“The expectations of life depend upon diligence; the mechanic that would perfect his work must first sharpen his tools.”
At the Center for Connected Learning and Computer-Based Modeling, I have been an active participant of the software development team, which is responsible for the continued development of the NetLogo modeling environment and programming language and the HubNet participatory simulation software. My personal contributions range from NetLogo language features (improved network layout primitives) to core engine work (developing a hybrid bytecode compiler/interpreter) to usability features (auto-save, auto-indention, quick-help). I have also contributed through participation in numerous planning meetings, QA testing, and constructive feedback.
I highly recommend NetLogo to anyone interested in studying agent-based modeling or complex systems. (And also to kids interested in learning computer programming.)
BehaviorSearchOne central outcome from my thesis research is the creation of a software tool for automating the exploration/search of agent-based model parameter searches. My goal in designing BehaviorSearch was to make a tool that was easy enough to use that it would make query-based model analysis accessible to a wide range of modeling practitioners. BehaviorSearch is open-source cross-platform software that interfaces with NetLogo, provides an easy way to design and run search experiments. BehaviorSearch is flexible about searching a combination of numerical (discrete/continuous), boolean, and categorical parameters, and it offers several search algorithms and search space representations to suit different needs. BehaviorSearch provides a GUI version, as well as a headless command-line version suitable for use on computing clusters, and has multi-threaded support for searching in parallel on multi-processor/multi-core computers.
For more, check out the BehaviorSearch website, which includes a tutorial, downloads, and more.