"Adam Messinger, Oracle vice president of development, told QCon that Oracle plans to offer a 'premium' edition of the JDK in addition to the open-source JDK," as was reported by The Register's Gavin Clarke. 

So, this tends to generate more questions than in answers. What will be the value added by the premium version, and is this something the Java community should be worried about? In his November 6th, blog post, Stephen Colebourne asks the questions everyone in the Java community is likely asking themselves:

• Will the premium and gratis version be released at the same time?
• Will gratis version still support the same range of operating systems?
• What extra features will the premium version have?
• Is this only management features?
• Or are there performance features?
• Who is the target market?  

The details are currently pretty slim. And while nobody can blame developers for raising an eyebrow over the announcement, this certainly isn't a reason to worry that they sky might be falling. Many vendors have generated revenue by taking the vanilla JVM and charging for various performance, scaling and memory management enhancements that significantly improve the platform. The hope is that any premium JVM offerings won't put those using the gratis version at an unanticipated competitive disadvantage.

Stephen Colebourne's Weblog

http://www.theregister.co.uk/2010/11/06/oracle_dueling_jvms/

When it rains, it pours. Oracle keeps piling up the brilliant PR and I'm sure that when the Java brand is finally well and good destroyed they'll look around surprised and say "what the h*** happened?".  

I have no problem with Oracle wanting to make money from their investment and I even agree with them sueing Google. But their moves are like an elephant walking into a room full of China. They alienate developers and Java champions left and right and create doubt about the future of Java. And in this case, it's irrelevant whether a premium Java actually does mean that free Java gets cripled in some way. It's enough that it looks that way.

Yeah, it takes a little while to adjust to the change. Blablabla. Enough already with the amateur hour. Get it together, will you?!

When it rains, it pours. Oracle keeps piling up the brilliant PR and I'm sure that when the Java brand is finally well and good destroyed they'll look around surprised and say "what the h*** happened?".  

I have no problem with Oracle wanting to make money from their investment and I even agree with them sueing Google. But their moves are like an elephant walking into a room full of China. They alienate developers and Java champions left and right and create doubt about the future of Java. And in this case, it's irrelevant whether a premium Java actually does mean that free Java gets cripled in some way. It's enough that it looks that way.

Yeah, it takes a little while to adjust to the change. Blablabla. Enough already with the amateur hour. Get it together, will you?!

Agreed. I'm still holding my breadth hoping it doesn't completely fall apart for Java under Oracle.

I think now would be a good time for a SpringSource/VMWare JVM offering. :)

Agreed. I'm still holding my breadth hoping it doesn't completely fall apart for Java under Oracle.

I think now would be a good time for a SpringSource/VMWare JVM offering. :)

Red Hat is  actually already set up to deliver (and do on Fedora/RHEL) a JVM through the Iced Tea project.  I don't see any reason why we couldn't pick up the MacOSX slack too...

When it rains, it pours. Oracle keeps piling up the brilliant PR and I'm sure that when the Java brand is finally well and good destroyed they'll look around surprised and say "what the h*** happened?".  

I have no problem with Oracle wanting to make money from their investment and I even agree with them sueing Google. But their moves are like an elephant walking into a room full of China. They alienate developers and Java champions left and right and create doubt about the future of Java. And in this case, it's irrelevant whether a premium Java actually does mean that free Java gets cripled in some way. It's enough that it looks that way.

Yeah, it takes a little while to adjust to the change. Blablabla. Enough already with the amateur hour. Get it together, will you?!

Agreed. I'm still holding my breadth hoping it doesn't completely fall apart for Java under Oracle.

I think now would be a good time for a SpringSource/VMWare JVM offering. :)

I'm not an expert on the JVM, but one has to ask, does Spring have the resources and expertise to effectively develop and manage a jdk release?

Writing Java software is one thing. Maintaining a huge code base like OpenJDK isn't an easy task and would likely takes years for a team to gain sufficient experience and familiarity to do the job effectively. There's 15+ years of code in OpenJDK, so not "anyone" can pick up the task and hit the ground running. The only viable solution I see is if some company buys the developers from Oracle. short of that, it feels rather impractical.

When it rains, it pours. Oracle keeps piling up the brilliant PR and I'm sure that when the Java brand is finally well and good destroyed they'll look around surprised and say "what the h*** happened?".  

I have no problem with Oracle wanting to make money from their investment and I even agree with them sueing Google. But their moves are like an elephant walking into a room full of China. They alienate developers and Java champions left and right and create doubt about the future of Java. And in this case, it's irrelevant whether a premium Java actually does mean that free Java gets cripled in some way. It's enough that it looks that way.

