AsciiMath

▲ AsciiMath(asciimath.org)

68 points by smartmic 10 hours ago | 20 comments

▲ gucci-on-fleek 31 minutes ago | parent | next [-]

The author of ConTeXt (a TeX format similar to LaTeX) has some interesting comments on AsciiMath [0] [1]. Its space handling looks especially problematic; the example given in [0]

  o ox x = xo
  a ax x = xa
  ooxx=xo
  aaxx=xa
  ac+sinx+xsqrtx+sinsqrtx+sinsqrt(x)

produces the following output

  o ⊗ x = x o
  a a x x = x a
  ∞ × = x o
  a a × = x a
  a c + sin x + x √x + sin √x + sin √x

Its handling of commas looks even worse, but it's tricky to demonstrate that in plain text.

[0]: https://tug.org/TUGboat/tb36-2/tb113hagen.pdf#page=3

[1]: https://github.com/contextgarden/context/blob/e9bd55ec/tex/c...

▲ tombert 5 hours ago | parent | prev | next [-]

I use Typst a lot now (which this reminds me of), and the equation support is generally very good, but the thing that gives me pause is that I'm afraid that there's going to be something missing, or worse than the LaTeX equivalent.

LaTeX has been the industry standard for the mathematical world for decades and as a result it has had the most work adding new notation or making nicer formatting.

For example, I needed to do a proof tree recently. Typically I would use bussproofs in LaTeX but I was using Typst, and while there is a package for handling proof trees in Typst [1], I think they're not very pretty, and as a result I ended up porting the document over to Pandoc markdown and doing the rest of my work there (which is annoying because Typst renders around ~1000x faster and has better tooling).

[1] https://github.com/SkiFire13/typst-prooftree

▲

dadoum 3 hours ago | parent [-]

I remember using curryst for my proof trees (a few months ago) and they looked fine if I recall correctly. But I agree that often using typst means searching for package that may not exist or is not working correctly since the ecosystem is not very mature currently.

	▲	tombert 2 hours ago \| parent [-]
		Hadn't seen curryst. Looking at the examples it looks ok. Maybe I should have used that and stuck with Typst.

▲ agnishom 6 hours ago | parent | prev | next [-]

One could just use the math fragment of typst, no?

https://typerino.com/

▲ BruceEel an hour ago | parent | prev | next [-]

Neat. Personally, I wasn't aware of mathjax, it's a bit of a revelation to learn you can do this. I like very much the fact that the ASCII side is highly readable - compared to say, LaTeX (to me, anyway!) - as something I could use in code comments.

▲ NIckGeek 3 hours ago | parent | prev | next [-]

I'm a big fan of AsciiMath and have been supporting it in my note taking program (MicroPad) since ~2016.

It was the key for me being able to write maths in a classroom/lecture theatre at the same speed (or faster) than those doing it by hand.

▲ MichaelNolan 6 hours ago | parent | prev | next [-]

What exactly is the distinction between this and mathjax? At first glance it looks like this is a wrapper on top of mathjax, or is it something different?

I guess it’s just more “natural” and less latex like.

$$\sum_{i=0}^n i^2 = \frac{(n)}{2}$$

sum_(i=o)^n i^2=((n)/2)

	▲	bobbylarrybobby 4 hours ago \| parent [-]
		asciimath does a lot of things automatically, like parenthesis sizing and auto-frac, that latex requires you to do yourself. \left and \right and \frac do add quite a bit of noise to simple equations.

▲ hota_mazi an hour ago | parent | prev | next [-]

What strange choices. For example, to express the sum from i=1 to n:

    sum_(i=1)^n

Why use the exponent sign to indicate the upper limit? Am I taking crazy pills here?

How about using some symmetry instead, e.g.

    sum(i=1)(n)    "Means: sum from i=1 to n"

	▲	zkmon 31 minutes ago \| parent [-]
		That's because the positioning of n is similar to that of exponent? As author says, this is more about expressing "visual rendering" using the text. Hence the term "ASCII math" like in ASCII Art.

▲ runarberg 4 hours ago | parent | prev | next [-]

Shameless plug: I made a competing library to asciimath called mathup

https://mathup.xyz

▲ adamnemecek 7 hours ago | parent | prev | next [-]

If you like this, check out typst https://typst.app

▲ bottlepalm 6 hours ago | parent | prev | next [-]

It's too bad MathML never supported ASCII math. The web might be a different place if it did.

	▲	Rendello 6 hours ago \| parent \| next [-]
		Asciidoctor renders AsciiMath to MathML [1], I imagine that it's a fairly common target after MathML (Core) was revived from the dead. While MathML is pretty verbose to write by hand, I think things like AsciiMath and LaTeX are on a different level of abstraction. You can't embed AsciiMath directly in HTML in the same way that you can't embed Markdown directly in HTML. 1. https://docs.asciidoctor.org/asciidoctor/latest/stem/asciima...
	▲	Mikhail_Edoshin 4 hours ago \| parent \| prev [-]
		XML as a standard has a concept of "notations": you can specify that contents of an element are written in such-and-such notation, which can be "asciimath" too. Of course at the XML level these are merely labels; you still need to agree on notation names and make the processors of that document to understand that notation. But as a foundation it is there. So technically a web could indeed be a different place: a network of XML documents where all the original notations are kept as they are and merely marked with tags that formally specify: this is AsciiMath, this is C, this is SQL, and so on.

▲ JoelMcCracken 7 hours ago | parent | prev | next [-]

This is cool. I could see myself using this for notes.

	▲	Rendello 6 hours ago \| parent [-]
		You can use it with AsciiDoc readily, if you use that [1]. With anything you could also use MathML in an HTML-passthrough block, but it's pretty verbose. 1. https://docs.asciidoctor.org/asciidoctor/latest/stem/asciima...

▲ lutusp 2 hours ago | parent | prev [-]

Great! Another LaTeX competitor, doubtless "better" for an obscure reason known only to its author. Especially appealing is the fact that, when embedded in a Web page, it must be translated into LaTeX syntax before rendering by MathJax.

The "AsciiMath" name reveals volumes, because prior to rendering, LaTeX code is already ASCII characters meant to represent math symbols. We just didn't call it that.

Oh well, a tempest in a teapot, soon to be forgotten. We can already tell a chatbot, "Show me the tensor equations of General Relativity, and render the result in LaTeX."

I close with the obligatory XKCD reference: https://xkcd.com/927/

	▲	forgotpwd16 11 minutes ago \| parent [-]
		>for an obscure reason known only to its author It's, in project's words, simple calculator-style syntax (can also call it simplified LaTeX subset) made to easily embed math on web pages by converting to MathML, with its existence predating MathJax by few years (and even MathJax's predecessor, jsMath). It was never meant to be LaTeX competitor. With last point, have noticed people most often use this xkcd strip opposite to what it means. It's about when, for a particular use case, one standard/tech/whatever tries to replace all others rather when one standard/tech/whatever attempts to fulfill a distinct use case.