Editorial: Remembrances of things past

The title of this editorial is borrowed from Marcel Proust’s autobiographical novel to describe a new category of papers, which we call “origins papers”, and which will appear in TOPLAS from time to time. An origins paper is a technical contribution that describes the genesis and development of an important idea in programming languages and systems. Athena may have sprung fully-formed from the brow of Zeus, but ideas in science often require several iterations and a lot of sweat and tears from many researchers before reaching a finished text-book form. An origins paper describes the historical milieu out of which an idea arose, credits the contributions of various people to the finished product, and may document the dead-ends that were explored by the research community before the big idea emerged from the mists. In older disciplines like Physics and Mathematics, entire books are written about the origins of important ideas like relativity theory, quantum mechanics, and calculus. There are not many origins papers and books in Computer Science, perhaps because our field is relatively new, but we believe that the time is ripe for our community to start documenting the origins of important ideas in our discipline. We see many benefits from origin papers. Knowledge of the history and development of an idea adds to the depth of one’s understanding, and may enrich and animate the presentation of that idea in the classroom. Important ideas are often discovered by people standing on the shoulders of others, so origins papers are an opportunity to acknowledge sung and unsung contributors, giving credit where it is due. Origins papers may also show young researchers how science is actually done, opening their eyes to the fact that science is an intensely collaborative activity, since the stereotypical solitary scientist thinking deep thoughts in a windowless cubicle is the exception rather than the norm. In this spirit, we have solicited a number of origins papers from the community on topics such as the origins of dependence analysis and loop transformations, multithreading and latency tolerance, and static single assignment form. Like other papers submitted to TOPLAS, origins papers are refereed carefully and contain innovative content of broad interest to the field. The TOPLAS website describes the criteria in more detail: http: //www.cs.utexas.edu/∼toplas/. In this issue, we are delighted to present the first of such papers, written by Davide Sangiorgi, on the origins of bisimulation, which is an important idea in process algebras.