Skip navigation

Most voters in the just-completed Java.net poll on Java EE 7 believe that Java EE 7 will be the most widely used version of Java EE within the next few years. A total of 354 votes were cast in the poll, which ran for three weeks.

The exact question and results were:

How long will it be before Java EE 7 is the most widely used Java EE version?

  • 14% (51 votes) - Less than 6 months
  • 18% (65 votes) - About a year
  • 32% (112 votes) - 2 or 3 years
  • 16% (58 votes) - Up to 5 years
  • 8% (30 votes) - More than 5 years
  • 7% (26 votes) - It will never happen
  • 3% (12 votes) - Other

To me, this poll both provides an overall community opinion (but, as I am always bound to say: this is not a scientific poll), while also providing some opportunities to speculate on what some voters might be thinking.

On the one hand, about a third of voters think Java EE 7 will be the most widely used edition with just one year. Now, that would surprise me, but it's good to know that many developers see Java EE 7 as providing sufficient impact that the companies they work for will quickly adopt it. Of course, you'd think that almost all of these voters are currently working with (or believe most others are working with) Java EE 6.

Add in the "2 or 3 years" votes, and you've got about 2/3 of the voters believing that Java EE 7 will be the predominant Java EE version a few years from now. This, to me, sounds reasonable, and likely. Java EE 7 is an important release. But, enterprise software vendors cannot afford to release new versions without thoroughly vetting the releases, which takes time. And, upgrading to a new version of Java EE takes time (which is why so many vendors are still developing their applications using Java EE 5).

For this reason, it's easy to agree with the 24% who think it will take "Up to 5 years" or "More than 5 years" before Java EE 7 becomes the most widely used edition. These people are very likely working on apps built on Java EE 5 or even Java EE 4.

Still, adding all these numbers up, this poll shows 88% of voters believing that, eventually, Java EE 7 will become the most widely used version of Java EE.

Where I start wondering is in the voters who selected "It will never happen." Why do they think Java EE 7 will never be the most widely used Java EE version? Do they think Java EE 7 is flawed in some fatal way? Or, do they think Java EE is destined for the dustbin of history? I wish some of these people had posted a comment!

And 3% chose "Other." I actually thought I'd covered all logically possible answers to the question, this time. I mean: "How long will it be before Java EE 7 is the most widely used Java EE version?" And my response options ranged from "Less than 6 months" to "More than 5 years" and "It will never happen"... Are their other possible answers to the question??? Yet, no one left a comment explaining their vote for "Other"...

Oh well... I think this poll produced an interesting result -- which is my objective in designing the polls. As always, I welcome suggestions for polls. Surely you have an idea for a poll, right? Surely there's something you'd like to ask the Java community! If so, contact me with your idea at editor -@- java.net. Do volunteer to help - I love working with others who have fresh ideas about potential polls!

New poll: So, this "cloud" thing - does it matter?

Our current poll asks Does 'the Cloud' change anything?. Voting will be open until Friday, July 12.


Subscriptions and Archives: You can subscribe to this blog using the java.net Editor's Blog Feed. You can also subscribe to the Java Today RSS feedand the java.net blogs feed.

-- Kevin Farnham (@kevin_farnham)

Last week, I attended the global live launch of Java EE 7, a webcast I wrote about earlier, titled "Introducing Java EE 7." The live webcast happened twice (on at 17:00 GMT on June 12, and 12 hours later at 05:00 GMT on June 13), thus enabling every Java developer in the world to attend the live session at some point during non-sleeping hours.

The value of attending the live sessions was the capability to participate interactively, via the chat app that was part of the webcast user interface:

But even if you weren't able to attend one of the live webcasts, if you registered for the event, you can still revisit pretty much everything that happened in the live webcasts, including reading the chat logs:

Now in progress: Virtual Developer Day

As I finish writing this post (about 17:00 GMT on June 19), I'm listening in on the latest Java Virtual Developer Day, titled Taking Java EE, SE, and Embedded to the Edge. This four-hour event started with a keynote address that provided an overview of Oracle's overall Java strategy. Now the breakout sessions are underway. Like the Java EE 7 launch webcast, there will be two editions of the Virtual Developer Day, timed so that Java developers around the world have an opportunity to attend at least one of the sessions during non-sleeping hours. The second VDD will happen at 08:00 GMT on June 25.


Subscriptions and Archives: You can subscribe to this blog using the java.net Editor's Blog Feed. You can also subscribe to the Java Today RSS feedand the java.net blogs feed.

-- Kevin Farnham (@kevin_farnham)

In a not entirely surprising result, a considerable majority of voters in the most recently completed Java.net poll indicated that they expect Java and other JVM languages to still be widely used in new software development projects 15 years from now. A total of 364 votes were cast in the poll, which ran for two weeks.

The exact question and results were:

Will Java and other JVM languages still be widely used for new software development projects 15 years from now?

  • 85% (308 votes) - Yes
  • 15% (56 votes) - No

The key phrase in the poll was "for new software development projects." Clearly, Java's installed software base is so huge that there will be work for developers maintaining and upgrading that code for decades into the future, just as has happened with COBOL and the large base of scientific code written in Fortran. But, how much new code is written in COBOL or Fortran today?

So, the question was: does the community believe Java and JVM languages are headed toward a COBOL/Fortran-like state 15 years from now? The great majority of those who voted think not (the usual caveat applies: this is not a scientific poll).

A huge advantage Java has over languages like COBOL and Fortran is, in my opinion, the JVM. This permits the creation of new languages that utilize the very proven infrastructure that's at the basis of Java. So, as the world of hardware and programming needs evolves, JVM-based languages can adapt in ways that COBOL and Fortran could not. So, COBOL and Fortran remained, in a sense, stuck in their era (though, assuredly, we have Fortran 95, with its adaptations for multiprocessor hardware, etc.). The JVM, however, provides a platform that's well-tested and proven, but also flexible enough to permit development of new languages designed to meet specific emerging needs.

That's a big advantage -- one that I doubt was thought about long ago when the JVM concept was invented. Then, the JVM was conceived as a means to write one set of code that could run on any platform. You might say that it's a fortuitous side-effect of this design that brings us languages like Clojure, Groovy, and Scala today.

New poll: Java EE 7 usage forecast

Our current poll asks How long will it be before Java EE 7 is the most widely used Java EE version?. Voting will be open until Friday, June 21.


Subscriptions and Archives: You can subscribe to this blog using the java.net Editor's Blog Feed. You can also subscribe to the Java Today RSS feedand the java.net blogs feed.

-- Kevin Farnham (@kevin_farnham)

Filter Blog

By date: