Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Recto – A Truly 2D Language

masatohagiwara.net

Post reply on HN
Print view View original

61–70 of 72 posts

Re: Recto – A Truly 2D Language

#61

Post by impulsivepuppet »

mhagiwara wrote: post #55
psychoslave wrote: post #32

>Virtually all the languages we humans use—spoken, written, or artificial (such as programming languages)—are fundamentally one-dimensional. Anyone having a look at a musical score can conclude this is wrong. Even when it comes to spoken language, intonation is conveying a lot, even in languages which are not classified as tonal. All that said, that's nice to see an exploration of exotic expression form. It doesn't n…

Hi! Author here. You're right that music and speech carry more than just raw symbols such as intonation, pitch, dynamics, etc. all add important information. But in the framework I'm using, both are still fundamentally “1D” because their core structure unfolds sequentially in time: notes in a score, words and phrases in speech. Even when additional features like accent or dynamics are present, they're layered onto a…

I find it quite intriguing to introduce "language-native" matrices and 2d blocks (which I still find difficult to wrap my head around.)

The reason why most people would more intuitively consider a music score as multidimensional has to do with parallelism or concurrency.

In theory, nothing is stopping you from creating a hyperarray language a la BQN++ (or dare I say QRP). Maybe I glossed over an example, but having proper pointwise application to hyperscalars feels like a must-have.

Second idea is to introduce process parallelism, which could actually make this form of syntax into an execution graph of sorts--could be quite promising!

Re: Recto – A Truly 2D Language

#62

Post by ofalkaed »

mhagiwara wrote: post #55
psychoslave wrote: post #32

>Virtually all the languages we humans use—spoken, written, or artificial (such as programming languages)—are fundamentally one-dimensional. Anyone having a look at a musical score can conclude this is wrong. Even when it comes to spoken language, intonation is conveying a lot, even in languages which are not classified as tonal. All that said, that's nice to see an exploration of exotic expression form. It doesn't n…

Hi! Author here. You're right that music and speech carry more than just raw symbols such as intonation, pitch, dynamics, etc. all add important information. But in the framework I'm using, both are still fundamentally “1D” because their core structure unfolds sequentially in time: notes in a score, words and phrases in speech. Even when additional features like accent or dynamics are present, they're layered onto a…

Western harmony can not be understood in one dimension, it is very much in two dimensions.

Re: Recto – A Truly 2D Language

#63

Post by almostgotcaught »

Sharlin wrote: post #58
almostgotcaught wrote: post #50

Earlier quoted context omitted.

> of course not a language in the same way English or Spanish are languages, but certainly a language in the same way mathematical formula, Lisp, or Java are languages. It is mostly certainly not - the defining feature of all languages (human, formal, mathematical, programming) is that you can define new words/concepts/primitives. Sheet music is notation.

Then regex, HTML, YAML, and so on are not languages either according to your definition. Yet they are languages according to formal language theory, a definition which I’d say is a bit more widespread than yours. Indeed what you propose roughly maps to (a subset of) context-sensitive languages in particular. Defining new primitived isn’t really essential. What matters is what you can express with the existing primiti…

> Yet they are languages according to formal language theory, a definition which I’d say is a bit more widespread than yours. Indeed what you propose roughly maps to (a subset of) context-sensitive languages in particular.

I love when people talk down to me as if I don't have a PhD in this shit lol. No actually what I'm talking about is Recursive languages.

> Then regex, HTML, YAML, and so on are not languages either according to your definition.

Do you see the M before the L there in both HTML and YAML? Do you know what it stands for? Markup. Markup is a funny word that colloquially means the same thing as annotate which bears a strong relationship to the word notation hmmmmmmmm.

Let me put it simply: if the force of the original comment is that context free languages can be 2d then I would say..... no duh.

Re: Recto – A Truly 2D Language

#64

Post by bawana »

Why stop here? Cant we endow macdraw with an LLM? With that kind of graphic- text language, you can draw structures (like you might on a napkin) with arrows resembling trees or flow diagrams that do computation. In this kind of macDraw-LLM language, we can use graphic primitives. As you zoom in, the primitive can reveal lines of code. Zoom out and you have a high level view of code blocks. Now those silly diagrams of transformer models actually become the code.

