CSTN VS na

My recommendation is to just write something, even if it sucks. That goes for any concept. You'll learn faster and better by interacting with the machinery yourself versus trying to interpret someone else's abstract understanding of the machinery. In this case, that means choose a simple language or write your own grammar to play with, and make a parser for it. The first real parser I made is a recursive descent parser that parses a relative of JSON. If you're curious, my code is available), but I was a lesser programmer when I wrote it, so don't take it as an example of how you must do things. Regardless, it does work. I've continued to use the character stream code in every text parser I've written since, with some improvements.

So far I've only documented the syntax.

I tried to distill down the most essential features of TS/JS (functional, prototypal) and then come up with new syntax and semantics that was minimal, orthogonal and hopefully easy to learn and use. The result is kesh and na.

Improved na's documentation. na is the underlying data notation format.