[andrew_eit]

I really wonder what percentage of difficulties that stem from math are due to its representation in the symbolic language used. By that I mean, so often, I am staring at a long, condensed equation. Factored to perfection by whomever was working on developing it. I stare at it for hours and just can't figure out what's going on.

Then after much tinkering with the parameters, exploring the limits, plotting graphs, slowly I look at it and recognize what is going on and realise "gosh that is actually really F**ing simple"

I wonder how much of "learning" math is really spent on decoding the representations we are provided, rather than understanding the concept they're meant to represent.

[lysecret]

That is a really good point. I spent some time putting ml equations into code and the code representation always seemed so much simpler to me. Many math equations seem a bit like too clever one liners in code haha

[andrew_eit]

"clever one liners"

That's exactly it! It's basic clean coding style where it's better to use meaningful names and sometimes write something in 2 or 3 lines that is more interpret able than to write a perfect one-liner that no one will be able to unpack.

In this regard I almost wish we had a de-facto standard in math of presenting the 'condensed' perfect form of the equation and a more chunkier version.

What I like is that recently I saw some researchers annotating the symbols in their equations [1].

[1] https://twitter.com/sibinmohan/status/1480583840858996743/ph...

[johnaspden]

This is how all of maths is. Before you learn it, it's incomprehensible frightening gibberish. After you learn it, you wonder how anyone can not understand ideas so simple and obvious.

How to get from one state to the other is the whole problem.

The terse symbolic language is someone's best attempt to communicate the beautiful simple idea in their head. More than any other discipline, I think, mathematicians write to be understood, to be clear.