What open source research information systems are there?

A piece of infrastructure often goes unnoticed – a Current Research Information System (CRIS) is such a piece of infrastructure in research organizations. In this blog, I explore what a CRIS is, what (open source) software provides this functionality, and what we can do to improve the situation.

What is a research information system?

A research information system helps organize information related to research outputs and the producers of it. Most academic organizations have such a system to keep track and report of who is working at the organization, what they produced, what additional interests they may have, and much more. It is also the foundation for internal tracking mechanisms that enable mechanisms for academic competition (both internally and externally). The image below depicts other pieces of information that a research information system is typically expected to track.

If you are at a research organization, you probably have used a CRIS at some point. You may know it through the yearly experience of manually entering that year’s outputs. You may also know it as the way you create or view public profiles, available on the organization's website. 

The CRIS is deeply embedded in a modern day academic organization – there is no escaping the CRIS as an employee at an institution. It is the basis of a lot of frustrations in my anecdotal experience, and yet it is taken as a given instead of something to change. It made me wonder: What alternatives are there, really?

What characterizes the CRIS landscape?

To understand what software implementations are offered as research information systems, I explored the landscape of available research information systems. Surprisingly, the available information is fairly sparse and hard to find. What I did find, is already quite outdated: The most recent reports date back to 2016 and 2018. The table below lists the most common software implementations providing CRIS functionality.

Some highlights from these surveys:

• The majority of institutions use the same software for CRIS and repository functionality (82%)
• A substantial part of institutions have in-house built solutions (21-28%)
• Pure (by Elsevier) is the most used CRIS (30-43%)
• Most CRIS are hosted on-premises (>50%)

Whether these numbers changed over the past 6-8 years is an open question. There is reason to speculate that since then, there may have been a strong push to move towards more cloud based solutions and accompanied data collection benefits for the centralized provider. The surveys already indicate there are geographical differences between Europe (more self-hosted) and North-America (more cloud based). A renewed investigation to compare whether a shift to cloud based CRISs occurred would be helpful to assess the situation.

Overall, these reports provide an indication that there are nine key research information systems, of which three are open source (33%) and five also provide repository functionality. But there is only one open source CRIS that also provides repository functionality: DSpace-CRIS.

82% of institutions use the same software for their research information systems and repository. Assuming they want to keep it that way, they could only consider software that also provides both the CRIS functionality and the repository functionality, meaning that for 82% there is only one open source option available.

What next for open source CRIS repositories?

As with any choice: One option is no choice. There are multiple open source research information systems available (e.g., VIVO, Profiles). There are multiple open source repositories available (e.g., HAL, Invenio). But there is only one open source CRIS repository available (DSpace-CRIS).

There already is demand for the things open source enables – digital sovereignty and source audits to enable privacy and reduce vendor lock-in. If we want these things for CRIS repositories, which are so deeply embedded in research organizations and their competitive nature, then we should start making more open source options available. This is perfectly feasible and desirable, especially when we consider the recent Barcelona Declaration on Research Information.

All research information systems are based on an open standard: the Common European Research Information Format (CERIF). This means that, in theory, migrating from one system (e.g., Pure by Elsevier) to another (e.g., DSpace-CRIS) is more than just possible – it can be built in. However, these migrations currently only happen once in a blue moon, because the opportunity costs are high and the system is so important to the functioning of a competitive university. Moreover, opportunity costs are kept high through vendors. This prevents organizations from wanting to undertake such an effort, even though they can recoup those costs with reduced licensing fees when switching to open source software.

The limited open source CRIS repositories is an invitation to innovate. Open source options should not only provide additional value as a CRIS repository itself, but it also needs to excel in its migration capabilities. The underlying open standard CERIF means this is technically possible. I would even speculate and say that an open source version of Pure (by Elsevier) would quickly take a large part of the market share, if it could provide a reliable migration assistant with it. To build something like this will require a commitment from universities and other research organizations to enable and invest in such an effort.

In a future blog post, I will expand on how creating investment funds for open source infrastructures will help universities and research organizations save money by moving away from a cost-covering model.

Update: We received two more CRIS through social media upon posting this blog - OSIRIS and gris.