[YeGoblynQueenne]

I'm thinking of giving markdown a try, for the reasons you say, but I'm not sure how the maths notation looks like when it's not rendered. I think it's all escaped characters, in which case it won't be as readable as plain text, even with rare glyphs that need special fonts.

Some text editors even let you enter rare characters with key combinations (frex, Vim's digraphs). The end result is pretty much WYSIWIG mathematical notation.

Here's an example from my machine learning class notes:

        ⎛ ₙ       ⎞

ƒ̂(x) = θ⎜ ∑ wᵢxᵢ ⎟

        ⎝ ⁱ⁼⁰     ⎠

So, that looks a mess on HN, but if you copy/paste it in a decent text editor, with a font that has all the necessary characters, you'll get a nice formula that you can keep around for as long as there's text editors supporting ASCII- and you don't have to rely on any special tools to render it properly.

That's what I'm going on about- but there's not enough super/subscripts in ASCII itself. :)

[teach]

I'm not sure you're using the term "ASCII" correctly. There are virtually no characters in ASCII that aren't printed on a key on a typical US keyboard.

Especially when you say "with a font that has all the necessary characters". ASCII is ASCII. All fonts (except Wingdings) have 100% of the ASCII characters.

Maybe you're pasting in Unicode; maybe you're using the Windows-only ALT-numeric keypad garbage.

But I suspect your "nice formula" that renders in a text editor maybe isn't as clean or standard as you think.

[YeGoblynQueenne]

Sorry, I meant "unicode", not ASCII, my bad. :)

I can't edit it unfortunately so it has to stay there, in perpetuity reminding me of my failings in life.

[bgar]

If you use Emacs' Org mode, you can actually convert parts of the buffer to LaTeX and have them inline. This solves the problem of equations being not as readable as you're writing.