Our journey towards open research support software
Ian Gifford - Head of Digital Development, University of Manchester Library Nilani Ganeshwaran - Senior Software Developer – Management, University of Manchester Library Tom Higgins - Senior Software Developer – Technical, University of Manchester Library
Since Linus Torvald announced that he was writing a free operating system as a hobby in 1991, Open Source Software has grown from a novel movement to a world-wide phenomenon which has transformed the software development industry and become (as pointed out by the Boston Consultancy Network last year[1]), “a strategic necessity in a fast-changing digital world”.
It’s taken a while for the benefits of Open Source Software to be realised within the academic community over and against a preference for expensive commercial software[2] and the open research movement is playing a critical role in its growing acceptance and effectiveness. But what about the systems and applications that provide administrative support to researchers and research support functions within higher education?
As an embedded software development team within one of UK’s largest research libraries, we have been developing innovative research support and administration software for over ten years. This includes applications to manage the University’s electronic PhD submission process, to manage the submission and compliance of open access research outputs, to provide access to advanced research data metrics and a large number of specialist scripts for interacting with large research platforms such as SciVal, Scopus and Altmetrics to name a few.
As developers we have long been aware of the benefits of Open Source Software but have often tended to do our development (as it were) behind closed doors. It’s interesting to reflect upon why this is, how much of it is habitual, how much because our user base and their requirements are locale-dependent in terms of business processes and perceived to be not relevant outside of the institution, how much because we are gifted with a talented development team who are capable of creating many a bespoke application without the need to collaborate externally.
Developing systems
There is a certain inertia with developing in-house systems which can impede good intentions when it comes to opening up and sharing code, this can include the necessity of integration with internal systems and the amount of time and effort needed to potentially modify software to be easily adaptable to and usable by the outside world.
However, over the last four years we have been involved in two major Open Source Software collaborations with Harvard University (ePADD+) and Cambridge University (CDCP). Both have helped to clarify for us in an immensely practical way the rich benefits of collaborative development and which, along with the growing open research movement, have helped to overcome much of our reticence in opening up access to our locally developed software.
We began making tentative steps in this direction through the open sourcing of some code we wrote to retrieve datasets from the Scholix API based on a light of target publication DOIs [3] which was shared at the OpenAire conference in 2020 [4].
As a further step we have recently open sourced the latest version of our Open Access Compliance Platform[5] an extensible tool for managing and monitoring the submission and compliance of open access research outputs, which has been recently modified to reflect the recent UKRI open access policy change on peer-reviewed articles[6].
We have also made a commitment to begin each new software project involving research support as open source from its inception which will help us avoid making excuses later on about the further work involved in sharing it.
Risks
One of the risks commonly associated with Open Source Software is around generating and sustaining a community of interest but our recent experience collaborating with other large Universities has shown us first-hand that despite our difference we all hold many challenges and problems in common and that working together, sharing our experience, and combing our resources, can dramatically increase the chances of us successfully negotiating many of those challenges and problems.
Increasing pressures on professional services staff to deal with constantly changing technical, structural, and political environments at work often present a temptation in the opposite direction, with an understandable but myopic focus on pressing local issues and a corresponding demand for immediate, local solutions. Whereas multi-institutional collaboration requires playing a longer game, taking a longer-term view, and requiring a sustained commitment and an openness and willingness to adapt and learn from others.
We have learnt that good intentions are often not enough and so we have chosen to take a more deliberately pro-active approach in seeking potential collaborators to work with us on future research support software projects which, aside from the benefits to the communities involved, we hope will contribute in a productive way towards fulfilling the goals the open research movement.
If you would like to work with us on such projects, please don’t hesitate to get in touch.