Re: Recto – A Truly 2D Language

#65

Post by Timwi »

Aw man, no mention of Funciton :) [0]

I believe I understand where the designer is coming from. Traditional languages use a 1D syntax to express a structure that is really not 1D (the parse tree). However, I don't think it's really 2D either. If you think about how an interpreter would have to proceed to execute Recto code, it would still parse it into an internal representation and then execute that, and this internal representation no longer uses the 2D layout of the code.

I'm not implying that it's a bad idea or anything. To the contrary, I think Recto is a cool idea. I'm just skeptical that simply making a language syntax two-dimensional automatically makes it easier to understand, code in, collaborate in, or implement a parser/interpreter/compiler for.

[0] https://esolangs.org/wiki/Funciton

Re: Recto – A Truly 2D Language

#67

Post by dako2117 »

almostgotcaught wrote: post #37
psychoslave wrote: post #32

>Virtually all the languages we humans use—spoken, written, or artificial (such as programming languages)—are fundamentally one-dimensional. Anyone having a look at a musical score can conclude this is wrong. Even when it comes to spoken language, intonation is conveying a lot, even in languages which are not classified as tonal. All that said, that's nice to see an exploration of exotic expression form. It doesn't n…

> musical score sheet music isn't a language, it's a notation. it's just a concise way to represent a waveform (a song). and it's certainly one dimensional because you either play the song forwards or backwards.

>written language

written language isn't a language, it's a notation. it's just a concise way to represent a waveform (a spoken sentence). and it's certainly one dimensional because you either read forwards or backwards

Re: Recto – A Truly 2D Language

#68

Post by dahart »

gmueckl wrote: post #44
Ferret7446 wrote: post #43

Earlier quoted context omitted.

It's all about semantics and representation. All data is (can be represented as) one dimensional. Musical scores are definitely on the one dimensional side of the spectrum however (e.g., see Lilypond).

Music notation is definitely not one-dimensional because it encodes simultaneous occurrence and simultaneous time-progression of independent pitches (e.g. a sequence of cords played legato).

Not necessarily true; this depends entirely on which music notation you’re talking about. Chord charts handle simultaneous pitches easily using a 1D flow of text along the timeline. Most music notations, including standard staff notation, use a layout that has a primary linear time axis, with lines that wrap, just like written text. Staff notation is different, but not that different from text. You can think of pitch as a secondary embedded dimension, but it’s certainly not necessary. You can also notate pitches and pitch combinations symbolically. The resulting audio, whether spoken or played musically, is 1-dimensional (in time) with respect to an idealized single point listener. So at some level, using anything more than one dimension is a type of feature expansion - it’s useful for notation and control, but adds new dimensions that are over-specified and unnecessary to the final result.

Re: Recto – A Truly 2D Language

#69

Post by Mars008 »

Timwi wrote: post #65

Aw man, no mention of Funciton :) [0] I believe I understand where the designer is coming from. Traditional languages use a 1D syntax to express a structure that is really not 1D (the parse tree). However, I don't think it's really 2D either. If you think about how an interpreter would have to proceed to execute Recto code, it would still parse it into an internal representation and then execute that , and this inter…

> how an interpreter would have to proceed to execute Recto code, it would still parse it into an internal representation and then execute that, and this internal representation no longer uses the 2D layout of the code.

Adding parallel execution will make it true 2D at execution time. Even though it's sequential code. More over it can be logically, architecturally 2D while being 1D from interpreter's standpoint. If it uses cooperative multitasking with 1 running task at a time. Or simply 1 core processor.

Re: Recto – A Truly 2D Language

#70

Post by apgwoz »

archargelod wrote: post #54
apgwoz wrote: post #48

In the very early 2000s, a coworker of mine, who was frequently banned from the Something Awful forums, shared with me a language, which I think was called “Path”, that was 2D. IIRC, execution of code followed a textual path through the file. I seem to remember it being sort of like a 2D brainfuck. You essentially drew a map to compute. When I look at Orca ( https://100r.co/site/orca.html ) I’m reminded of it, but it…

A very simple query "PATH programming language" with SearXNG returns [1]. I think it matches your description perfectly. > 2003 > PATH is a programming language that is similar to Brainfuck. This intuitive language requires a simple grasp of two dimensional space, one requirement that almost no other programming languages have. [1] https://sourceforge.net/projects/path/ [2] https://pathlang.sourceforge.net/

Yes! This is it, thank you!
Post reply on HN