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

The Crystal Programming Language

crystal-lang.org

Post reply on HN
Print view View original

11–20 of 180 posts

Re: The Crystal Programming Language

#11

Post by asterite »

kibwen wrote: post #10
asterite wrote: post #5

Earlier quoted context omitted.

It has global type inference, there's nothing dynamic in the language. You can specify type restrictions to allow overloading methods, for example, or doing multiple dispatch. But in the general case you don't specify types (except for generic type arguments) and the compiler figures out everything.

I'm curious how possible it is in practice to write large software that relies exclusively on global type inference. Does the compiler itself go without ever explicitly specifying a type?

Yes, you can browse the source code if you want, check it out and compile the compiler. It has between 30k and 60k lines of code (because you have to consider it also includes the standard library) and on a Macbook Pro 2015 it takes less than 10 seconds to compile (in non-release mode).

We'll have to wait until we get project that's larger than the compiler, but we believe there's still room for performance improvements.

Re: The Crystal Programming Language

#12

Post by trengrj »

I've been looking at the benchmarks https://github.com/kostya/benchmarks and am pleasantly surprised about the speed. It definitely smokes Ruby, but also is usually faster than Go. I know that all benchmarks are relative but Crystal seems a great language from a performance viewpoint.

Re: The Crystal Programming Language

#14

Post by logicchains »

asterite wrote: post #11
kibwen wrote: post #10

Earlier quoted context omitted.

I'm curious how possible it is in practice to write large software that relies exclusively on global type inference. Does the compiler itself go without ever explicitly specifying a type?

Yes, you can browse the source code if you want, check it out and compile the compiler. It has between 30k and 60k lines of code (because you have to consider it also includes the standard library) and on a Macbook Pro 2015 it takes less than 10 seconds to compile (in non-release mode). We'll have to wait until we get project that's larger than the compiler, but we believe there's still room for performance improveme…

It seems to have a fair slew of type-system features, so how does it manage to compile so fast, compared to for instance the Rust compiler, which doesn't do global type inference?

Re: The Crystal Programming Language

#15

Post by MCRed »

I love that we're getting new languages lately, but almost all of them seem to be ignore the significant new requirement of our age: parallelism & concurrency.

Specifically, you need lightweight processes and no-shared-memroy architecture.

While the number of cores on a machine is remaining relatively low, the number of machines in a system are going up.

Erlang got this right and build a lot of infrastructure around it (OTP) and while you can't replicate that infrastructure quickly you can get the fundamentals right.

Simply getting this wrong rules out a lot of languages from consideration (because why learn a new language that is going to be obsolete, or only chosen by people who don't understand how to build systems?)

It's a lot easier to get this in when you're new and can make major changes to the language. Once you start to solidify it would break things- this is why Go's fake concurrency is a tragedy and a huge missed opportunity.

Re: The Crystal Programming Language

#17

Post by rifung »

MCRed wrote: post #15

I love that we're getting new languages lately, but almost all of them seem to be ignore the significant new requirement of our age: parallelism & concurrency. Specifically, you need lightweight processes and no-shared-memroy architecture. While the number of cores on a machine is remaining relatively low, the number of machines in a system are going up. Erlang got this right and build a lot of infrastructure around…

I largely agree with what you are saying, but doesn't it only apply to languages that will be used in back end systems? If you are just making a phone app or something, it doesn't seem like parallelism is nearly as important.

Re: The Crystal Programming Language

#18

Post by seanmcdirmid »

asterite wrote: post #5
kibwen wrote: post #2

Crystal looks like a neat language, and it's fun to see how many entrants there are in the modern renaissance of scripting languages that compile to native code. I wish there was some more documentation about it though... for example, does the bullet point "Never have to specify the type of a variable or method argument" from the home page imply gradual typing as per Dart and TypeScript, or is it something else entir…

It has global type inference, there's nothing dynamic in the language. You can specify type restrictions to allow overloading methods, for example, or doing multiple dispatch. But in the general case you don't specify types (except for generic type arguments) and the compiler figures out everything.

What algorithm are you using? How do you deal with subtyping and paramatricity?

Re: The Crystal Programming Language

#19

Post by reagency »

kibwen wrote: post #10
asterite wrote: post #5

Earlier quoted context omitted.

It has global type inference, there's nothing dynamic in the language. You can specify type restrictions to allow overloading methods, for example, or doing multiple dispatch. But in the general case you don't specify types (except for generic type arguments) and the compiler figures out everything.

I'm curious how possible it is in practice to write large software that relies exclusively on global type inference. Does the compiler itself go without ever explicitly specifying a type?

GHC does global type inference, and has very sophisticated types.
Post reply on HN