I'm a linguist/psychologist/evolutionary/cognitive science type working in academia. The work I do almost inevitably uses natural data like linguistic corpora, agent-based simulations, or experiments where data is collected with human participants.
Traditional software for behavioural experiments, like E-Prime and its descendants (e.g., PsyScope, PsycoPy), are very much built around the idea of psychological experiments as a series of trials. I found this type of software didn't work for me because it focused on precision timing (important for many types of experiments, but not mine), a very traditional stimulus-response model, and a super-boring visual environment for the participant. But experimental results tend to be interesting, so something had to be done. I had to learn to program.
I consider my experiments to be more like games, and I want my participants to be players. Although many of my experiments are in fact rather traditional, my goal is for them to become more and more game-like as I learn to be a better programmer, designer, and scientist. To get more control over my experiments I started off by learning Processing. I primarily do data analysis in Python or R and use ggplot for static visualisations, but for more fun/dynamic visualisations I am learning d3.js. This site is for displaying some of my ongoing experiments and visualisations.I'm happy to share code where I can, but be warned: I am an entirely self-taught programmer, and I started fairly late in life, so my code is far from perfect (although it usually works). If you have feedback about the interface of my experiments - if it is confusing or annoying - let me know and I'll try to implement it (time permitting). Any specific comments or suggestions about my code including dodgy math, n00b technique, or use of namespaces can be directed here.