Comment by Hexstream | Hacker News Reader

Comment by Hexstream | Hacker News Reader

Ok, that was painful to read by moments.

"Ruby is probably the most powerful programming language on the planet for creating DSLs, or domain specific languages."

If Ruby has anything over Lisp in the DSL domain (besides popularity), I'd love to know.

"A web development platform as powerful and syntactically concise as Rails could only have been done in Ruby."

Idem.

"This power does not come without a price. Ruby is slow. Ruby is generally recognized as being twice as slow as Python, and at least an order of magnitude slower than C++ or Java. Charles Nutter, one of the creators of the JRuby VM for the Ruby language, recently posted an article about optimizations that could be made to the VM, and he improves Ruby performance by as much as twenty times by removing much of Ruby’s power and breaking compatibility. If anything, it shows us skeptics that there is no such thing as a free lunch."

Lisp has all (?) the crazy features you love so much about Ruby and more, yet it's very fast (try SBCL, it compiles to native x86).

Ruby has an advantage over Lisp: syntax

Many programmers like syntax, since coding AST-style can be tedious sometimes. Sure, you can create your own reader macro with Lisp, but that is harder than overriding some methods in Ruby. Rails is an example that opinionated software has an edge sometimes and Ruby compared to Lisp is opinionated in terms of syntax.

I don't see how the fact many programmers like syntax makes it better.

I also don't see how "coding AST-style" can be tedious, especially since it's so much easier to automate the process with macros. Granted, you'll want an editor with support for s-expressions.

The same way why ease of use of something makes that something better (or more pleasant to work with).

Bingo for the parent!

There are indeed some people who say "uh, nice syntax, readability, uh, big deal". Those are the folks who fail to grasp the point that programming is a human endeavor and that having your code read by other humans is a crucial element.

Lisp may indeed attract a large group of nice, intelligent people who don't get this. That is a further, fatal problem in the language.

Ruby has many problems itself but it gets that readability is important and it gets that, by extension, the humans relations that a programming language generates are important. Lispers don't seem to see the problem in an exclusive attitude and essentially closed community. The ruby community understands the need to have a good community open to people even if the Ruby community doesn't always succeed at this. It all comes together

This (and the rest of the article) is just ignorant cheerleading — an attempt to justify the lackings of the author's chosen technology, and thus resolve cognitive dissonance.

Lisp isn't quite as dynamic as Ruby, but Javascript certainly is, and it can be as fast as Java.

I agree that Lisp is a better choice for DSLs than Ruby. If you want a very dynamic (late bound) language that is also fantastic for DSLs, have a look at Io (http://iolanguage.com/).

"Lisp isn't quite as dynamic as Ruby"

?! Please substantiate. And of course when I say "Lisp" I really mean "Common Lisp".

I'd also like some substantiation around the claim that Javascript can be as fast as Java.

Rhino?

I'm guessing; I have no numbers on how it compares to actual Java.

That's like saying that ruby is probably as fast as c because the interpreter is written in c.

http://tinyurl.com/daom93 - shootout numbers rhino v. java

"That's like saying that ruby is probably as fast as c because the interpreter is written in c."

I was under the impression that Rhino was doing some compiling to byte-code (unlike what MRI does, but what perhaps like JRuby does).

Live and learn.

WTF is with modding people down for saying they took a guess, were wrong, and now learned something? Does that in any way add to the discussion?

Thou shalt not mock Lisp or be in any way associated with one who does:

http://news.ycombinator.com/item?id=545859 http://news.ycombinator.com/item?id=545851

Speculation on the behavior of Rhino equals mocking Lisp?

Wow; who knew?

Yeah, I noticed that as well... Kind of bizarre. Especially since Rhino was a pretty good educated guess; I went and took a look at it as well since I know there are some pretty performant languages running on the JVM that aren't Java.

For whatever it's worth, I think the problem Javascript is going to have with speed is that it is so dynamic that it makes it difficult for compilers to optimize. If it ever got type hints it would probably be much easier to make fast.

Sorry that sounded as snarky as it obviously did. I thought I was just drawing an analogy. Anyways...

The semantics of javascript are so different to Java (javascript functions are not like java functions everything is dynamic etc.) that you can't really expect the performance of javascript code to be comparable to Java just because it's hosted on the JVM, even if it did compile to Java byte code.

The Javascript equivalent would have to do so much more work than the Java version to support all the crazy crap you can do in javascript (ad hoc adding of properties to almost any object at any time).

Thanks for the clarification.

I did pull that out of my ass, but Nitro and V8 perform polymorphic inline caching. Many execution paths are inlined as machine code -- java won't be any faster for these execution paths.

Can you add and remove slots from a closure? Names in closures are bound at the time that they are created. There is no instance_eval in Common Lisp.

However, I am pretty sure that you could build such a beast in CL, as well as "closure"s with addable/deletable slots.

Yes you can build it in C as well, that doesn't make C dynamic.

Reply on news.ycombinator.com