Mike Gualtieri from Forrester says that "Java is a Dead-End for Enterprise App Development." Interesting premise, interesting reasonings, no real conclusion though.

Here're some core takeaway statements:

"Java has served its purpose, but now it is time to move forward."

His justification for this is wierd. He says Java bungled the presentation layer, which makes no sense - because Java isn't a single person or group. If he's going to blame someone, is he blaming Apache, for Struts 1? After all, Apache now has tapestry, struts 2, wicket, a JSF implementation, and probably others - is he saying that THIS is who bungled the presentation layer?

Or is he talking about JSF?

He says that there is a steady stream of new UI approaches reflecting "Java lack of leadership in the presentation layer," but to me, all of those new UI approaches are a reflection of Java's leadership in the presentation layer. Lack of leadership would be using whatever API microsoft crapped out last month.

It's mostly FUD, after that - "java's never been the only game in town," and he's right, Java's just the most successful one. Java's new boss is the same as the old boss. I like the Who but I don't think Oracle's going to really screw up Java enough to make it so it's no longer valid for enterprise development. Java's a 20 year old language based on C++, so what? Clojure is based on LISP, does it suck too?

His main takeaway is that app developers should create a three-year application development strategy and roadmap. Funny, that's something Forrester does, isn't it?

I hate those guys sometimes.

In my previous job, I received a lot of reports from Forrester. I think they are just nice statistics for making C-level people happy.

I just read the article. It really sucks. It criticizes a bunch of mixed things (technical and commercial issues for example) and proposes using tools as solution (mixing apples and oranges). Why in the world you should abandon Java in order to implement BPM?


It would be nice if it explained which kind of languages can't be implemented on the Java platfor:

"You can invent as many new programming languages as you want, but they must all be implementable in the underlying platform."


I would like to see a Forrester report which predicted the raising of Spring in 2004, the adoption of JSON protocol, etc. I think they doesn't have a clue of the future of technology and I'm happy for not being forced to read its reports anymore.

I would like to see a Forrester report which predicted the raising of Spring in 2004, the adoption of JSON protocol, etc. I think they doesn't have a clue of the future of technology and I'm happy for not being forced to read its reports anymore.

It's a popular trend to throw stones, and Forrester once again pitches sensationalism instead of research.

Nonetheless, if you look at the lone supporting graph in the article, Java has <b>advanced</b> significantly over the past three years in terms of companies using it for custom-developed applications.

Peace,

Cameron Purdy | Oracle Coherence

http://coherence.oracle.com/

It's a popular trend to throw stones, and Forrester once again pitches sensationalism instead of research.

+1 and try to name a serious contender to replace it.

Cameron, Just because something is widely used doesn't mean it is the best way forward. I am really surprised you do not know that. Data always tells us the past and seldom predicts the future. Its a good thing everyone doesn't think the way you do about adoption data otherwise noone would have adopted Coherence.

Data always tells us the past and seldom predicts the future.

What a great slogan for a research company.

Javalobby used to do this a lot.   Point to some random blogger praising RoR and see how many Java developers get their panties in a bunch.

This is not about adoption statistics. His point is about java breaking the "productivity" linear path (2gl->3gl->4gl) and making application development complex, using the web app development as a vehicle.

Java took us several years back from visual basic & power builder "productivity" age. Hence, "productivity" tools might stage a come back, using cloud as a vehicle.

Disclosure: I belong to Orangescape PaaS - a visual cloud app development platform.

thanks,

mani

Yes, but you're not talking about Java, you're talking about web development vs desktop development.

I agree wholeheartedly that productivity has nose-dived due to the web (I used to code using Delphi). Java is not web development, however. The problem you are talking about is the web platform and the difficulty in abstracting GUI-creation from the web mess that is HTML, CSS, JavaScript and Ajax.

I'd bet he has no real-life JEE experience whatsoever. What a load of bulls**t.

Wojciech, Thanks for asking about my Java experience. I have developed dozens of Java applications over the past 10+ years including travel reservation systems, banking front-ends, and eCommerce systems, auction billing systems, call center apps, and many others. I have used many of the most popular frameworks that we all know.

