I've been in the industry for too long (though not in low-latency) - I'm doing this to break out of the industry! And I think the finance game is long overdue for a good disruptive technologic event (not implying my humble project is it). R is a personal choice for data analytics (I love ggplot2). But the haskell thing is more than that. It's been done before (http://www.starling-software.com/misc/icfp-2009-cjs.pdf) and even goldman's has a thriving erlang/OCaml hacker ethic (http://www.zerohedge.com/article/aleynikov-code-dump-uncover...).