Scala - is it going to consign Java to legacy code

Tim Robbins

httpete wrote: »

I've started a new thread about regarding the presentation layer for Java EE web apps as I am writing one at the moment - http://www.boards.ie/vbulletin/showthread.php?p=82469910#post82469910

Regarding EJBs, I'm not sure what you mean when you say they are tied to your App Server.

1. Isn't the EJB spec the same no matter what App Server you use, and hence your code will be the same?
2. And even if not, how often would you be migrating an app to a different App Server in any case, I reckon it would be a very rare occurrence?
3. Is that the only reason to favour Spring over EJBs, as EJBs seems very simple to use, just annote the bean with @Stateless for example and inject it with @EJB.

(Note: If this question seems trivial it's because I am a PHP and Java SE programmer and I am writing my first Java EE application.)

To use EJBs you need an APP Server. You want o upgrade to a newer spec means you need to upgrade your App Server. Most companies won't want to do that.

Whereas when you want to upgrade Spring versions you don't have to worry about such matters.

And the way things are going now, everything should be stateless.
EJBs are dead and never really got going. They failed initially because the specs were awful, than they were beaten by Spring. Then when EJB3.0 copied a lot of Spring's ideas - Spring is going out of fashion.

Tim Robbins

conor.hogan.2 wrote: »

Anyone who calls Java dead I would ignore.

What other languages have you looked at? Don't mean that in a idiot way. But, what I meant by Java = dead is that it offers no architectural advantages to alternatives when it used to over major ones. So it is dead in the sense of been relegated from something that offered major architectural advantages to something that is just around because a lot of people know it.

C still offers architectural advantages. IT is not just around because people know it.

conor.hogan.2

Tim Robbins wrote: »

But, what I meant by Java = dead is that it offers no architectural advantages to alternatives when it used to over major ones.

So, you meant it was not dead. You know what dead means, and that is not it.

ChRoMe

conor.hogan.2 wrote: »

Anyone who calls Java dead I would ignore.

I think its the dumbest statement I've ever read in this forum.

conor.hogan.2

ChRoMe wrote: »

I think its the dumbest statement I've ever read in this forum.

It is up there.
But whatever your views on it, by any definition it is not dead.

Tim Robbins

conor.hogan.2 wrote: »

So, you meant it was not dead. You know what dead means, and that is not it.

Nothing in software is ever dead then. Sure EJB 1.0 will live forever too. And sure don't ATM's still use Windows 95? This is becoming silly.
Most of IT work in Ireland is second rate. So, yes there will always be Java work here but it will become less and less glamorous.

ChRoMe

Tim Robbins wrote: »

Nothing in software is ever dead then. Sure EJB 1.0 will live forever too. And sure don't ATM's still use Windows 95? This is becoming silly.
Most of IT work in Ireland is second rate. So, yes there will always be Java work here but it will become less and less glamorous.

What other countries have you written code in? I've done it in 3 different ones, currently writing this from London Bridge.

Its the same anywhere, that's the whole point of the skill.

Tim Robbins

ChRoMe wrote: »

What other countries have you written code in? I've done it in 3 different ones, currently writing this from London Bridge.

Its the same anywhere, that's the whole point of the skill.

The top stuff in US is way ahead. London a good bit further ahead than here.

I am not sure where this conversation is going. I am not interested in a competition.

ChRoMe

Tim Robbins wrote: »

The top stuff in US is way ahead. London a good bit further ahead than here.

I am not sure where this conversation is going. I am not interested in a competition.

Its just people trying to explain the fallacy you are suffering from.

srsly78

Google US are way better than Google Ireland dontcha know.

Tim Robbins

ChRoMe wrote: »

Its just people trying to explain the fallacy you are suffering from.

I have worked with Java since 1999. This is beginning to start like a flame war. Attack the post not the poster.

I am saying Java will remain around for legacy projects, it offers no major architectural advantages, most start ups or innovative new projects will look to move away from it (or they are making a big mistake using it). On the back end, it did nothing of any significance to leverage multi-core or cloud. Client side, it cannot offer the same benefits as JavaScript - which destroys it. Could you imagine gmail written in Struts or JSF?

