Show HN: LCON – Ludicrously Compact Object Notation | Hacker News Reader

Show HN: LCON – Ludicrously Compact Object Notation

github.com

35 pointsar-nelson24 comments

Suggestion: Show both the JSON and LCON representation for the examples. This would help illustrate the difference between the formats.

The examples are all LCON versions of JSON documents that are out there on the web. I provided a link to the JSON before each example. Side-by-side JSON and LCON would have been nice, but you can't do floats in Markdown, and putting the JSON and LCON one after the other would have made the overly-long README even longer.

Easy fix: just align the JSON examples out to the right of the LCON ones inside of your MD code blocks.

1. What problem are you solving with this? This really seems like a solution in search of a problem.

2. There's no emitter, just a parser. This suggests to me that there's no such thing as a canonical LCON transformation for a given input. This is bad.

I guess the canonical JSON could be considered the canonical LCON...

1. It's a byproduct of another project I've been working on, a programming language that's essentially JSON Lisp. I tried both YAML and CSON as syntax for the language, and neither worked quite like I wanted. The features that make it useful for the language are:

- Easy chaining of one-element maps. `a b c d` = `{a: {b: {c: d}}}}`

- Only a few reserved characters, so just about anything can be an unquoted string (symbols, etc.)

- Parenthesized maps

2. This is the first version. An emitter will probably be in the next version. The canonical version is essentially "JSON, but with unquoted strings where possible, no colons, and parens instead of braces".

I love yaml because I can see it, and it (usually) makes sense. Same thing with Markdown. They both visually represent the data/content in a manner that my brain gets.

This seems less obvious (or slower to digest) but I've stared at yaml and markdown for a while now.

This is compact, sure, but I have a hard time reading it.

I'll stick with JSON.

Agreed, but I'll stick with YAML.

I use json when I need to read it and messagepack when I need to send it over the network.

Fair enough; this seems to be the most common complaint I've seen here. There's nothing stopping you from including colons, at which point it would look exactly like YAML and CSON, but then I suppose it would have very little added value beyond YAML or CSON. For me, the killer feature is chaining one-element maps on a single line (CSON can do that too, but it doesn't have unquoted strings).

While that might be convenient, it gives me a hard time comprehending what the actual structure of the map is.

I like how this is presented to be ludicrously compact, but it uses double quotes "" instead of single quotes ''. Everybody knows that single quotes take up less space!

If you're really looking for efficiency, take a look at baconification ;)

http://pastebin.com/RGRfewBY

...comes in handy for those occasions when the courts order you to hand over some data - especially effective when printed.

For anybody who wants a little bit more structure than LCON but less clutter than JSON, I'd recommend checking out CSON.

https://github.com/bevry/cson

This project was partially a response to CSON. CSON was almost (but not quite) good enough for my purposes, so I made something that was.

There's a few really awkward corner cases that are much clearer in YAML (and, it seems, LCON). The most significant one, to me, is lists of objects, which are perhaps not so common in code literals (hence why people don't tend to run into it in coffeescript), but do tend to come up a lot in structured documents. In JSON they are trivial:

    [
      {
        "a": "b",
        "c": "d"
      },
      {
        "e": "f",
        "g": "h"
      },
    ]

In YAML also trivial and very concise:

    - a: b
      c: d
    - e: f
      g: h

But in CSON (if I remember correctly), I think this looks odd and confusing by comparison:

    [
      a: "b"
      c: "d"
    , # <- note dedent, without this it comes out as one object not two
      e: "f"
      g: "h"
    ]

Now, in each of these you can just revert back to JSON for the difficult part, but that kind of defeats the purpose. I think YAML-the-syntax strikes a really nice balance between eliminating unnecessary punctuation and having solid structure, and LCON seems like it does as well. CSON errs too far on removing punctuation without a lot of thought put into the remaining structure.

If I understand correctly, the above example in LCON would be (at most minimal):

    . a b
      c d
    . e f
      g h

I actually really dislike this use of . and would like to know a rationale for it, to be honest.

This was one of the reasons I designed it the way I did; I think it's a nice middle ground between YAML and CSON.

As for the ., I chose it over - or * for the bullet character because I designed LCON as the syntax for a JSON-Lisp programming language, and - or * could conceivably be used as prefix operators for math functions (* [5, 6]), leaving . as the only "bullet-like" ASCII character left.

The more I write LCON, the more I realize that the . can be difficult to see (though it's still not as bad as CSON's dedented commas...), so I can see your point, I'm just not sure what other character to use. In an earlier version, I had used :: as a bullet, but that ended up looking ugly.

I'd say just use hyphens like YAML, but I guess I get the keyspace-advantage of using a dot. Still an aggravating change.

If I was working in a language that used isomorphic json-like, I think I'd expect it to have call signatures more like smalltalk (object verb subject) than lisp (verb object), so calling '-' or '*' would seem strange to me.

But I don't know anything about this other project of yours other than what you've said here.

For another compact format (exclusive to homogenous collections), checkout JSONH https://github.com/WebReflection/JSONH

I'm waiting for ubjson http://ubjson.org which is probably the most compelling json replacement I've seen yet.

How does this compare to JSON when gzipped? I imagine any size benefit of stripping colons and parenthesis is negligible compared to a good compression.

The goal is to save keystrokes, not file size.

Reply on news.ycombinator.com