In recognition of the contributions to computer science that John Vlissides made during his lifetime, SIGPLAN has established the John Vlissides Award. The award is presented annually to a doctoral student participating in the OOPSLA Doctoral Symposium showing significant promise in applied software research. All doctoral candidates participating in the annual OOPSLA Doctoral Symposium are eligible. The award includes a prize of $2,000.
The John Vlissides Award was established by generous contributions from the following companies, individuals, and organizations: IBM, Addison-Wesley, William Pugh, and SIGPLAN.
The selection committee consists of the following members:
The committee is chaired by the OOPSLA Program Chair. The OOPSLA Program Chair shall adjudicate conflicts of interest, appointing substitutes to the committee as necessary.
The winner of the Vlissides award is Gustavo Soares of the Federal University of Campina Grande, Brazil. His proposed thesis is entitled “Automated Behavioral Testing of Refactoring Engines”
Hari Pyla’s work is motivated by the need to leverage concurrency to achieve performance, and the challenge of avoiding deadlocks in concurrent shared memory programs. He has introduced Sammati, a new approach that transparently and deterministically detects and recovers from deadlocks, without requiring modification to source code. This approach not only avoids errors, but also allows more effective composition of lock-based code. Hari’s system also supports coarse-grained speculative parallelism, supporting difficult-to-parallelize applications and achieving significant speedups in practice.
Márcio Ribeiro’s research addresses the problem of maintaining software defined using “virtual separation of concerns.” The virtual separation of concerns technique presents software with features defined using code between #ifdef and #endif (as in the C preprocessor) to the programmer in a way that hides various features that are not of immediate interest. However, when maintaining such software, it is easy to break code in hidden features. The solution approach uses a specialized dataflow analysis that takes features and allowed feature combinations into account. The results of this analysis are used to warn the programmer if there are dependencies between the changed code and any hidden features. These ideas are being implemented in a tool that will allow independent development of parallel features.
Tudor Dumitras’s research identifies and addresses the leading causes of both unplanned failures and planned downtime resulting from software upgrades in distributed systems. Traditional fault-tolerance approaches concentrate almost entirely on responding to, avoiding, or tolerating unexpected faults or security violations. However, scheduled events such as software upgrades account for most system unavailability and often introduce data loss or latent errors. Tudor defines three abstract properties required for improving the dependability of software upgrades. Building on empirically derived insights on current upgrade practices and problems, Tudor’s work harnesses the opportunities provided by emerging technologies such as cloud computing to simplify large-scale upgrades, to allow upgrades to be executed efficiently online, and to improve their dependability.
Ciera Jaspan’s research targets the programming difficulties associated with using software frameworks. Plug-in writers need to satisfy a variety of framework constraints, and Ciera is directly motivated by the kinds of plug-in errors that frequently arise in today’s frameworks. She has defined a new abstraction for expressing framework constraints as well as lightweight annotations and an associated static analysis to enforce this abstraction. Ciera is focused on making the right tradeoffs to ensure her solution is useful in practice.