Refactoring aims to improve the quality of a software systems' structure, which tends to degrade as the system evolves. While manually determining useful refactorings can be challenging, search based techniques can automatically discover useful refactorings. Current search based refactoring approaches require metrics to be combined in a complex fashion, and produce a single sequence of refactorings. In this paper we show how Pareto optimality can improve search based refactoring, making the combination of metrics easier, and aiding the presentation of multiple sequences of optimal refactorings to users.

Virtual Machines (VMs) with Just-In-Time (JIT) compilers are traditionally thought to execute programs in two phases: the initial warmup phase determines which parts of a program would most benefit from dynamic compilation, before JIT compiling those parts into machine code; subsequently the program is said to be at a steady state of peak performance. Measurement methodologies almost always discard data collected during the warmup phase such that reported measurements focus entirely on peak performance. We introduce a fully automated statistical approach, based on changepoint analysis, which allows us to determine if a program has reached a steady state and, if so, whether that represents peak performance or not. Using this, we show that even when run in the most controlled of circumstances, small, deterministic, widely studied microbenchmarks often fail to reach a steady state of peak performance on a variety of common VMs. Repeating our experiment on 3 different machines, we found that at most 43.5% of ⟨VM, benchmark⟩ pairs consistently reach a steady state of peak performance.

Domain specific languages (DSLs) are mini-languages which are increasingly seen as being a valuable tool for software developers and non-developers alike. DSLs must currently be created in an ad-hoc fashion, often leading to high development costs and implementations of variable quality. In this paper I show how expressive DSLs can be hygienically embedded in the Converge programming language using its compile-time meta-programming facility, the concept of DSL blocks, and specialised error reporting techniques. By making use of pre-existing facilities, and following a simple methodology, DSL implementation costs can be significantly reduced whilst leading to higher quality DSL implementations.

Abstract. Though there has been nearly three decades of work on program slicing, there has been comparatively little work on slicing for state machines. One of the primary challenges that currently presents a barrier to wider application of state machine slicing is the problem of determining control dependence. We survey existing related definitions, introducing a new definition that subsumes one and extends another. We illustrate that by using this new definition our slices respect Weiser slicing's termination behaviour. We prove results that clarify the relationships between our definition and older ones, following this up with examples to motivate the need for these differences.

Model transformations are increasingly recognised as being of significant importance to many areas of software development and integration. Recent attention on model transformations has particularly focused on the OMG's Queries / Views / Transformations (QVT) Request for Proposals (RFP). In this paper I motivate the need for dedicated approaches to model transformations, particularly for the data involved in tool integration, outline the challenges involved, and then present a number of technologies and techniques which allow the construction of flexible, powerful and practical model transformations.