Groovy Zone - Comments for "Groovy will replace the Java language as dominant language"

@SergeEver heard of SOA?

jalexoid — Fri, 15 Feb 2008 15:49:24 -0500

@Serge

Ever heard of SOA? SCA? Maybe Layered Architechture or MVC rings a bell?

I have a fairly large project in production witch has WEB UI in RoR and backend in Java... Works FINE. Frontend developers only have the interfaces and noone cares that it is considered ONE project, that we have 2 disticnt teams and use 2 languages.

So your "VERY bad idea" becomes a beautiful, productive team of specialists...

@SergeMan.... You are pi**ed

jalexoid — Fri, 15 Feb 2008 15:43:26 -0500

@Serge

Man.... You are pi**ed with Sun... Don't like Netbeans? Use Eclipse, use IntelliJ... Not like Sun owes to create th best IDE....

And the decisions are not lethargic, but conservative... This is what happens when there are a LOT of parties with vested interests and HUDGE investments.

@SergeIf you say "only a

jalexoid — Fri, 15 Feb 2008 15:38:56 -0500

@Serge

If you say "only a look"

You could have noticed that Scala integrates with Java. It is stated in the first paragraph on http://www.scala-lang.org/.

@Tim In my experience,

SergeBureau — Fri, 15 Feb 2008 15:19:12 -0500

@Tim

In my experience, mixing language in a project is a VERY bad idea.

The reason I put Groovy as an exception is that in fact Java is "included" in Groovy. So programming in Groovy does not impair the Java part.

All the other's cannot claim that. This is why I agree with the statement that Groovy looks more like a Java 2.0.

In general mixing languages in a project produces a mess, very difficult to debug and maintain, not forgetting difficult to read.

Overall, this has been an

goekenator — Fri, 15 Feb 2008 14:20:46 -0500

Overall, this has been an amazing discussion. I have a question - why do any of the languages have to defeat the other? If I am writing an application, might I not choose Java, Groovy, Scala, bash script, perl, or Ruby for different aspects of the same project?

I'm not sure why I need to be a Groovy programmer or a ____ (fill in the blank) programmer. Why am I not just a programmer and trying to resolve issue using the best approach I can find and understand?

I certainly seems that the Groovy folks are seeking peaceful co-existence with others.

It may seem to be hard but many people such as Martin Fowler and Joel Spolsky seem to believe in a multi-lingual future.

If Ruby has the best ATOM library but Groovy has the best GUI framework, wouldn't I possibly choose both Ruby and Groovy as the implementation for these different responsibilities?

"They keep cranking about

tug — Fri, 15 Feb 2008 12:56:26 -0500

"They keep cranking about 10x faster than Sun. Groovy has huge perf probs though...when a language is outperformed by Ruby, you know you have problems!"

I think you may have that back to front. I think it's an achievable goal for a dynamic language like Groovy to be less 10 *slower* than Java on current JVM implementations and have said so repeatedly on the Groovy list. I have never heard anybody say that any dynamic language on the JVM would be faster than Java it's just not a credible statement.

If you follow the Groovy list you will see that some progress is being made on improving performance but there are still substantial issues to be addressed.

In principle Groovy should be quite a bit faster than JRuby (because it's a better fit for the JVM not because the JRuby people are not competent, they are very competent) at the moment it isn't and that's because the Groovy run time system has a bunch of crud in it because it was developed along with the language whist JRuby and Jython had the advantage of starting with an existing language.

The Groovy team may or may not be able to evolve the existing runtime system into one with good performance. My view is that they will not be able to do this via evolution (i.e. they can make it faster but not dramatically faster). However they have already done better than I expected and I may well be wrong about this.

>As far as waiting for Sun's

philswenson — Fri, 15 Feb 2008 12:36:52 -0500

>As far as waiting for Sun's blessing, let's be real.

>I have defended Sun's decisions for years, but it is now incredibly slow to get anything done.

You have a point... Microsoft seems much faster and more agile than Sun when it comes to updating the platform/language. Perhaps all the ceremony and bueracracy of the JCP is to blame?

> Not in Java 7, so le'ts wait 3 years and maybe ...

Ya, and Java 7 still hasn't been speced out I think. So add 2 years from whenever they settle on the features.

> Groovy provides it, in spite of Sun's.

>Where is the Netbeans support for Groovy ? Are you still waiting for Sun ? I am done defending their decisions, there lethargic.

I like Groovy. Feature wise, it's great IMO. They keep cranking about 10x faster than Sun. Groovy has huge perf probs though...when a language is outperformed by Ruby, you know you have problems!

When you have millions of

SergeBureau — Fri, 15 Feb 2008 12:04:23 -0500

When you have millions of Java coders available, having a similar syntax is an overwhelming plus.

As far as waiting for Sun's blessing, let's be real.

I have defended Sun's decisions for years, but it is now incredibly slow to get anything done.

2 years between releases is much too long, where are closures? where is AOP support ? ...?

Not in Java 7, so le'ts wait 3 years and maybe ...

Sorry but I am fed up, it seems more to me that the success from Ruby and Groovy are forcing the hand of Sun, otherwise nothing would have happened.

So I do not want to loose Java's good points, but I also want more.

Groovy provides it, in spite of Sun's.