Yeah, it takes a little while to adjust to the change. Blablabla. Enough already with the amateur hour. Get it together, will you?!

Agreed. I'm still holding my breadth hoping it doesn't completely fall apart for Java under Oracle.

I think now would be a good time for a SpringSource/VMWare JVM offering. :)

I'm not an expert on the JVM, but one has to ask, does Spring have the resources and expertise to effectively develop and manage a jdk release?

Writing Java software is one thing. Maintaining a huge code base like OpenJDK isn't an easy task and would likely takes years for a team to gain sufficient experience and familiarity to do the job effectively. There's 15+ years of code in OpenJDK, so not "anyone" can pick up the task and hit the ground running. The only viable solution I see is if some company buys the developers from Oracle. short of that, it feels rather impractical.

Agreed. The point was made as a joke with the smiley face and all. But you may never know.

I know we are both joking, but some people have suggested spring as a viable option.

Then again, you never know. VMware could make a deal and acquire the jdk team.

Writing Java software is one thing. Maintaining a huge code base like OpenJDK isn't an easy task and would likely takes years for a team to gain sufficient experience and familiarity to do the job effectively. There's 15+ years of code in OpenJDK, so not "anyone" can pick up the task and hit the ground running. The only viable solution I see is if some company buys the developers from Oracle. short of that, it feels rather impractical.

If you wanted to witer a JVM implementation 15 years ago, the only tools you had were ViM/emacs, make and C compiler.

A lot of things have changed since then. You can now grab LLVM (supported by a lively community) and you instantly get a good JIT compiler. You can check TraceMonkey JIT for dynamic optimizations blowing HotSpot out of water. You can check Apache Harmony and draw experience from PyPy VM.

So the task of writing a respectable JVM is now several magnitudes easier. A graduate student can do this in 1-2 years. In fact, there are already several JVM implmentations for LLVM. Making them production-quality is a bigger task, but certainly doable.

Writing Java software is one thing. Maintaining a huge code base like OpenJDK isn't an easy task and would likely takes years for a team to gain sufficient experience and familiarity to do the job effectively. There's 15+ years of code in OpenJDK, so not "anyone" can pick up the task and hit the ground running. The only viable solution I see is if some company buys the developers from Oracle. short of that, it feels rather impractical.

If you wanted to witer a JVM implementation 15 years ago, the only tools you had were ViM/emacs, make and C compiler.

A lot of things have changed since then. You can now grab LLVM (supported by a lively community) and you instantly get a good JIT compiler. You can check TraceMonkey JIT for dynamic optimizations blowing HotSpot out of water. You can check Apache Harmony and draw experience from PyPy VM.

So the task of writing a respectable JVM is now several magnitudes easier. A graduate student can do this in 1-2 years. In fact, there are already several JVM implmentations for LLVM. Making them production-quality is a bigger task, but certainly doable.

I agree that writing JVM has gotten easier over the last 2 decades, but I don't agree it's several orders of magnitude easier. TraceMonkey JIT is for javascript, so that type of JIT is quite different than JVM JIT, atleast from my limited understanding. LLVM is an interesting project and VMKit is doing good work, but getting something production quality takes a long time. I wouldn't be surprised if it takes 4-6 years. To achieve that takes a lot of money and fulltime developers, which often is the limiting factor for open source projects. I used to spend about 30-40 hours a week working on open source stuff, but that's very difficult to sustain over a long period of time. After about 5-6 years of that, developers that aren't paid to do open source tend to move on to other things.

Agree. Oracle is f**ing with the Java world.

I think it is time to move on. How knows..

I could be smoking pot, but if you wanted support from Sun for the jdk, you had to buy a support license. Of course you could always file a bug entry, but Sun explicitly stated customers had higher priority. From a support perspective, having 2 separate releases makes it simpler. The stack dump will tell the engineers exactly which version produced the error.

It would be great to have Oracle spell out exactly what the differences are to dispell any misunderstanding. If it's the same as Sun, it's just a much ado about nothing. If the non-supported version is missing significant API or functionality, that would definitely cause a problem for everyone.

Quoting Gavin Clarke (generally not a fan of Oracle, to say the least) who was in turn quoting a mis-typed tweet of someone watching a PowerPoint? Come on ..

