There has been some buzz around a new language for numerical, and potentially statistical, computation: Julia. Another interesting project is gsl-shell which offers interactive data analysis and visualization with GNU Scientific Library.

Julia

Julia is actually available on GitHub and a binary for OS X can be grabbed from there. That was just for testing purpose, because I expect Julia to be evolving very rapidly in the next few months and I don’t want to recompile Julia every time.

Among its most interesting features are the web REPL with d3js-based visualization, and transparent FFI support. That’s a very biased summary, which comes from my looking for similar facilities with Common Lisp or Scheme at the moment.

gsl-shell

Beside Julia, gsl-shell offers a full interface to the GSL via a REPL based on the Lua JIT compiler.

It has been updated to version 2.1.1 few months ago. Older versions required to patch lua using the LNUM patch (see Lua Power Patches). However, it is no longer needed. I have had difficulty compiling it from scratch, but the Git version works just fine. There’s a static 2D and 3D graphics interface (based on agg) whose syntax is very simple yet very efficient.

I’m currently playing with GSL Shell for computational statistics. More to come later, probably.