How we locate wild animals with a functional program

Ryan Newton
MIT

This talk will describe how we use the ML-like stream processing
language, Wavescript, to implement distributed sensor network
applications.  Stream processing is a good fit for distributed problems,
as it exposes parallelism in the form of explicit producer/consumer
tasks.  This talk will focus on the particular benefits of functional
programming in the stream processing context: from building generic,
modular libraries, to capturing streaming design patterns and embedding
"mini langugaes" that capture different streaming models.

This talk will require minimal familiarity with functional programming.
It will be of particular interest to those working with real-time,
embedded, and data intensive programs.  It will argue the utility of
Wavescript in particular to a variety of real-world problems, and argue
more generally that functional programming enabled useful techniques.
In particular, we focus on the tricks that functional programs can play
that are difficult in other streaming languages that extend C.