Nial Array Language | Hacker News Reader

Nial Array Language

nial-array-language.org

50 pointsxept18 comments

I'm planning on attempting to learn an array language. This might be a good starter?

It depends on your objectives but for most people it would be better something with support for array programming like numpy, Julia or Matlab since the ecosystem is better. Otherwise I would try J.

Learn APL. Back to the roots. You’ll be pleasantly surprised. https://xpqz.github.io/learnapl

I’m learning BQN while solving AoC puzzles.

While this is my first array language and I cannot compare with others I like BQN very much:

- excellent documentation

- small but excellent community

- context free grammar

- good text editor support (I’m using neovim)

- FUN!

I had started with J and find out about Klong later, a nice niche language. I bought the book to learn about array programming in general and can recommend it: http://t3x.org/klong/book.html

A bit of past discussion:

QNial7 Array Language Announcement - https://news.ycombinator.com/item?id=14862405 - July 2017 (1 comment)

The Nial Programming Language - https://news.ycombinator.com/item?id=9074506 - Feb 2015 (9 comments)

The extreme amounts of whitespace and text zoom makes it extremely difficult to get a sense of the language.

After going to the intro, then to a chapter, then zooming to 67%... I can almost browse through some code examples and see the language. But even here, each example has extraordinary amounts of whitespace.

There needs to be a single page with substantive lines of code, at c. 67% zoom, which shows off simple-to-advanced features in one go.

Like most array programming languages, they center on the repl. It's not that the snippets have excess whitespace, it's that they're expressions being evaluated, one after the other. http://learningj.com is similar.

At least with LearningJ, the examples tend to be short because the language is more terse than others. Reading J is more like reading math than a programming language. And so the snippets tend to be smaller. Nial may be similar.

I use scala, python, etc. often -- they have terse repl output.

EDIT: https://imgur.com/a/Qs1ukGM

100%, 67% zoom, python comparison (similar examples)

I find this comment odd. It looks fine to me: https://imgur.com/5fW1OON

https://imgur.com/a/Qs1ukGM

100%, 67% zoom, python comparison (similar examples)

I wish there was a language tour, otherwise I don't know what this is.

There's a Documentation link at the top of the page [1]. I think Nial Introduction [2] is what you'd like to read first.

[1] https://www.nial-array-language.org/ndocs/

[2] https://www.nial-array-language.org/ndocs/intro/index.html

This is quite strange. It seems they are hiding the syntax on purpose, you need to dig into the documentation until you see a code example.

I'd ascribe it to ineptitude rather than malice. Either way, expecting first-time visitors to be patient and understanding enough to go looking for syntax/EBNF unaided is a mistake.

The docs themselves are weirdly written:

> The session begins with an assignment of the string ‘hello’ to the variable A. A is the variable; gets is the assign action; the string ‘hello’ ( a list of characters) is the expression.

The most egregious part about this rambling mess is that "gets" is not in the example.

It appears in later examples on the same page. I suspect that := was added as syntactic sugar but docs were not updated.

It is awesome on the quirkyness-o-meter that they (if I'm right) started with "gets" (which to me does not feel assignment-like) then sugared it into := which is very common, yet kept the name.

The GitHub README gives more context for the somewhat inconsistent state of the docs. It's an open source release of a language that had been a commercial spinoff from academia decades earlier, and now gets only occasional updates from two people as a spare-time project:

> The Nial language was developed by Mike Jenkins and Trenchard More in a collaborative research project supported by Queen's University at Kingston and IBM Cambridge Scientific Center from 1979 to 1982.

> [...]

> Nial Systems limited licensed the interpreter from Queen's University and marketed it widely. Mike Jenkins continued to refine both the language and its implementation. In 2006 Mike released Version 6.3 as an open source project to encourage continued development of Nial.

> In 2014 Mike started working with John Gibbons to develop a 64-bit version and to add capabilities that John needed for his work. The decision was made to target the open source for Unix-based platforms and release it on GitHub.

Reply on news.ycombinator.com