Software sustainability – together we’re stronger

By Neil Chue Hong, OMII-UK.

Back to index page

Previous article: News in Brief

Next article: One giant leap for Taverna

Research is about exploration, innovation, the discovery of the new. If software is to remain relevant, it must it must be able to adapt to the changing needs of the researcher, which means that it is impossible to divorce the software from the research environment. This tells us that the best people to sustain research software for the long term are the researchers themselves. Any third-party organisation aiming to take on and maintain a key piece of software without involving the software’s research group is setting itself up for failure.

Sustainability is more than just making something available in a repository: preserving something is not the same as sustaining it. Software environments change – witness the many system updates we see each week – and we need to ensure that the research software moves too. Likewise, it is not sustainable to have only a single developer responsible for the software. At worst, this fails the hit-by-a-bus test. At best, it is difficult for one person to discover, understand and meet the differing requirements placed on software by different domains.

A common approach is to open source the software in an attempt to get more developers involved. In our experience, and the experience of others, this alone is not enough. It certainly isn’t the route to free effort that some hope will lead to sustainability.

The approach that OMII-UK has developed over the last few years is a collaborative one. We work with key members of the research communities to reduce fragility, enhance comprehensibility, improve extensibility and increase the software owners’ confidence to make the changes that their research demands.

We work with research groups to help enhance the sustainability potential of their key software products. We have developed an approach that focuses on both the software and the community in which the software is used. Software improvement sees us enhance the software’s structure, modularity, commentary, documentation and test processes. This makes the software easier to pass from generation to generation of researchers. With community improvement we focus on enhancing communication and work to bring key people into the decision-making processes that drive software development. This grows the community around a software project and helps to remove the barriers that prevent more people from contributing to it. Programmes like Engage have shown that, with the right approach, much can be done in a short amount of time and with a minimum of resources – other articles in this newsletter (see pages 4 and 8) are a testament to this fact.

The sustainability of important research software requires partnership between research and software specialists. Likewise, sustainability is not free. By working together and pooling our effort we are stronger, and our software is more sustainable.