Regarding having a "premium" edition of the JVM, that was announced in September. See: http://www.oracle.com/us/corporate/press/173782

Oracle can definitely do a better job communicating around Java, and we're trying to fix that, but in this case there has been communication, and it's publicly available.

Peace,

Cameron Purdy | Oracle Coherence

http://coherence.oracle.com/

There's no significant change from how Sun sold support for Java. I can understand people being fearful and having doubts, but all of this "sky is falling" is really counter productive.

I guess it drives traffic to serverside, so it's good for pageviews.

peter

Apart from what we don't need which is all the browbeating, going back on your word because suddenly you find yourself on the other side of the fence, naked greed and all the negative press, there is also something that we do need, and that is some real innovation. 

Right now, that means the mobile space. And it's just killing me that the words Java and Android are not uttered in one breath (except by those in the know). One area that could foster a whole new area of innovation is taking some of the UI development tooling on Android to the desktop and the web. AFAIC if there's one area where .NET shines and Java sucks is productivity towards the front end: just getting stuff done. Taking the Android UI as a basis for a new cross platform UI toolkit would give the Java UI the excitement it never had and that JavaFX will never ever produce.

This requires a close working relationship between Google and Oracle. Which includes taking care of this lawsuit in whichever way possible. Honestly, if you want innovation, you're talking Google, not Oracle, and although I understand the lawsuit from a legal point of view (I'm on of those many legal people with a J2SE certification), from a business point of view, this needs to be put to rest asap. Oracle should be thanking it's lucky stars for Google's innovactions on Java in the mobile space, take a more long term perspective and get in bed with them as soon as they can. 

Can't blame Oracle for wanting to monetize Java (and MySql) but somehow they have crossed a "product" tipping point and I can only expect less technical and more marketing innovations to come. For the first ~seven years Java was accessible - you could see the innovation exploding, the community exploding, and all under the benevolent support of Sun.  free+open+supported+quality was the total package and product. With Oracle chipping away at the first two aspects, Java seems like an old mule with a yoke on its neck that Oracle hopes can plow the back 40 before selling it for glue.

Google seems best positioned to pick up the Java "access with innovation" burden although it may not be Java.  Hey, if it's free for me and  sells more eyeballs on ads for Google, I'm in.

Can't blame Oracle for wanting to monetize Java (and MySql) but somehow they have crossed a "product" tipping point and I can only expect less technical and more marketing innovations to come. For the first ~seven years Java was accessible - you could see the innovation exploding, the community exploding, and all under the benevolent support of Sun.  free+open+supported+quality was the total package and product. With Oracle chipping away at the first two aspects, Java seems like an old mule with a yoke on its neck that Oracle hopes can plow the back 40 before selling it for glue.

Google seems best positioned to pick up the Java "access with innovation" burden although it may not be Java.  Hey, if it's free for me and  sells more eyeballs on ads for Google, I'm in.

Maybe if this Oracle lawsuit against Google doesn't work out for them (i.e. Google), then we might see Google putting their full weight behind powering up and using Python since as I understand it, they pretty much use it a lot next to Java.

http://www.tiobe.com/index.php/content/paperinfo/tpci/index.html

Andrew -

Can't blame Oracle for wanting to monetize Java (and MySql) but somehow they have crossed a "product" tipping point and I can only expect less technical and more marketing innovations to come.

Where on earth are you pulling this stuff from? Oracle has totally ratcheted up the engineering investment in Java since acquiring Sun Microsystems. That investment will be showing up in the next several releases, i.e. starting in Java 7 and Java 8.

Java remains free, open and quality. If you want support, there will be a price for that. (Sun called it "Java for Business"; it's not a new concept with Oracle.)

Peace,

Cameron Purdy | Oracle Coherence

http://coherence.oracle.com/

Andrew -

Can't blame Oracle for wanting to monetize Java (and MySql) but somehow they have crossed a "product" tipping point and I can only expect less technical and more marketing innovations to come.

Where on earth are you pulling this stuff from? Oracle has totally ratcheted up the engineering investment in Java since acquiring Sun Microsystems. That investment will be showing up in the next several releases, i.e. starting in Java 7 and Java 8.

Java remains free, open and quality. If you want support, there will be a price for that. (Sun called it "Java for Business"; it's not a new concept with Oracle.)

Peace,

Cameron Purdy | Oracle Coherence

http://coherence.oracle.com/

 

But Did Sun sell a different JVM for poors garage based developer vs. rich folks who by things to mostly use 10% of the features ?

As you mentioned there was communication in the past. Also we all understand ORacle has to make money out of java, otherwise why would Larry buy it in the first place.