So we must have been living in different universes, or maybe you're one of those my-ruby-on-rails-can-do-anything-for-the-enterpise-so-you-dont-need-java-any-more evangelists?

Give me examples of platforms that match java in areas such as popularity, simplicity, scalability, performance, number of free, open-source enterprise-class libraries, engines, interoperability capabilities or choice of different runtimes (ie. appservers, containers).

The fact that evolution of the Java platform has stagnated recently does not make all that is around it obsolete. Don't you agree?

I've been working on Java & C# and ocassionally using Oracle BPEL/BPM. First of all I don't find any significant reason why C# would be a replacement for Java or fora that matter any other language. And BPM replacing Java is the most insensible statement I had heard recently. This guy doesnt know java or BPM!

The fact about BPM is, it's just a concept or tool to deceive corporates. I've just completed a a bull crap BPM project, where the IT director of this company was saying , we are adopting BPM becasue in a couple of months we'd be delivering software's just like how "Sausage machines" does. After a couple of months the guy who made that statement was fired for  wasting a million £££'s.

Cheers

Aneesh

oh dear

So accoring to him we should all focus on tools such as Microsoft Lightswitch ,WaveMaker and traditional, but updated 4GL tools such as Compuware Unifaceand Progress OpenEdge

Jelmer, I am not recommending any of those tools per se. But, I am saying that there is too much complexity and a cacophony of frameworks necessary to develop Java business applications. The tools I mentioned attempt to abstract away some of the plumbing to allow the developer to focus on what is really important: getting the requires and user experience perfect.

So, you're saying that having a wider choice of solutions is worse. Interesting.

Absolutely! But the point could've been made without this title.

The problem is also the rapid pace of technology (web/HTML -> mobile/WML -> AJAX -> smartphones), which does not lend abstractions to mature. So developers have to fall back to hard core programming to catch the next wave instead of waiting for productivity tools to evolve.

BTW, i missed an other important point: developers love coding. They look for the first opportunity for the productivity tools to fail to go back to coding. While they love reporting back a bug on a coding based abstraction. especially when it is transparent i.e. an open source like spring or hibernate.

Bottom line, developers hate anything that makes them dump. Fair. isnt?

Sorry! That should have been dumb. (not dump) I was dumb :-(

> BTW, i missed an other important point: developers love coding. They look for the first opportunity for the productivity tools to fail to go back to coding.

Definitely! I know I do. I think it may become true, in the future, that for productive development of standard enterprise software, 4GL tools will become the best solution (I don't think it's a foregone conclusion, however). In which case, it will be time for me to move on to greener pastures. And let's face it, enterprise software is not rocket science or very interesting in 80% of the cases.

Mike, I would given you five stars if you just wrote "Java is dead, becasue I don't like Java" Or probably "The Guy used to Sit next to me always use to code java better than me".

Can you please tell me how on earth do you think Java is coplex in writing business Logic? Could you explain the what business and what logic bother you in terms of Java?

This artlcle to me is more like a a Biochemist giving an expert article about the economic porblems in Europe!

Best Regards

Aneesh

Forrester will say anything as long as someone pays them. They really have zero technical credibility with anyone that actually understands technology.

Peter, I have 3 software patents (you can look it up), have developed dozens of Java applications and know several computer languages. Saying that I have no technical expertise is an easy way for you to dismiss the arguments that Java is not the end-all be-all for business application development.

Peter, I have 3 software patents (you can look it up), have developed dozens of Java applications and know several computer languages. Saying that I have no technical expertise is an easy way for you to dismiss the arguments that Java is not the end-all be-all for business application development.

Mike, that is not what he said.  He was not questioning YOUR experience or expertise or ability to submit patents. He basically said that things that come from Forrester can't be trusted or that they should be taken with a grain of salt because of who pays them and that those with technical expertise/experience can and will do this.  It is not about you. 

That being said ... Having patents and having coded for years and in multiple languanges does not equate to expertise.  I don't know you so I cannot say what your actual expertise is.  But I do know of people in similar situations ... and they have years of experience but not "expertise" let alone ability.  And your "article" does really make me wonder.  My 15+ years does not come the same conclusion. Of course i have zero patents.

