Facebook.

[danielskatz]

Well, there was a chance to elicit activity ;-) Also, it's obviously a bit different from the CSPs because it was written a few months before and intended as a handy checklist.

Thanks @katrinleinweber - I was not familiar with this site, or its history. I still don't really know who runs it, or how active it is.

[mr-c]

@danielskatz It is a product of @louisepb @dimazest @izaromanowska and myself from the Collaborations Workshop 16 Hackathon. It is not part of a resourced effort and has essentially stayed the same since launch: https://github.com/mr-c/shouldacite/graphs/contributors

[danielskatz]

As a reminder, the context of the "shouldacite" guide is software one has used in ones research. If you used it, there is little reason not to cite it, except if it was a piece of commodity software that is freely interchangeable for another. The "shouldacite" website is a guideline to help authors get to "yes" except for this one case (and to help them figure out how to cite the software).

I agree with this.

If you are using software in your research that asks you to cite it, and you disagree with that request, then you should probably use another piece of software. I don't see Microsoft Excel requesting software citations anytime soon.

I don't understand why the author's request matters. Why is this any different than citing papers, which we don't only cite if author wants us to?

Likewise if the software you use is from the academic world with an explicit mention of public funding, then even lacking a specific request to cite it, you should do so. Again, if you disagree, then you should probably use another software package.

Again, I don't understand how this matters. Again think about the parallel with other works, such as papers or data. Cite them because you used them and they were part of your work, not because of which institution the author is at or who funded them.

In either case, if you can't find a replacement software package, then it is likely that this software is critical to your research. Therefore it needs a citation :-)

I agree with this, too.

[danielskatz]

@danielskatz It is a product of @louisepb @dimazest @izaromanowska and myself from the Collaborations Workshop 16 Hackathon. It is not part of a resourced effort and has essentially stayed the same since launch: https://github.com/mr-c/shouldacite/graphs/contributors

I knew I should have stayed for the hackathon day :)