Stack Exchange network consists of 183 Q&A communities including Stack Overflow, the largest, most trusted online community for developers to learn, share their knowledge, and build their careers.
@sh1: Two of my most highly voted answers are quite heavy on the math: stackoverflow.com/a/6661242/524368 and stackoverflow.com/a/5257471/524368 – both of them are lacking in the quality of the mathematical typesetting. In case of the "how to calculate normals" answer I was able to make use of Unicode to somehow typeset it. The answer on how to derive a tangent space mapping tries to use ASCII art to typeset matrices and fractions, but it looks ugly (in fact it's that later answer, which lack of proper math typesetting which made me ask about adding support for TeX/LaTeX at SO).
@sh1: And then there's of course the broader issue that, thanks to the Curry–Howard correspondence writing a program is technically the same thing as writing a mathematical proof. So stating that programming and asking questions on StackOverflow is about how to call library functions is highly ignorant. And given that some languages (like Coq which allows you to write mathematical proofs and translate it into programming language source code), yes math typesetting is absolutely required.
@sh1: Some of the mathematical notation, for the most part linear algebra can be typeset using unicode symbols. But already something as mundane as superscripts or subscripts is impossible with just markdown and unicode. Sometimes you want to typeset a partial differential equation, that can be found in a book or a paper about a particular graphics rendering technique and want to explain how to translate that into library calls.
@sh1: To properly use a library one must understand the underlying principles that are to be achieved with doing these calls. Take computer graphics programming for example. A recurring question is, why normal vectors muse be transformed using a different set of operations (i.e. library calls) than regular vertex positions. Another recurring question is, why a separate tangent space transformation is required for normal map bump mapping. Just giving the recipe is only half the answer. It's also essential to understand "why". And the most concise way to explain the "why" is through math.
Yes, I'm aware of this, but I was puzzled by the -100; of course at Mat already pointed out, the user could have cast votes on 10 of my answers. Silly me.
