Before openly sharing your data, you need to find out if you actually have the right to share it. To do so, you will need to research the data sharing policies of anyone who may have a claim to your data:
- Your institution (start with your institutional Intellectual Property policy)
- Your collaborators’ and their institutions
- Your funders
- Your institutional review board (if you have human subjects data)
- Rights holders of anybody else’s data you used in the development of your data set And that's just to name a few.
You can see where this can easily get confusing and convoluted. We are not lawyers, so if our advice is contrary to that of your institution’s legal counsel, please defer to them. We’ll try to cover the basics with some guidance on how to get to specifics.
1. COPYRIGHT - In GENERAL, in many countries, facts cannot be copyrighted.
This rule usually applies to raw observational and experimental data, without any originality, skill or judgement applied. It gets more complicated when data consist of creative expressions (e.g. images, narratives, music which all can be copyrighted) or when you are combining data created in different countries with different copyright laws. (See Sharing Research Data and Intellectual Property Law: A Primer for more information.)
2. INTELLECTUAL PROPERTY (IP) - Intellectual property rights USUALLY rest with the funder of the research. Read your organization’s intellectual property policy (check with your Office of Research or librarian if you don’t know where to find this) to see who maintains intellectual property rights of data produced by employees and students. (See What Is Intellectual Property document (pdf) put out by the World Intellectual Property Organization for more information on intellectual property.)
3. FUNDER MANDATES - Funder mandates USUALLY trump organizational policies. In the past decade, more and more funders (private and public) have begun requiring statements or plans for sharing data from funded research. Since funder grants are usually contracts between the funder and an organization, generally they take precedence over any organizational policies but you will want to check with your Office of Research or librarian to make sure. Read your funder grant proposal guidelines to see what your funder requires for data sharing.
4. PRIVACY / ETHICAL CONSIDERATIONS - In GENERAL, these will trump all of the above.
- Institutional Review Board (IRB) - if your research is regulated by your institution's IRB, you may have very specific rules about what you can and can’t share- or, indeed, what you MUST share and how. Consult with your IRB office for more details. For advanced users interested in talking to your IRB about changing approaches to data sharing, consult this Sharing Guidebook produced by one of our Mozilla Science Fellows, Jason Bobe.
- Disciplinary Norms - Each discipline differs slightly in their culture of data sharing- with what is shared and where it is shared, varying even within sub-discipline. It really depends on the data being produced. For example human subject data should always be de-identified before sharing, data that could be used in exploitative or dangerous ways should be shared with care, etc.
All of these issues can make it very unclear to someone else whether or not they can use your data in their research. The best way to make this clear, is to assign a license to your data.