The specs move way to slow (very little language features in last 7 years).
It has been either very slow to address its problems (EJB, JPA, Concurrency) or just never addressed it problems (J2EE architectures). It is now owned by a company that specialise not in languages but Databases. What have they done? Spent a fortune investing in ADF - you think that's any good? Even java.net has become a poor site when it used to be one of the best.

When it was written by James Gosling, he wanted something with simple garbage collection, had threading built in and was portable. He did a brilliant job but then other problems came along for which there are better solutions.

A Java top to bottom stack would be a really bad architecture now and pose a risk to an organisation. See Martin Fowler's thoughtworks who says the exact same thing.

There are occasional exceptions for example, Hadoop was written in Java.
And of course plenty of software projects which don't need superb architectures because they stuff they are doing is simple . But even in the make it simple space, there are better alternatives, Ruby / Rails, Groovy / Grails or even something like Force.com.

Technology changes and there will be some who don't want to move. It all comes down to return on investment. Once, companies can make more money doing things without Java they will move. Some are already there, others will go much slower.

ChRoMe

Tim Robbins wrote: »

Attack the post not the poster.

I don't think you know what the word fallacy means.....

In addition, the JVM supports multi core, however when people start using marketing terms like "cloud", it has no space in a technical discussion.

What amuses me the most..... the gmail backend is Java

Tim Robbins

ChRoMe wrote: »

I don't think you know what the word fallacy means.....

In addition, the JVM supports multi core, however when people start using marketing terms like "cloud", it has no space in a technical discussion.

What amuses me the most..... the gmail backend is Java

OK so we can't have a technical discussion and mention cloud then?
FFS?

Also, the JVM abstracts out multi-core. Saying it supports it doesn't really answer the question how well / unwell it makes use of it. When you are in Java you can't get a handle on a specific core.

I'd ask again to try to stop this turning into a flame war.

croo

Tim Robbins wrote: »

Also, the JVM abstracts out multi-core. Saying it supports it doesn't really answer the question how well / unwell it makes use of it. When you are in Java you can't get a handle on a specific core.

there are better alternatives, Ruby / Rails, Groovy / Grails

But since Groovy/Grails is also running in the JVM it would suffer the same?
I kinda look on groovy as part of the java stack anyway.

I still like the java stack though I agree that the combination of; first the years of uncertainty around SUN followed by Oracle ownership has not done the "official" java stack any favours but unofficially the like Spring & Ajax have done goods things and continued to move the stack forward. OK so technically ajax is javascript is not in the "java" stack but really javascript was always there when the application client was to be web based.

srsly78

Tim Robbins wrote: »

OK so we can't have a technical discussion and mention cloud then?

Not without a lot of sniggering no.

Tim Robbins

croo wrote: »

But since Groovy/Grails is also running in the JVM it would suffer the same?
I kinda look on groovy as part of the java stack anyway.

No the JVM is here to stay. It's class.

This is what I am saying the architectural advantages from Java (for example the portability of the JVM) are no longer major selling points because alternatives groovy, scala etc user them.

I still like the java stack though I agree that the combination of; first the years of uncertainty around SUN followed by Oracle ownership has not done the "official" java stack any favours but unofficially the like Spring & Ajax have done goods things and continued to move the stack forward. OK so technically ajax is javascript is not in the "java" stack but really javascript was always there when the application client was to be web based.

JavaScript / Ajax killed of the idea of thin client web frameworks.
It is possible to do stuff with JavaScript it is not possible to do with Java alone.

Spring on the other hand removes the need for a lot of boiler plate code and plugs gaps in J2EE. But, it is possible to do everything in Spring with Spring.

Tim Robbins

srsly78 wrote: »

Not without a lot of sniggering no.

So if a customer puts all their CRM data in the cloud and says give me your best are you just going to snigger as well?

srsly78

Yes. Putting stuff on a distributed network drive is hardly a new innovation. When talking to the customer tho you have to drop lots of buzzwords like cloud, just try really hard not to giggle.

Sparks

Tim Robbins wrote: »

So if a customer puts all their CRM data in the cloud and says give me your best are you just going to snigger as well?

