learning first data vis

I got the idea for this 2by2 matrix by this great video. The idea is simple: Some software is hard to learn, some is easy to learn. Some tools can offer a lot of functionality, some can only offer a little. One might assume that with every hard-to-learn and therefore complex tool, the most complex/individual results are possible. And Prezi at the top-left side and d3.js at the bottom-right side show, that this is a valid theory. 

Pen and paper, however, invalidate this theory (although one could argue that it's pretty hard to LEARN how to use a pen when you've never used one before). And all these tools are not on one line between "hard but lots of functions" and "easy but little functions". They are subtle differences between them. 

The best use of this matrix is maybe, to be clear about what to learn first when entering a new field. You maybe don't want to start with the hard stuff, but want to start with the easy-to-learn tools that show you best what's possible in the field.