Joss: JOSS site should specify content license

[arfon]

Currently we don't say what license covers JOSS papers. My suggestion is CC-BY https://creativecommons.org/licenses/by/3.0/us/

/ cc @openjournals/joss-editors for thoughts.

[kyleniemeyer]

Yeah, agreed.

Perhaps we should use the 4.0 international version (https://creativecommons.org/licenses/by/4.0/), although tbh I'm not sure of the difference.

On May 13, 2016, at 12:20 PM, Arfon Smith [email protected] wrote:

Currently we don't say what license covers JOSS papers. My suggestion is CC-BY https://creativecommons.org/licenses/by/3.0/us/

/ cc @openjournals/joss-editors for thoughts.

—
You are receiving this because you are on a team that was mentioned.
Reply to this email directly or view it on GitHub

[jakevdp]

👍 for some flavor of CC-BY

[katyhuff]

If the code is GPL or some other copyleft type license, is the paper a "derivative work" and therefore bound to be GPL (or whatever)? I assume not. . . but I am not sure. Anyone know? That might limit our ability to accept papers concerning GPL code...

[katyhuff]

(I am also in favor of CC-BY... I use 4.0 for my own stuff... but only because it seems most up to date. I also don't know the difference between 3.0 and 4.0.)

[kyleniemeyer]

Oh gosh I hope there aren't any GPL issues... Although I would think that if other software papers that rely on GPL software don't need to maintain GPL-ness, then we wouldn't either.

[kyleniemeyer]

Maybe we should recommend permissive licenses for things to be published in JOSS just to avoid the question...

[jakevdp]

If I've learned anything from all my reading-up about software licenses, it's that _nobody_ actually knows what "Derivative Work" in GPL means.

[arfon]

If the code is GPL or some other copyleft type license, is the paper a "derivative work" and therefore bound to be GPL (or whatever)? I assume not. . . but I am not sure. Anyone know? That might limit our ability to accept papers concerning GPL code...

I'm pretty sure that's not the case but I'll get some professional advice...

[arfon]

Also, seems like CC-BY 4.0 is a good choice here.

[acabunoc]

From what I understand, CC BY 4.0 is preferred - it works internationally among all jurisdictions

[labarba]

Open-source licenses (GPL, MIT) apply only to code, not to other content like a paper. Similarly, CC licenses apply to content like a paper (and any other creative works) but not to code.

JOSS should state clearly that the copyright of papers remains with the authors and that they license it under CC-BY.

[katyhuff]

@labarba is totally right.

[arfon]

OK, I've had some great feedback on this from a colleague at GitHub. So here's the proposal:

• We clearly state that authors retain copyright and papers as licensed as CC-BY
• If authors choose to write code snippets in their papers (I'd argue we don't ever want this as we recommend that people leave API documentation in the docs) then the code snippets are MIT-licensed (regardless of the license of the core software package)

BTW, this is similar to the approach that Stack Overflow has taken with code vs content.

Thoughts?

[arfon]

OK, I've attempted to address this in https://github.com/openjournals/joss/pull/122. Comments/improvements very welcome 😸