Is the client a developer? I think the point he's making is that the inter-developer lexicon is very different from the developer-client lexicon, and "cloud" is the abused buzzword of the moment.

Looking back on the last few posts though guys...

[Mod]Keep it civil please folks[/Mod]

ChRoMe

Tim Robbins wrote: »

Spring on the other hand removes the need for a lot of boiler plate code and plugs gaps in J2EE. But, it is possible to do everything in Spring with Spring.

Your reasoning is all over the place, you appear to make no distinctions between frameworks and languages...

Tim Robbins

srsly78 wrote: »

Yes. Putting stuff on a distributed network drive is hardly a new innovation. When talking to the customer tho you have to drop lots of buzzwords like cloud, just try really hard not to giggle.

It is a new innovation in if you can develop in it and get much simpler scaling options - and everyone is being charged on per usage.

So you customer roles this way and you are asked to provide a technical solution to some business functionality? Do you still giggle?

Tim Robbins

ChRoMe wrote: »

Your reasoning is all over the place, you appear to make no distinctions between frameworks and languages...

Somebody else mentioned Spring.

If you read what was written, I was comparing what it meant for Java compared to Ajax / JavaScript.

Tim Robbins

Sparks wrote: »

Is the client a developer? I think the point he's making is that the inter-developer lexicon is very different from the developer-client lexicon, and "cloud" is the abused buzzword of the moment.

Looking back on the last few posts though guys...

[Mod]Keep it civil please folks[/Mod]

OF course it is abused.

But, the architectural meaning of the Cloud: rapid deployment and different charging mechanisms are very relevant.

For example, look at Salesforce and their force.com offering. Oracle could have given us an equivalent Java offering but gave us nothing.

It will means that CRM systems that say were deployed on WebLogic and integrating with Salesforce can now ditch WebLogic and just developed in force.com.

And when deployed in force.com, scaling is a lot easier.

conor.hogan.2

Tim Robbins wrote: »

J2EE.

JEE, if you are starting out for the last few years.

ChRoMe

Tim Robbins wrote: »

OF course it is abused.

But, the architectural meaning of the Cloud: rapid deployment and different charging mechanisms are very relevant.

For example, look at Salesforce and their force.com offering. Oracle could have given us an equivalent Java offering but gave us nothing.

It will means that CRM systems that say were deployed on WebLogic and integrating with Salesforce can now ditch WebLogic and just developed in force.com.

And when deployed in force.com, scaling is a lot easier.

So really what you are advocating is people deliberately tying themselves to a proprietary infrastructure?

srsly78

Yeah but it's proprietary CLOUD infrastructure so totally different.

ChRoMe

srsly78 wrote: »

Yeah but it's proprietary CLOUD infrastructure so totally different.

My apologies

Tim Robbins

conor.hogan.2 wrote: »

JEE, if you are starting out for the last few years.

Well yeah but a lot of industry still call it J2EE. Most of the core ideas are from that. By the time JEE came out, Spring had won the race.

Tim Robbins

ChRoMe wrote: »

So really what you are advocating is people deliberately tying themselves to a proprietary infrastructure?

In most cases no. It depends on the specific problem and architectural challenge.

croo

Tim Robbins wrote: »

Somebody else mentioned Spring.

If you read what was written, I was comparing what it meant for Java compared to Ajax / JavaScript.

Well I mentioned Spring because it more than anything has been pushing the Java Stack forward.

But I don't see how Java is competing with Ajax? I work an an app with a java backend, swing front end and a ZK front end.

JavaScript / Ajax killed of the idea of thin client web frameworks.

Like the ZK & GWT frameworks? Perhaps I just mis-undertand?

But, it is possible to do everything in Spring with Spring.

Is that a typo?

I only looked at Scala very briefly about 4-5 years ago. I could see for certain types of problems - I think someone mentioned earlier "programming problems as opposed to say business problems - that's kinda what I thought but it was a brief look. Perhaps you might point to something that Scala does that Java just cannot do - by that I mean a problem it can solve that java can't.

To me the attraction of java is not the language, but the huge eco system that's available around it. Things like (Spring) that took years to evolve. Not matter what problem (or business problem sat least because business software in the area I deal with) you have it's been addressed before!