A fun post exploring the differences between a number of statistical computation packages. As one of the commenters said, this is an awesome flame war! The comments are very informative, with all sorts of historical information explaining how and why certain packages turned out the way they are. I use JMP at work, which is owned by SAS, but is not near as common as these other packages. JMP is pretty easy to use, and looks good, which is its major selling point. I find that I often with other programs were as good at making and analyzing graphs as JMP is. JMP is excellent for exploratory data analysis. However, it is less powerful than most of the packages analyzed in the post, so it will probably remain a smaller player.
h/t John D. Cook