Open access: Creative Commons and data licensing

Creative Commons and Open Data Licenses

While a Data Access Committee, Data Access Terms, and Data Use Agreement Template are best for medium and high-risk data, you can consider an open data license for low-risk data which allows you to define what others can and cannot do with your data. This most often is a Creative Commons license.

Attempting to obtain permission from copyright holders can be a frustrating process. You must track them down and obtain permission. And if you are a copyright holder, repeated requests to use your work can be an irritation. Potential users can try relying on fair dealing exceptions, but this can be a risk. To circumvent this barrier of attempting to gain permission, and removing the ambiguity around fair dealing, owners may apply a license or waiver to their copyrighted works.

A license is a way for rights holders to give permission to use a work, or part of a work, subject to conditions, in a way that would otherwise infringe copyright for that work.

Creative Commons (CC) are the most common form of license you encounter. This guide, for example, uses a CC license.

CC licenses are template licenses, useful for all kinds of creative works—including, in some cases, research data. If you are the rights holder for a work, all you do to apply a CC license to your work is state which form of CC license it has and provide a link to that license.

Aside from the ease of adoption, the attraction and advantage of CC licenses is the way they communicate permissions and restrictions on reuse. This is thanks to the four conditions that can be selected to form the license.

1. BY: Attribution

The primary condition is “Attribution”. This is the “BY” condition. BY means that anyone who reuses the work must give credit to the owner of that work. BY is found on almost all forms of CC licenses.

The potential problem with a BY requirement is “attribution stacking.” So, if you use different works, you must acknowledge all contributors to each work from which it is derived and indicate the types of changes you have made.

This usually is not a problem, but if you are using a lot of different works, or works with many contributors, the list of names to acknowledge and modifications to state stack up to the point where they might become unwieldy.

2. NC: Non-Commercial

NC means for non-commercial reuse only. What is non-commercial? CC defines NC as “not primarily intended for or directed towards commercial advantage or monetary compensation.”^[1]

Note, the NC clause refers to use, not user. If the use is not directed or intended to produce a commercial advantage, it does not matter who the user represents. Likewise, an example of an academic who uses NC licensed data to create an app that is then sold and generates a profit is violating the NC clause.

If there is a commercial advantage to the work, it’s best to look outside CC licenses or even to stick with “all rights reserved.” Unless you can provide strict enforcement and oversight of non-commercial use, NC conditions can be problematic, as it is unlikely a commercial user is willing to pay to use an NC licensed work.

3. SA: Share Alike

SA is Share Alike. Anything that uses SA licensed work must be shared under the same conditions. So, you can’t, for example, use CC-BY SA licensed work and then place additional restrictions like adding an NC condition.

The problem with SA conditions, sometimes termed “copyleft,” is they prevent works being combined with those released under more restrictive licenses, especially for forms of work that encourage reuse, repurposing, derivatives. A CC BY-SA work—attribution only and share alike—could not be combined with a CC BY-NC work. Neither the SA nor NC condition would be satisfied—once the SA label is applied, everyone who uses your items in the future has to keep the exact same license.

4. ND: No Derivatives

Finally, there is ND: No Derivatives. This means that if you alter, transform, build on, or remix the content, you cannot share the materials created. Essentially, a look but don’t touch condition. Potentially useful for fixed works, but again, if wanting to prevent derivatives is an intention, then stick with all rights reserved.

The main goal of a CC license is to make the content as widely available as possible. CC licenses are a liberal form of sharing, even with NC and ND restrictions. A dataset with sensitive information, or any work that requires some form of restriction beyond non-commercial reuse or enforcement of restrictions, is not suitable for CC licensing.

Remember CC licenses are not revokable. Once applied, they remain applied for as long as copyright exists.

CC0: Public Domain Contribution

An alternative to licensing is to employ a waiver. A waiver sees the rights holder give them away and put a work into public domain. Creative Commons Zero (CC0) is such a waiver.

The advantage of employing a waiver is it removes ambiguity on either side as to who can use something or how. There is no need to worry about attribution stacking or any other obligations or restrictions on the part of the user, and no one else can subsequently assert or reassert ownership once a work is public domain.

But waiving copyright does involve a loss of control on the part of the copyright holder(s), as you cannot require attribution or specify any other terms of use. Nor can you reclaim rights once you have waived them. It should also be obvious that public domain dedications are inappropriate for personal or sensitive information.

Data Access Terms and Data Use Agreement Template

If data are not appropriate for releasing under a CC license, it may be shared under a Data Access Terms with users signing a Data Use Agreement. These are terms tailored to the needs of types of data in a particular repository. They often require action from the potential user, such as registration and acknowledgement of usage conditions before the data can be accessed. Such data usually prevent sharing with anyone other than the individual(s) applying to use the data, prohibit attempts at re-identification of participants, as well as the NC condition found in some CC licenses. Sometimes terms are more restrictive if data are sensitive. These might include barriers like requiring certain qualifications, providing details on a specific project for use, and specific access requirements which can include technical requirements and sometimes physical requirements.

Templates for Data Access Terms and Data Use Agreements are included earlier in this toolkit!