On the other hand, maybe your article is just poorly titled and worded and your true point is being obfuscated.  Can Java do things to be more productive? Sure. Can we learn something from 4GL's? Sure. Is Java a Dead-End for Enterprise App Development? No.

We all have to put food on the table - but we also do plenty of pro bono work too. Take a look at my keynote a Linuxcon this year. Ask yourself this - if mike was engaged in pay for play, wouldn't it make sense for him to shut up and toe the line of the big Java vendors? Let's reinject some analytics into the debate please.

We all have to put food on the table - but we also do plenty of pro bono work too. Take a look at my keynote a Linuxcon this year. Ask yourself this - if mike was engaged in pay for play, wouldn't it make sense for him to shut up and toe the line of the big Java vendors? Let's reinject some analytics into the debate please.

I understand we all have to put food on the table. Each person chooses how they do that. Perhaps a less sensationalistic approach would result in a "more balanced" critique. Saying Java is a "dead-end" just doesn't come across as well reasoned critique in my book, but that's my own bias. Then again, a toned down paper wouldn't get as much attention as something juicy.

We all have to put food on the table - but we also do plenty of pro bono work too. Take a look at my keynote a Linuxcon this year. Ask yourself this - if mike was engaged in pay for play, wouldn't it make sense for him to shut up and toe the line of the big Java vendors? Let's reinject some analytics into the debate please.

I just re-read the article again. Honestly, higher level abstraction isn't the solution for everything and a "higher level language' isn't a general purpose solution either. I've been focusing on business rule technology for over 7 years now. Things like BRMS, BPM, workflow, coreography, production rule engines, dsl and BPMN. I've looked at 4GL in the past, but never implemented anything with it. I don't know if it works in practice, but from my experience with business rule technology of various forms, it's not for everyone.

I agree that Java has warts and needs to work on making things "more productive", but that's different than being a "dead-end".

Peter, I have 3 software patents (you can look it up), have developed dozens of Java applications and know several computer languages. Saying that I have no technical expertise is an easy way for you to dismiss the arguments that Java is not the end-all be-all for business application development.

Mike, that is not what he said.  He was not questioning YOUR experience or expertise or ability to submit patents. He basically said that things that come from Forrester can't be trusted or that they should be taken with a grain of salt because of who pays them and that those with technical expertise/experience can and will do this.  It is not about you. 

That being said ... Having patents and having coded for years and in multiple languanges does not equate to expertise.  I don't know you so I cannot say what your actual expertise is.  But I do know of people in similar situations ... and they have years of experience but not "expertise" let alone ability.  And your "article" does really make me wonder.  My 15+ years does not come the same conclusion. Of course i have zero patents.

On the other hand, maybe your article is just poorly titled and worded and your true point is being obfuscated.  Can Java do things to be more productive? Sure. Can we learn something from 4GL's? Sure. Is Java a Dead-End for Enterprise App Development? No.

That is exactly what I meant. Credibility isn't the same as ability. I for one don't trust anything forrester or gartner says, without evualating a piece of technology for myself. I'm glad that mike has 3 patents to his name. That doesn't make me blindly give mike credibility in my book. Credibility has to be earned.

...for you to dismiss the arguments that Java is not the end-all be-all for business application development.

Obviously it is not, but that is far from your statement; that java is a dead end. It seems you are getting a grip now, understanding that java is a contender in the race, and not at a dead end.

... You cross a snake oil salesman and a false prophet?

It seems the author does not know anything about software development at all. Just lots of bull***t!

In the chart he compares Java, with Spring+Hibenate, the same way he compares Java with .Net!!

Nice Try!!

I think Mike Gualtieri does not have technical background neither competence to do a very good research and present some usefull findings to the community. So, what is the best wat to get the spotlights? Just publish something totally wrong :-). I think he wants to do something like Nicholas G. Carr with "IT doesn't matter".

Renato Guimarães.

Forrester should just outsource their research to Redmonk.

James, Stephen and Michael are excellent analysts and you should follow their research. I think if you talked to them I hope you'd find that they share the same opinion of their colleagues at Forrester.

It's been dying for 30+ years. NT is taking over, anytime now!

