3 Replies - 12605 Views - Last Post: 22 September 2010 - 10:16 PM

#1 Programmist  Icon User is offline

  • CTO
  • member icon

Reputation: 252
  • View blog
  • Posts: 1,833
  • Joined: 02-January 06

The Next Big JVM Language

Post icon  Posted 22 September 2010 - 01:39 PM

I read Stephen Colebourne's blog post with a healthy dose of skepticism, but still found it very interesting. He makes some good points about the target "audience" for the Next Big JVM Language (NBJL). While I'm sure that many will snicker/sneer at the idea that we should be targeting the "least common denominator", don't forget that a language's adoption often determines its future use in the job market. If a language doesn't make it to the mainstream you can only partly blame the average developer for not wanting to get out of his/her comfort zone (After all, not all professional developers love it the way you and I do). The language itself can't escape some blame itself for just not being suitable enough. It's a shame because there are some great languages out there, but it's unlikely that most of them will be highly-sought-after in the job market in he near future.

Edit: BTW, the Fantom language they are talking about was formerly known as the Fan programming language. The name was changed because Googling "Fan" or "Fan Language" did not give good results. I prefer the new name because a language named "Fan" seemed pretty dumb to me. Why not just name it "Blender", "Toaster", or "Chair." :)

Is This A Good Question/Topic? 3
  • +

Replies To: The Next Big JVM Language

#2 Dogstopper  Icon User is offline

  • The Ninjaducky
  • member icon



Reputation: 2876
  • View blog
  • Posts: 11,050
  • Joined: 15-July 08

Re: The Next Big JVM Language

Posted 22 September 2010 - 01:50 PM

He made some good points, but something he said was just wrong:

Quote

2) Primitives and Arrays. Both these features expose low level details from bytecode. They break the "everything is an object" model. The lack of generics over primitives is a classic example of this. The correct solution is a language where the source code does not have exposed primitives or arrays, and the compiler (or perhaps the JVM) works out if it can optimise it for you.


Primitives are not objects as he says, however, Arrays are objects...
http://java.sun.com/...arrays.doc.html

I certainly think the JVM is a very powerful machine that will not die off anytime soon, even if Java might (which I also suspect won't happen). I am beginning to enjoy Clojure myself, so it is certainly interesting to see the side-by-side comparisons.


Thanks for sharing!
Was This Post Helpful? 0
  • +
  • -

#3 Raynes  Icon User is offline

  • D.I.C Lover
  • member icon

Reputation: 611
  • View blog
  • Posts: 2,815
  • Joined: 05-January 09

Re: The Next Big JVM Language

Posted 22 September 2010 - 04:03 PM

I read that yesterday. I think he discarded Clojure and Scala way to quickly and unfairly. Especially Clojure. It looks to me like he saw a piece of Clojure code with a couple of parens in it, screamed "OMG PARENTHESES" and decided it was irrelevant.

I hate these sort of blog posts in general. Nobody (seriously, *nobody*) knows what is going to happen in 5-10 years, and in my opinion, trying to play Nostradamus makes you look silly.

When I mentioned this on twitter, Brain Carper replied with this, which I think pretty much sums it up.

Quote

It's fun to make predictions or talk about things no one knows anything about, because no one can argue with you & you sound smart.


I think that if we focused more on the merits of these individual languages (Java, Clojure, Scala, Fantom, etc) and on their interop rather than on what will replace what, and what's the Next Big Thing, we'd be a lot happier and get a lot more done.

This post has been edited by Raynes: 22 September 2010 - 04:06 PM

Was This Post Helpful? 0
  • +
  • -

#4 Tom9729  Icon User is offline

  • Segmentation fault
  • member icon

Reputation: 180
  • View blog
  • Posts: 2,641
  • Joined: 30-December 07

Re: The Next Big JVM Language

Posted 22 September 2010 - 10:16 PM

Honestly I am very sleepy so I didn't read the whole article, perhaps I am just agreeing with the author.

I don't think we should have a radical shift from Java syntax, which is why I don't think Clojure/Scala will really catch on. Groovy is nice because Java is valid Groovy, but I'm not sold on the benefits of dynamic languages.

My biggest gripe with Java are the little inconsistencies in it's implementation. I don't like primitives vs wrapper classes in a language that is supposedly all about OOP. I don't like that String is a special case when it comes to operator-overloading, and I don't like that strings are immutable.

Arrays are weird, they are objects but they have properties (like length) instead of methods.

Closures aren't that important but would be nice to have. I don't have any complaints about generics. I agree 100% on having properties to replace getters/setters (which IMO are a messy nightmare, and the reason why so many people are lazy and just use public variables).

It might be a good idea to add operator overloading to the language. STL containers in C++ are very natural to use because of this.

</rambling>

On a side note it would be cool to have 3d graphics support ship with the JDK (preferably in the form of Java3d).
Was This Post Helpful? 0
  • +
  • -

Page 1 of 1