Where is the Netbeans support for Groovy ? Are you still waiting for Sun ? I am done defending their decisions, there lethargic.

Certainly Groovy took

tug — Fri, 15 Feb 2008 11:19:18 -0500

Certainly Groovy took elements from Ruby. However, Builders were taken by Ruby from Groovy (and the Groovy developers were very happy to give something back to Ruby).

"I and every other java

tug — Fri, 15 Feb 2008 11:15:43 -0500

"I and every other java developer I know have ZERO interest in Groovy. Scala is an obvious choice."

Perhaps you should widen your circle of acquaintances :)

Groovy and Scala are not really competitors. Scala is a high performance statically typed language. The target of Scala is Java not dynamic languages like Groovy/JRuby/Jython.

Saying that Scala is superior to Groovy makes about as much sense as saying that a bicycle is superior to a rowing boat. I.E. it's committing a category error.

> had a look at Scala, and

philswenson — Fri, 15 Feb 2008 11:14:50 -0500

> had a look at Scala, and it will remain only a look. I totally hate the syntax.

...and similar rants about ruby syntax.

I coded in nothing but Ruby for 8 months a couple years ago... at first I found the syntax a bit weird, but then learned to really like it (for the most part anyway). I think syntax should be driven by what makes sense, not legacy. People have the ability to learn new things, esp programmers who's jobs probably are in flux more than any other career.

Everyone keeps saying Groovy is so much more like Java.... but I have to say I find Groovy more like Ruby than Java.

The arrays and hashes are much closer than the java equiv. The closure syntax is very very close. Reg ex language syntax is close. Builders are very similar. Both have properties. I see a lot of influence in Groovy from Ruby.

Anyway, as to the original topic I think for a language to supplant Java on the JVM Sun will have to get behind it and proclaim it the successor.

@Ed I had a look at Scala,

SergeBureau — Fri, 15 Feb 2008 09:19:23 -0500

@Ed

I had a look at Scala, and it will remain only a look. I totally hate the syntax.

As far as static versus dynamic, it is the same debate as interpreted vs compiled, the advancement in optimizers make those not that relevant points.

So you can keep Scala.

Synthax might be a matter of taste, but integration with actual language is nuch more important, and this is where Groovy is impeccable.

edaj wrote:This post is

gr34134 — Fri, 15 Feb 2008 08:13:11 -0500

edaj wrote:

This post is UTTER GARBAGE. Groovy is a pathetically slow and ugly language.

If any langauge is worthy of succeeding Java it is Scala.

http://www.codecommit.com/blog/scala/scala-for-java-refugees-part-1

It is statically typed and offers the performance of Java with syntax more elegant than Ruby.

I and every other java developer I know have ZERO interest in Groovy. Scala is an obvious choice.

Although the post is extreme and I don't agree with the "replace" assertion, your above comments also fall into the garbage catagory. Groovy is currently the most popular alternative language for the JVM by a long way, so your friends are probably living in a hole.

The fact that you think Groovy is ugly is subjective and your opinion, as is your assertion that Scala is more elegant than Ruby. Many find Scala's syntax very cryptic, again their opinions are subjective, each to his own.

As for performance dynamic languages will always be slower than statically typed languages, but the Groovy team is working on optimisations now and hope to achieve C Python speeds.

This post is UTTER GARBAGE.

edaj — Fri, 15 Feb 2008 07:43:51 -0500

This post is UTTER GARBAGE. Groovy is a pathetically slow and ugly language.

If any langauge is worthy of succeeding Java it is Scala.

http://www.codecommit.com/blog/scala/scala-for-java-refugees-part-1

It is statically typed and offers the performance of Java with syntax more elegant than Ruby.

I and every other java developer I know have ZERO interest in Groovy. Scala is an obvious choice.

I'm surprised that nobody

jbulck — Thu, 14 Feb 2008 12:03:09 -0500

I'm surprised that nobody mentioned the "Java - next generation" scenario: Java2, Groovy and other influences will replace themselves...

Java is going through a tough period... there are plenty of good idea's, the community is as vital as ever, but the body is getting tired and doesn't support big changes anymore. The never ending discussions on closures indicate that we've learned our lesson well: instead of adding new features sloppily, it might be better to *not* introduce these features at all. At Javapolis, the BOF on Closures was next to the BOF on Groovy. Both talked about closures, but the atmosphere was very different...

Imho, I'd prefer that the smart guys start working on a new generation of Java. Java3 has been coined before. The Javaposse talks about it from time to time as well. Forget about backward compatibility. Use JDK 7 as a start. Fix whatever doesn't smell right. And most importantly, add the good ideas from Ruby, Groovy, Scala, etc in an uncompromised fashion. This could/should add a new dimension to Java - dare I to say: somehow similar to evolution from C to C++ ?....

True, to a large extend, the result might be very Groovy-like... it wouldn't be called Groovy, though. Just Java3 or Java** or Java-whatever. If Groovy wants to grow its acceptance dramatically, it needs to change its name anyway. A wanna-be enterprise standard called 'Groovy' is just too hard to sell.

I admire the work done by the guys behind JRuby, Groovy and Scala, but I don't think they are going to beat todays Java all by themselves. I hope that at least one of them participates in creating the new Java.