Software Attribution: can we improve the reusability and sustainability of scientific software

The traditional model of scientific credit is based on publication from a time when reproduction was expensive and cross-referencing costly. Recently, the open data movement has enabled other objects created in the course of research, such as software, to be considered as important research outputs. Technological infrastructure has reduced the traditional costs incurred in attribution and citation, and spawned entire fields of bibliometrics and alt-metrics. In my talk I will look at some of the continuing challenges of extendingattribution, citation and credit to software, and the implications for software reusability andsoftware sustainability. These include the impact on skills and career paths, as well assoftware quality and fragmentation.