Any new programming language or environment comes with its own set of classes to model collections. History shows that having a standard way to manipulate a group of objects is a nonnegligible strength. However, having a robust collection library is not an easy task; bugs may be found, even after decades of intensive use.
One way to improve the quality of a collection class library is to structure its unit tests and extend their coverage, which requires discipline. Wolfmaier et al. argue that identifying specifics of the collection library and using code quality metrics are helpful when writing new tests. In short, defects are easily identified.
Clearly identifying the different concerns with collections is challenging, and the authors do a good job. Even though the paper provides only preliminary results, people interested in collection libraries and testing in general will find it relevant.