I would not dare to predict the death of Java by showing a graph where Java-based technologies (including both JEE and open source frameworks like Hibernate or Spring) sum up to an impressive and growing adoption rate of 71%...

So after reading the article I actually went to look at some of the suggested products and saw tools where one could drag in a database connector on to a screen and then place UI elements on the screen and connect them to fields in the connector...

Haven't I seen that before?

Was it 10 years ago?

Or maybe 20?

Technology that old must surely be dying?

We all know enterprise development is just simply displaying database fields to a screen.  These tools are garbage.  Nothing but super expensive, proprietary tools, which require massive custom development using the vendors proprietary XML like syntax, which only the vendors super expensive services team knows.

After you spend boatloads of money, the applications wont scale or perform well, and they you will bring in a consulting company to "fix" the application by rewriting from scratch in Java.  In the end the project will cost 20x what it would have cost had you just used industry standard enterprise tools and languages which do not lock you into a single vendors proprietary tech.

I understand that the title is provocative. So some these reactions are natural.

But seriously, isnt there anyone else, who thinks visual basic was easier than java for an average developer i.e. for the vast majority?

Visual Basic felt so much easier because the problems you were solving were so much easier.  When you have a fat client making database calls the only shared resource is the database.  You have a whole desktop to yourself for caching, writing files, etc...  Not to mention an entire CPU and memory to abuse as you see fit.

Contrast that to the web world where you have strict restrictions on the client, everything is shared on the web server and database servers which have to support multiple simultaneous users.  You can not just abuse the CPU, memory and disk if you want your application to scale out.

I agree, that there is some impedance mismatch - at the presentation layer. (But not because of scaling, caching stuff). HTML was meant for document presentation. But it was used as application presentation layer (which it was not meant for), which complicated app development.

Now that we are in the HTML5/RIA world, do you think it is possible to simply application development ? Ofcourse, not just at the presentation layer but also at business logic, business process, integration layers (ORMs did a good job already at datamodel - better than VB)?

Its about time we started to have a rating system we could use for analysts like this - Mike Gualtieri would get zero out of 5 for producing clue free thinly veiled marketing fluff.

Rob Davies

FuseSource

Now that I've read the article I have to ask.... seriously Forester, how can an article that contains the statement "Clear standard alternatives to Java and C# for custom-developed applications do not exist" be named Java is Dead end for Enterprise? And then there is this brilliant conclusion... "Application development teams should create a 3-year application development strategy". WIth no clear alternative to Java and C#, what could be in the plan?

Not really an analysis but more like fodder for a flamewar.. Are you really that desperate for publicity?

Kirk

Started to read artice but hoestly was quite hard to force myself to read everything. There are no strong and grounded base in his statements, just superficial and provoking ones. It's non sense to predict Java's death.

car rental

Thanks for all of your feedback and discussion on my post http://tinyurl.com/2fl69zs Prior to becoming an analyst at Forrester, I designed and developed Java applications (and some .NET ones too) for many large firms and some startups. As someone who made his bones developing operating system and scientific apps in the 80's I can appreciate Java. And, as I say in my post it served its purpose. But, we need to find a better way. The data in my post shows that Java is widely chosed for custom-developed applications. I put that in there to show that I am not saying that Java is dead. On the contrary it is thriving. But, I do think it is a dead-end because firms that hold on to it will be less cometitive than firms that move on. 

so rather than discuss the issues people are pointing out regarding the fallacies of your report, you reply with more uninformed market speak?  How are firms who use Java going to fall behind?  Give some specifics?

If you don't, you confirmed everyones points about your report being useless pixels.

1. How do firms that invest in Java only go mobile? How do the use it on all the mobile devices that are coming out?

2. How do firms that want to modify the JVM to make it run better in a cloud based environment do so without getting sued by Oracle?

3. How do developers get apps into Apps store that won't support the JVM?

I'd appreciate any insight you can provide. 

1. How do firms that invest in Java only go mobile? How do the use it on all the mobile devices that are coming out??

Very simple, the web browser.  There is no sdk which works on "all the mobile devices that are coming out".  How does this relate to java being irrelevant?

2. How do firms that want to modify the JVM to make it run better in a cloud based environment do so without getting sued by Oracle?

