hhhmmm...at one time or another, a few years ago, I visited some of the webpages for the software listed in the link...at the end, I simply chose Python (numpy, scipy, matplotlib, pyQt, etc).
As a non-user of the software, a brief homepage visit and maybe a trial usage, I have the following speculative opinions...
Sage is huge, a beast, a conglomerate of many other programs (dozens) unified under a common python interface; needless to say, one would need to really know how to use those other individual programs in order to use them through Sage.
Scilab and Freemat are a-la-matlab kind of programs. Scilab seems rather popular with literature, conferences, etc.
Euler seems for mathematicians, fancy plots, reminds me of Mathematica.
Julia is picking up good, seems designed for good performance, very friendly language but I am not sure what kind of shape is in regards to libraries...I would like Julia to be my programming language of the future...we shall see...keeping an eye on it.
...for now, I am sticking with Python...many libraries not only for data analysis but also for plotting, and even for quick application prototyping, I am talking enabling some of us non-programmer engineers/scientists fairly easily and quickly build basic GUI-based applications (pyqwt, enthought, pyQt, pySide). Besides, once you learn Python, you will also be able to script many other applications that have chosen Python as their scripting language.
Anyway, that's just my non-educated opinion, no need to sue me.
Lastly, do you know anybody in the field that you are looking to get into? Have you asked anyone there what they use?
