> How would any closed source (secrets) Mathmatica > be a better option than an open source language (Sage)? > Really want to understand the thought process. P

> How would any closed source (secrets) Mathmatica > be a better option than an open source language (Sage)? > Really want to understand the thought process. Probably because it does something Sage can't do. I can think of several reasons why I'm attached to Mma. 1) It has a great help system. Having 3 to 100 examples per help entry, and a See Also section in every single one f them is invaluable, at least for me. I almost can't use any other language's help system after using this one, because it's so good. Good yet simple; nothing prevents others from maintaining at least Examples-driven documentation but almost nobody does that, and I'm not aware of a single other documentation model that would include [consistently] both Examples and See Also sections. With its interactivity and tutorials on top, MMA's DocCenter is completely unbeatable. 2) Sage is for mathematics, Mma is for everything. That's what Wolfram says, and that's what programmers usually giggle at. Ask non-programmers, like me, about it. I use Mma for everything, and I can't imagine using Python instead. I actually feel rather weird explaining this on HN. Have you read about Blub Paradox in “Beating the Averages”? Mathematica vs Sage is Lisp vs Python. Of course I'd never use Sage. Wolfram language is really good: I use it as a basic markup for every piece of structured data that I need to write down in everyday life. I can't think of any single alternative here: other markup languages I know of are way too clumsy for this. 3) Mma has a unique interface, a mix of GUI and CLI. Nobody else has that. IPython Notebooks is a (surprisingly late) attempt to mock it, but for me it looks a bit like an exercise in sympathetic magic: for example, they've copied In and Out labels that are actually Mma expressions, that is, a part of its language. Not that it's an essential part but still, it looks like IPython authors were rather clueless about what they tried to imitate. Mma notebooks themsevles are written in Wolfram language while IPython Notebooks are written in that miserable excuse of a markup called JSON. Mathematica is Enterprise-class emacs. It's general, powerful, AND also user-friendly and good looking, unlike emacs. I can be sure that lots of people will use it who would never dive into a black hole of a terminal [until absolutely neccesary], yet all the power of CLI will be with them when they use Mma. Well, this sounds like the kind of people I want to write software for. As for closed source, Mma is just too good for that being a crucial drawback. Besides, a truly mathematical result never depends on anything external so it doesn't matter whether you use black boxes while producing it, or not. If only mathematical result is correct and valuable, then it /must/ be self-contained.