So java is a dead end because you can not modify the JVM?  You lost me on that one.  If Java does not perform for your specific needs, do not use it.  There is no development paradigm which is a silver bullet.  No matter what these tool vendors say when they suck away all your money.

3. How do developers get apps into Apps store that won't support the JVM?

Again, no idea how Java not being the sdk for every device out there from phones, mp3 players, televisions, video game consoles, etc... means it is a dead end.  What language runs in every app store? 

You offered nothing which explains what java is a dead end for enterprise development.  By your line of questioning, every language out there is a dead end because there never will be a single language that does everything for everybody. 

I supposed you are going to say some super expensive, bloated, proprietary, tool can magically make a program run on all devices with all their different specs?  LOL.  I have some ocean front property in Afghanistan to sell you...real cheap...only $10 billion USD.

Not seeing a lot of drivers other than Java sucks eight ways.  Usually something needs to drive this change.  Picking Java as a focal point seems to be missing the bigger picture of devices, clouds, and continued monetization of technologies and channels as drivers. Offering up the business wants to keep slapping the Staples "easy button" just doesn't justify the argument. 

Specially when you state "Swing is a nightmare" and "Hibernate, Spring, Struts, and other frameworks reveal Java’s deficiencies rather than its strengths"

I worked with java the last 10 years, I consider myself as an expert in Swing, Hibernate & Spring, and even when I fully agree with you that sometimes these frameworks & technologies can be a nightmare, I also can tell you that 100% of the times the issues are because self called "expert" programmers doesn't really know the core of the technologies and the companies like to contract cheap programmers to do complex things letting them to "learn" during work.

So, yes it can be a real nightmare, as well as create a simple "Hello world" by someone who knows nothing about java.

In (my) IT maturity cycle, I don't think you could get to highly productive Enterprise app development tool without a strong, mature, low-level language and platform to build on.  Things were going that way with client/server until the internet changed the dynamic, like client/server changed green screen, like green...

Now that TCP/IP driven systems (when you get down to it) and open standards have matured, we can reinvent VisualBasic again. Spring is doing it for VmWare cloud, Google provides IDEs for GAE cloud, Redhat, MS, yadda yadda. Problem is, Java may not do so well in non-enterprise solutions against vendors that want to control their delivery channels through proprietary devices and cloud platforms.

When I think of Ent. app development, I think license costs, seats, and lock in.  With the availability of mature, open source frameworks and accessible cloud platforms (no Google, GAE isn't portable enough) I just can't see trading perceived short-term productivity for long-term lock-in.  Look at SalesForce - they have a new word for everything.  

"not dead yet" and Holy Grail seem to go together. And Java is far from dead and IT will always chase after silver bullets and non-technical Holy Grail solutions but never quite get there.

Enjoyable discussion, but really, there's nothing much to talk about in that article.

The penny dropped some time ago that domain specific languages offer a faster route to getting things done than a generic, all purprose language like Java. And hence the popularity of newer, sexier languages like Scala, Ruby, etc... What we take from I don't know how many years of Java (14? 15?) is really the Java Virtual Machine, and the multiple languages and technologies it can support. I really derive great benefit from the ability to use the JVM *platform* wherever I like, but choose the *language* I prefer, while always having nuts-and-bolts Java to fall back on when I need it. For example, I'm using Camel right now to do cool integration flows with very little code, but whenever I need to do something fiddly I can pull in a little of of Java to save the day.

If you think about it, Java's biggest contribution is that it has become platform that manages to cohabitate  innovation in languages and frameworks for those who love to play, while also providing a pretty solid (if, perhaps a little dull these days) programming language that almost anyone can use to get stuff done.

So I don't think that Java's a dead-end from a technology perspective. Maybe the real problem for Java is the uncertainty around it's future under it's new owner; if Java is made open - and really open - then I think the 'dead-end' argument will evaporate for me. But if we all get scared to use the Java or the JVM in case we get smacked by Some Big Company then sure, Java will be dead in the water before long.

Agree! JVM is not going anywhere. Most of these productivity tools might infact be using JVM. IMO, what he probably meant is the decline of usage of java - a low level language for enterprise app development. Not JVM.

