In fact, Mathematica is not as flawless as WRI would like you to believe; just read sci.math.symbolic. Free software holds up pretty well, as it does against other famously well funded monopolists.

- is irreproducible, un-reviewable science Science?

Jordi was following the implications of an everyday scientific tool being completely, permanently closed. His points still stand (rephrased as questions):

- what does it mean not to know how that tool works?
- what does it mean for scientific results when their reproducibility depends on a closed tool?
- what does it mean when, inevitably, that tool is no longer available?
- what does it mean to put a paywall (Mathematica license) in front of scientific activity, access to data, methods, and results?

These are new issues (in the history of Science), and essentially unexamined. (I and friends had been deeply concerned about these questions before I found out about Jordi's essay.)