What exactly would count as copyright infringement in a replication paper?

Question

I have been following the review process of a replication journal

In particular, it has an open review which can be seen.

On a particular submission they are having a debate about whether copying equations and their explanation in a paper counts as copyright infringement.

I would like to ask this question here, since the answer seems non-obvious.

Equations, by themselves, since they are ideas, should be free from copyright infringement. But to have equations, explanations, replica of figures, and discussion, would basically be a copy of the paper, and I imagine that would be copyright infringement. Where exactly can a line be drawn?

Note: There are questions on stack exchange which ask similar questions, but often in context of building up on previous research. Since the point of such a journal is just to replicate, it would seems that the aim is to build a freely available copy of the existing (perhaps paywalled, copyrighted) material, that can be freely accessed, and this is different intent than regular articles.

Also, answers regarding plagiarism aren't much help in this case as the point is to do "explicit plagiarism" with proper attribution.

Crossposted to law.SE

Answer 1

After asking permission to Elsevier (I'm editor in chief for ReScience), they confirmed (in this specific case) that equations can be re-used without asking permission to Elsevier:

Our Global Rights group has confirmed that these equations are not copyrighted; therefore, you can merely acknowledge the 2012 Journal of Theoretical Biology as the source of the equations but do not require written permission from Elsevier.

Answer 2

From what I have learned about copyright law (IANAL), the answer is non-obvious to the point that even a lawyer could only guess. Leaving aside the issue of how jurisdiction would be determined for the moment, and assuming that the US copyright law (and in particular Fair Use) applies: The problem is that the way copyright law is structured, copying the explanation of the equations would constitute copyright infringement on its face. Fair Use is merely a defense that may be applicable, and it depends on the balancing of multiple factors (the four factor test) which is by necessity somewhat subjective. The only way to truly definitively determine whether a specific act of copying is exempt by fair use or not is to have a court render judgement on this specific use.

It may be that in practice, this specific case is clear-cut enough that a lawyer could with confidence predict based on precedent if this case would be covered by fair use or not, but for that you would have to ask law.SE (in fact, you should probably ask there to make sure there aren't any errors in the above as well).

Answer 3

Great question. Firstly, I am not a lawyer and for anything approaching a formal legal answer you'd need this to be migrated to Law.SE. With that caveat, I'll have a go.

Original content is covered automatically by copyright. Under the 1886 Berne Convention which first established the recognition of copyrights among sovereign nations, copyrights do not have to be asserted or declared, as they are automatically in force at creation: an author need not "register" or "apply for" a copyright in countries adhering to the Berne Convention.

Certain categories of work cannot be copyrighted. For example, a list of the materials required for a protocol cannot be copyrighted, at least under US copyright law (reference). However, the protocol itself can be.

Usage of copyrighted material is permitted under certain conditions, such as fair use. I'll probably expand on this later, but basically while fair use is not clearly defined it could fall within the usage described here.

If you did that, you should make it clear what part of the M&M in your paper is quoted from the source (and clearly attribute it).

For categories of work that are covered by copyright, unless covered by e.g. fair use reroduction of a substantial portion of the work is prohibited by the laws of copyright without a special license.

A free license or open license is a license agreement which contains conditions permitted to the user from the holder on a specific list of uses for his work. As originally defined (and thus with respect to software, which was the original target of these licenses) a license is free if it gives users the following freedoms:

• Freedom 0: The freedom to run the program for any purpose.
• Freedom 1: The freedom to study how the program works, and change it to make it do what you wish.
• Freedom 2: The freedom to redistribute and make copies so you can help your neighbor.
• Freedom 3: The freedom to improve the program, and release your improvements (and modified versions in general) to the public, so that the whole community benefits.

(Freedom 2 is the most relevant for our purposes.)

Licenses meeting these criteria and commonly in use by academic publishers include Creative Commons licenses such as CC-BY-4.0 (which is used by all PLOS journals).

TL;DR: the only way to guarantee you won't get sued for copyright infringement is to target studies published under open licenses, or get explicit permission from the copyright holder.