These new abstractions might run on JVM and java can always be a fallback. Hopefully, without the fallacies of the code generation tools - including MDA - a fiasco in the name of productivity.

If anyone on this platform has experience using a platform like TIBCO BuisnessWorks they will appreciate the fact that core Java development will take a back seat in many Enterprise scenarios. Having said that, it is important to realize that even platforms like BW are built using Java - so Java is not dead neither is it anywhere close - what is shifting is where and how it will be used going forward.

We use TIBCO BW and have mixed results.  If used purely for application integration it works really well.  When it started to be used as a big hammer to develop full blown applications with complex business logic, it really broke down.  One of the biggest issues was the tool was being used by people who did not have any distributed development experience.  The tool was being pushed because management was sold the bill of goods that people with little development experience (specifically distributed development) could quickly build applications.  The problem was (and is the case with any technology) that you can still design/architect poor solutions that are hard to test, debug, support, etc.  So the perceived productivity gains absolutlely killed teams in terms of reliability and support.

The problem I have with many of these "productivity tools" is that many of them do not provide things like: 

1) good impact analysis
2) easily isolate testing of business rules
3) The ability to mock behaviors in testing
4) robust packaging and deployment environments

I believe some of these tools have their place, but using them as big hammers to me is dangerous.  Too many times have I seen when there are issues with the tool (and you will run into them), and management throws a rediculous amount of money at a consultant from the vendor.  Alertnatively, we could have hired a couple really good java (or C#) developers for half the price to put together a robust and supportable solution.  Also, it seems like have to pay for every add-on and adapter in these types of tools (I know, that is how they make money and I would do the same).

Why does anyone listen to analysts, if they really understood software they would bein the development game instead of just talking about it.

if they really understood software they would bein the development game instead of just talking about it

Such emotional/irrational/hasty generalization are the ones that justifies the need for analysts. It makes your CXO doubt your objectivity in judgement. So can we get back to the point, please? 

Java is one of the advanced and good tool for App development.This is good information.

.<a href="http://www.fprinting.com/Full-Color-Sticker-Printing.htm">Full color sticker printing</a>

Is Mike Gualtieri shareholder of Microsoft? Or is he one of the many paid writers/bloggers of Microsoft? What gain does he get to sacrifice his name's integrity for that crap article?

Been hearing that "Java is dead" story from Java comptetitors espcially coming from the .net camp since year 2000. Now it's 2010 and Java is still alive and kicking.

Not defending him or anything, but to be honest, I think Mike Gualtieri is saying that C# is a dead end as well ;-) So this is not pro or against Microsoft, Oracle or whatever. If anything, his rant seems to be against generalist low-level languages like Java and C# -- and yes, in this context, Java is low-level.

Not defending him or anything, but to be honest, I think Mike Gualtieri is saying that C# is a dead end as well ;-) So this is not pro or against Microsoft, Oracle or whatever. If anything, his rant seems to be against generalist low-level languages like Java and C# -- and yes, in this context, Java is low-level.

Also, if you read his comments, he also includes Ruby, Groovy, Scala, etc in with Java and C#

Most of these languages built on JVM were quite low level. Because it was the developers who called the shots: http://manidoraisamy.blogspot.com/2010/11/java-revolt-of-working-class-against.html

Of course Java is a dead-end, but ultimately, EVERYTHING is. C, C++, Pascal, COBOL, etc etc etc etc - the software development road is littered with "dead ends" that served their needs very well until something better came along.

I don't share your conclusions that Java is dead NOW however. Honestly, I disagree with most of the conclusions in the article - "framework is a weakness", "bungled presentation" etc. These are not symptoms of a flawed architecture, they are indications of a vibrant community with an honest ambition for constant improvment. Your dream of some new, shiny paradigm that is the be-all-end-all platform/language/technology/whatever for all enterprise needs is just that, a dream. And one you can chase all the way to the end of the rainbow and you still won't find any pot of gold. New, interesting and extremly powerful languages and technoligies pop up all the time, but they all have a specific application where they are usable. There are no silver-bullets and 10 year down the road, we will be saying "Rails and Scala are the new COBOL".

