Contributor Role Taxonomy to specify what people did

[wkerzendorf]

I like this contributor role section in my papers because there are so many ways to contribute to science and it gives a clearer picture of what everyone did. Maybe something for the Astropy paper.

[wkerzendorf]

@pllim I should also say - one can just remove the parts that do not apply.

[nmearl]

Hey @wkerzendorf, can you expand a bit on how you're imagining this taxonomy would apply? Are you suggesting that authors/contributors be categorized in this way? There are quite a few that don't seem particularly relevant, and it's not clear if this is meant to clarify the subject of an author's contribution only recently, or over the lifetime of the project?

I'm not quite sure that this provides anymore specific context than simply having everyone be considered a paper author, since many of the paper contributors who have responded would fit into multiple categories. Likewise, I feel like we'd need to add explanations of what each category means for the taxonomy to provide useful context.

Any clarification would be helpful on how you've seen this approach used/applied in other papers (or your own)!

[wkerzendorf]

I didn't come up with these categories - this is used by nature and science and other papers https://credit.niso.org/ So not all of them are relevant - one can just remove the non-relevant ones. People are in multiple categories and that is fine. So your name can appear in many. I would say why not over the lifetime over the project. I've used this here: https://arxiv.org/pdf/2105.07910.pdf (my students paper).

[adrn]

Thanks @wkerzendorf -- this is an interesting idea that I haven't seen before, but I don't think it scales well! I think we would end up essentially copying the author list into "Software"... So, maybe best for smaller collaborations?

[wkerzendorf]

@adrn there would be many people in software but some people in software and writing draft and editing draft. Maybe we can have a section "education". BUT ... this was just a suggestion and if that doesn't fit - don't worry and just close it.

[eteq]

To me the purpose of the "taxonomy" approach in journals like science or nature are to help communicate what the conventions are in a given field about e.g. "first author is PI" (e.g. chemistry) vs "first author actually did the work" (e.g. astro). It's not as clear to me we need to do that in a paper that's intended more for "within our domain" consumption.

I also think it's highly problematic to try to put people in external boxes like this given that in Astropy many people make many roles, and we already have our own taxonomy suitable for our local conditions (i.e., the teams page).

That said, I agree with core idea that we want to give the reader some way to know all of the above! So I'd be inclined towards a much simpler taxonomy that's basically "writing", "reviewing", "non-paper astropy contributions (see the team page)", which we have historically done by setting the author order to be "alphabetical in groups". I think as long as we explicitly state that in this spot we don't need the more complex taxonomy here.

[adrn]

Thanks for the idea @wkerzendorf -- we discussed this amongst the paper coordinators and decided against this, at least for this article.