Take the 2-minute tour ×
Academia Stack Exchange is a question and answer site for academics and those enrolled in higher education. It's 100% free, no registration required.

I would like to post an repository on github that includes data and analysis code in R. The analysis and data forms part of a journal article submission. I'm happy for people to view the code and data prior to publication (in particular, it might be an easy way for reviewers to examine the code). However, I do not want anyone publishing analyses of the data prior to acceptance and publication of the journal article. After publication, I want to encourage people to re-analyse, re-use, re-publish (e.g., with a GPL licence with an encouragement to attribute).

Obviously, I could just keep the data and analysis code a secret until after publication, but I thought an appropriately worded licence might be more appropriate. I thought about just writing in plain English that the work is copyrighted at this time, and will be converted to GPL at a later date following publication.

  • Is there a standard way of licensing data and code so that people cannot republish the data and code until the corresponding publication has been published?
  • Or is it better just to keep the data and code secret until after publication?

UPDATE: I suppose there is a legal perspective to this, but I know that academia has its own norms and conventions regarding attribution and respecting the wishes of authors. So I'd be particularly interested in answers framed in that context. I.e., My broader aim is to be the first to publish my own research, generally get attribution, but also allow others to build on that work. So I'd be interested also in what is considered good practice when you are in the situation of wanting to share data and code while an article is being peer reviewed, but not wanting to lose your right of first publication.

share|improve this question
3  
This doesn't directly answer the question, but one possible workaround (depending on your perspective) would be to publish a preprint when you release the code, thus establishing your priority and giving people something to cite if they use your code. –  David J. Harris Jan 30 at 7:24
2  
Also, it's not totally clear to me that the legal system is the best way to deal with the issues you bring up. Asking people not to scoop you might be just as effective as legal threats. Especially since it's unlikely you'd ever actually sue anyone for infringement. That said, I think it's an interesting question, so (+1) –  David J. Harris Jan 30 at 7:26
3  
As a side note, your code is always your copyright (unless it is not because of a hiring contract), regardless of it being released under any GPL license or not. –  dgraziotin Jan 30 at 8:47

6 Answers 6

Honestly, I doubt that copyright protection is of any use. The copyright on the code itself does not* extend to data produced using the code. Similarly, the copyright you would hold on the paper submission covers* that particular presentation of the data, but not the data itself. So as long as someone is physically able to view your data, I believe there is no legal avenue to prohibit them from doing their own analysis on their data.

However, for another researcher to take the output of your code and perform their own analysis in an attempt to scoop you, when they know you have your own paper doing the same analysis pending, is ethically very questionable. If it came to light that something like this happened, I think the academic community would strongly frown on it. That's a very strong incentive for any other researcher not to do this, and so I personally wouldn't worry about it.

Besides, if it does happen, the other researcher will have to cite your code anyway so you still get credit. This can actually be a good thing. I would recommend including in your code a notice of the form

Please cite the following reference if you use the results of this code in a publication:

[reference to paper or code]

(see this example from my own publication history) If you're worried about someone else abandoning all pretense of ethics and just using your results without citing them, rest assured that it is very difficult to pull that off, and it constitutes academic fraud, which is a career-ender if it's discovered.

And finally, as a practical matter, you have a very large head start on anyone else who might want to publish an analysis of the outcome of your code. Don't underestimate the time and effort it takes for someone else to go through your code in enough detail to learn what it does and figure out how to use it enough to generate original results, and then to write and submit a paper and have it reviewed, typeset, and published.

*informed layperson speculation; see a lawyer for a definitive statement

share|improve this answer

A1 : The legal department at your university or company should be able to help you out with this type of request.

A2 : This depends on the journal to which you are submitting. Read their rules on copyright of submitted software and publication beforehand.

Other comments / ideas :

I have software freely available that I have submitted to a journal before publication (although, this may not have been the best idea) and software that is also ready and that I use that I could make freely available, but have not as of yet.

An alternative would be to say that interested parties may email you for access to the code before publication happens.

You could always publish the data (or paper explaining the data) on a pre-print server so that you cannot be scooped and then also publish the corresponding code. This will outline clear dates as to who published the research first.

share|improve this answer
    
I'd like to point out that some universities might not have a legal department (or might not care about these matters if they do). –  Trylks Jan 30 at 11:01
    
@Trylks -- Very true! This is completely dependent on your institution. –  nagniemerg Jan 31 at 17:57

You write that you

do not want anyone publishing analyses of the data prior to acceptance and publication of the journal article.

I am not a lawyer, but I don't think there's any license to enforce that. Once you publish data, other people are free to work with that data, and publish their analyses (although they might not be able to republish the original data, at least not in the same format as you did). As Wikipedia writes, copyright does not cover information itself, it just protects the way that it is presented, or the verbatim description. If you want to prevent others from publishing analyses of your data, I think the safest way is to keep the data confidential until you publish your paper.

On the other hand, if you don't put a license on the data, you have whatever copyright protects there, so others will not be allowed to re-publish the data in the same format. Therefore, I think the risk that others publish an ernest journal based on your data is not that high, unless it's really spectacular data.

The code is a different issue of course. You hold copyright on that, and if you don't put a license on it, others won't be allowed to republish it.

share|improve this answer

To address your revised question:

My broader aim is to be the first to publish my own research, generally get attribution, but also allow others to build on that work. So I'd be interested also in what is considered good practice when you are in the situation of wanting to share data and code while an article is being peer reviewed, but not wanting to lose your right of first publication.

The usual practice in my field (mathematics) is that when the project is finished, you submit it to a journal, and simultaneously post it to arXiv, a public preprint server. This establishes your priority in two ways, and lets people start building on it immediately.

share|improve this answer

Your right as a owner of intellectual property, allows you to release and allow reuse of that "property" in any way you feel correct. The use by a third party of your "property" is not allowed unless the party has a way to demostrate that you have somehow allowed it. The way you allow it is up to you. Obviously that creates an enormous exercise for the judge, that could be called to decide on the fact. Essentially there is a big legal hole, because I could decide for example to sell a book and to allow the purchaser to read it ONLY ON NIGHT from 8pm to 11:59pm. That means that if you use the book in other times you are penaly responsible.

So that is the reason that the legislator should take some decisions because if not a lot of problems may arise. The limitions that the author may impose, may be illegal etc. Most important is that fact that licenses may be so long and articulated that no normal person may be able to understand them and so that person shouldn't be liable for the infringement of them.

Any way my opinion is that you may impose any rule you may like. That is a one paty action and the other party should demonstrate that she is legaly using your work.

share|improve this answer

Who prevents you from releasing the code under GPL only after the article has been accepted for publication?

In such case, probably you cannot add GPL header to the files in advance but unless it is very large project, this should not be a really big problem (you may write some script to add headers if it is really a lot of files).

If you setup a public repository on GitHub, by doing so you allow to view and fork it. If this seems not acceptable for you, publish on the university server instead and only publish to GitHub when you open source it. You can add any restrictions you want as long as it is your fully owned code. Just I am not sure if it will be easy to enforce these restrictions if somebody does violate.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.