That whole "Application development teams should create a three-year application development strategy and road map to include architecture, process, tools, and technology" is just bunk. Assuming that this is not already constantly being done by every half-way competent team is farcical. I submit that at the end of the day a great many teams come to the conclusion that given all the input parameters (legacy support, availablilty of skilled developers, performance, stability, robustness, maintainability, felxibility etc), Java is still very much a relevant way to go. Throwing in some buzzwords like "Cloud Computing" won't change this (the "cloud" is nowhere near as revolutionizing to software development as it might be to powerpoint presentations).

Also

Deputizing dinosaurs like IBM or TIBCO to demonstrate some new, innovative way forward is a bit comical...

1. He is just a puppet.
2. His nose is growing.

Is this guy working for Apple or Microsoft? Should we switch to C#, F# or another language that is coming down the road? Just count how many open-source projects written in Java? > 50%. Is it dead? NO!

http://blogs.forrester.com/mike_gualtieri/10-11-23-java_is_a_dead_end_for_enterprise_app_development#comment-8768

Mr Gualtieri may want to check his facts a little better in the future.  Some of his timelines and assertions don't seem very accurate at all.  His statement that java was designed in 1990 is misleading, and neglects to mention that the first release did not come out until 1995.  Also, in this 1995 timeframe, java did not have anything in the way of serverside web centric APIs and so forth available.  These came along later (servlets 1997) with the release of Servlets, and later on Java Server Pages (1999).

Microsoft's active server pages (asp) predated java centric web development by a year or two and was the predominant solution for dynamic sites early on.  This is the platform that dropped the ball, but hey they were tied to weak nt servers.

What does he mean with his incredibly biased statement that "Oracle is continuing with Sun’s same failed Java policies"?  Is he referring to the Java Community Process.  IMHO, the JCP has been and is an incredible success without parallel in programming language history with an amazing history of collaboration and platform improvements.  Anyone ever heard of a PowerBuilder or .Net Community Process?

How about his statement about the "Internet forced developer productivity and 4GL’s to take the back seat".  Has he ever tried any of the incredible programmer productivity tools and capabilities in Eclipse or Netbeans to name a couple, or some of the amazing things available with tools like seam? 

I'm a little suprised an organization like Forrester would allow such slanted uninformed drivel to be posted.

Yes, I already heard this before, Java is dead .net is here then Java is dead ROR is here. What else dude? Java guys are smart they always find ways to preserve the java integrity.  

Forrester analysts are required to get their reports mentioned in the press.  Part of their annual appraisal process is to count the number of mentions (I'm not just guessing this, I know it to be true).  The best way to get that number up is to say something outrageous.

I think the analyst has made a category error here.  The issue is not the Java programming language, nor the huge set of Java related tools.  The issue is web development.  It seems to me that web-based application development has set us back years in terms of application complexity and maintainability.  New tools, attempt to claw us back towards simpler, more maintainable software  -  but it is like the fly crawling along the elastic band:  for each 1cm the fly crawls, the far end is being stretched another 2cm.  In other words the demand for increased complexity in our web apps is advancing faster than our improvement in tools. 

BTW The fly does actually get to the end of the elastic band (assuming it's infinitely stretchable!) but it takes a very long time.

Enterprise App Development is not easy - to say that JAVA is a dead-end is not a very founded statement. Of course you have to have expert knowledge in many frameworks but even if you are not using Java, you have a load of fundamental IT concepts to keep in mind. (ACID, Reliability et al.) If you do this in an other language you will also have to think about lots of things and probably code quite a big lot yourself. And everybody who has coded a failover safe database connection pool knows what I am talking about.

that´s my 2 cent

Java is today the richest language, which allows fast and efficient production of software, mainly applications, I see nothing better around

Java is good for its pursosal.

It gives plenty of options like no other language ( and some of them lack).

The future will be on-top-of it or an evolution of it.

Lots of people understand and like it.

If one try to reinvent it, it will end up in a platforma with a bunch of new APIs, all similar to it ( RIA, MVC, PERSISTENCE, WEBSERVICES, BPM, and so on...).

Technology demands a mininal of knowledge and complexity ( simpler not simplest).

Don't let this publicity talk ofuse all the great work that has been done.

Gabriel