Open Research Principles

This week we will focus on project planning and some related issues. This includes data management, ethical matters, and copyright issues. These topics have become more important with the general transition to Open Research.

In the following, we will use the term Open Research instead of Open Science. That is because we find “research” to be more inclusive than “science”. Many scientists identify as researchers, but not all researchers work in science. This is the case among researchers working in the arts and humanities, particularly those in the performing arts.

Why is Open Research Important?

We perform research because we want to learn more about the world. This is not possible if we do not share the knowledge we generate. Sharing has been a core research principle for centuries. After all, research builds on research. Publishing articles and books are ways of sharing the knowledge coming out of the research. Ideally, these publications should also be published so that they are widely accessible by others.

When working empirically, as we do when collecting motion capture data, the data is in itself an important part of the research output. The data supports the publications, and the publications explain the data and how it was collected. That is why there is an increasing focus on sharing data. If the data is openly available, others can check that the analysis and interpretations in publications are sound. Others can also use the data in new studies.

In addition to sharing data, it may also be necessary to share information about the methods used and the source code to scripts used in the data collection or analysis. This is, again, to allow for checking the reliability of the analysis. Others may also use the methods and scripts.

The Open Research ideal is that everything should be “as open as possible, as closed as necessary”. That also includes sharing educational material, like we are doing in this course.

Open versus Free

When we talk about Open Research, particularly Open Data, it is important to keep in mind that “open” is not the same as “free”. The two concepts overlap, but you may have free data that is not open and open data that is not free.

The general recommendation is that data should be both open and freely available for reuse and redistribution. Then it is necessary to equip the data with a permissive license, such as the Creative Commons licenses (CC). We will talk more about licenses in a later step.

One of the most common objections to opening the research process is that other people will steal your ideas, data, code, and so on. However, if everything is tagged correctly, time-stamped, and given unique IDs, it is impossible to steal anything. Everything will be traceable. And plagiarism algorithms will quickly sort out any problems.

The building blocks of Open Research?

The figure below summarizes some of the building blocks in an Open Research paradigm:

As can be seen, all parts of a research chain could be open. Some parts are easier to do than others. For example, there has been a growing focus on making all publications openly available (Open Access). There is now also a push for making data openly available (Open Data). This is more challenging, mainly because of privacy and copyright issues. We will talk more about both of these later.

In addition to making typical research outputs available, it is also possible to think about opening the research process. For example, it is possible to use Open Notebooks to explain how the research develops and track the steps taken to conduct the research. This could be done on web pages or blogs, on specific platforms (such as Open Science Framework) or with more computational tools (like Jupyter Notebooks).

Another way of opening the research process could be better documentation of the methods used (Open Methods). This could be in textual form or as part of making open-source software available (on platforms like GitHub and Gitlab).

Summing up

As the quick run-through of the different parts of the building blocks has shown, it is possible to open the entire research process. Much experimentation is happening these days. For example, the sharing of source code and data has come a long way in some communities. Some journals even refuse manuscripts without complete data sets and source code.

The main argument for opening all parts of the process is “sharpening” the research process. You cannot be sloppy if you know that it will be exposed. Open research fosters quality research.

In the rest of this week’s steps, we will look at various ways to open the research process.