Idealism vs Realism in programming?
Idealism strives for abstraction, expressiveness, productivity, portability, …
Realism is driven by implementation, efficiency, performance, system programming, …
From the beginning, we were certain that AND is possible. Consequently, we became hybrid programmers, applying the programming paradigms that fit to the problem. We integrated our numerically optimized C++ engines into the idealist Mathematica (Wolfram Language) that also is very pragmatic.
It supports several levels of interfacing to C/C++ programs, it invokes C compilers from within Mathematica, provides SymbolicC - manipulate C syntax in Wolfram Language, calls C programs from Wolfram Language and Mathematica from C programs, provides LibraryLink - loading a function from a dynamic library into Mathematica.
On the side of idealism the Wolfram Language is a highly developed symbolic language, which unifies a broad range of programming paradigms, using its unique concept of symbolic programming to add a new level of flexibility (references are: Symbolic Expressions, Lists and Tables of Expressions, Functional Operations, Rules & Transformations, Pattern Matching, Logic & Tests, ..).
UnRisk exploits this idealism and realism in the UnRisk Financial Language - by providing symbolic financial constructs atop the Wolfram Language and extending Mathematica's interpreters into the universe of derivatives and risk analytics.
It is so simple, but as I have pointed out in writing about music is like dancing about architecture you need to try it.
Picture from sehfelder
