1

I finished a code-project that's based on several papers. One of those papers has many errors in it's mathematical formulas, but presents an idea I believe to be unique and have implemented.

This paper therefore needs to be cited. However, I worry that people might read it and believe I implemented the erroneous math or base their own work on this paper and run into the same difficulties I had to find and overcome.

What I'd like to do, but could not find a common standard / best practice for, is comment the relevant reference, e.g.:

[1] Exemplarius, Journal Of Samples, 2005 (This paper presents a novel idea, but contains many mistakes in its derivation and execution. Cross reference any formula with other papers!)

I don't think a full discussion of the paper in the text is warranted, nor do I think adding this snippet to any mention deep in the documentation is sufficient.

Improve this question
New contributor
GammaSQ is a new contributor to this site. Take care in asking for clarification, commenting, and answering. Check out our Code of Conduct.
2
  • 5
    I don't see any problem with stating this in the body of the paper, or in a footnote. If you want your readers to know something, then tell them in a place where they expect it, which is certainly not in the bibliography. Commented 8 hours ago
  • 2
    And be a little less harsh in how you write this. Instead of pointing out “many mistakes,” write something gentler like “we implemented a version of the paper formulas that corrects some oversights and errors.” Commented 6 hours ago

2 Answers 2

Reset to default
4

The reference list/bibliography only means you cited something, not that you endorse or agree with it. The proper place to discuss, critique and possibly warn about a reference is the place of the text body in which it is actually cited.

If you feel this isn’t appropriate in the text body, then it’s definitely not appropriate in the reference list either.

Improve this answer
2
  • Does this hold for scientific code / software projects as well? At least personally, one of the first things I look at is the bibliography to get a better understanding of what approach the project is trying to take. Commented 8 hours ago
  • 1
    @GammaSQ If you want to have a scientific code / software project, instead of just a code / software project that happens to have scientific applications, by all means follow established practice of scientific writing. I don’t see why a bibliography would work differently for code/software; you would equally cite prior work that is worse and not used anymore, wouldn’t you? Commented 7 hours ago
0

I don't know your field, but you mention you're reading papers with mathematical formulas. I've read many mathematics papers that cite a previous work and acknowledge directly that that work constains mistakes, say where the mistakes are, and explain how it is still OK to use the formula/Theorem statement that they are using (e.g., by re-proving a particular formula in a different way, or by clearing up a typo in the original formula). As long as you do this in a respectful way, and check very thoroughly that the errors you are pointing out are actually errors, this is a very positive thing for the community.

The important part is to not do this in the bibliography, but rather in the body of the text where you mention the citation, just like how you would discuss any other reference in any other context.

Improve this answer

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.