BUT this is definitely goign to press panic button for some folks. Where is java going, do i need to download community version and then wait for next release to get bugs fixed ?  is it goign to mySQL route..... forking java is very likely. spring/apache/jboss - someone please help.

When sun was at the helm of it all i didnt have to worry about such a thing. From a engineering giant we are going to a marketing / perceptions giant.

Only product which is really good from oracle is the database software. it speaks for itself. Everything else is mostly useless aCQUIRED CR*P. ( BTW coherance isnt bad at all, but it got way too expensive after oracle bought it. I m happy with terracota ....for now:) ).

Once again Cameron - we are skeptical and please let us be. You can defed oracle as much as you want. You have been doing that for last few months. We will see what comes out of java 7 and java 8 and then we will see if oracle is worth the trust.

P.S. So when is Larry buying RIM(blackberry). The money machine of Larry has to generate millions in mobile market. Dont miss the boat(yacht in case of Larry).

Shawn -

We will see what comes out of java 7 and java 8 and then we will see if oracle is worth the trust.

Agreed. I'm personally expecting to see some announcements in the immediate future that will put some of the concerns that have been raised to rest.

Peace,

Cameron Purdy | Oracle Coherence

http://coherence.oracle.com/

Within a week? A month? A year?

Within a week? A month? A year?

Subscribers of the ?Premium JVM will be informed this week.  ;)

When you sue Google, and drive Apple to drop support for Java, and put two platforms at risk overnight, you have to admit the product has changed.  Java is now not predictable.  The intangibles have changed.  

I'm just a developer looking for a benevolent dictator. You work for Oracle, post frequently to TSS, stay on message, and dilute FUD when needed.  I get it.

Maybe time will tell.

Peace.backAtCha();

-andrew

Andrew -

When you sue Google, and drive Apple to drop support for Java

I can't comment on the lawsuit.

I have good reason to believe that the Apple story will have a happy ending, and that Mac OSX will remain a premier platform for Java.

I'm just a developer looking for a benevolent dictator. You work for Oracle, post frequently to TSS, stay on message, and dilute FUD when needed.  I get it.

Posting here isn't part of my job. ;-)

I care about the future of Java. I'm working to make sure that its future is bright. More importantly, a lot of people at Oracle (including many from Sun) are working to make sure that Java has a bright future. Now we need to make sure we don't lose the community that made Java so special in the history of our industry.

Peace,

Cameron Purdy | Oracle Coherence

http://coherence.oracle.com/

Oracle already has two flavours of JVM : Free (Hotspot) and Premium (JRockit) and they cannot afford to maintain two code lines. As per this news from Infoworld and this makes perfect business sense for Oracle to merge these code lines to make a single JVM and hence they will probably keep two versions of JVM - one free and one premium which has advanced functionalities such as Mission Control. So not to panic, I believe!

I smell fish in the air! Lot's of bad news against Java!  Oracle PR force, where are you? I see only one friendly face - Mr Purdy!

Debu Panda

http://www.starviewtechnology.com - The complete platform for Event Driven World

Too bad Java is not open standard and Oracle is in full control over its development adn execution.

Seems like we will not have othe choice but to leave Java the platform completly. It is a shame, since JVM is top notch, but now, Java is not more then MS offerings :(

Too bad Java is not open standard and Oracle is in full control over its development adn execution.

Seems like we will not have othe choice but to leave Java the platform completly. It is a shame, since JVM is top notch, but now, Java is not more then MS offerings :(

For the record, I don't work for Oracle. All of this non-sense about Sun vs. Oracle is silly to me. If Sun didn't go out of business in the first place, we wouldn't even be having these silly discussions. If IBM bought Sun, we'd likely see the same kinds of silly discussion. Being an open standard doesn't really mean anything in my mind. It's how the platform/language is managed day-in-day-out and year-to-year that makes the difference. There are plenty of useless open standards that do very little.

I understand some people have irrational fears of Oracle, IBM and Microsoft, that's normal for humans. At some point though, it's "nice" to get some sanity back and "try" to be a little bit objective. I'm just as guilty as anyone else of being irrational and emotional, but is there really a need to switch platforms?

On a daily basis I use both .Net and Java. There are things each platform is good at and suck at. No single platform will ever be the final solution. Only time will tell if Oracle does a good job of guiding java platform.

Too bad Java is not open standard and Oracle is in full control over its development adn execution.

Seems like we will not have othe choice but to leave Java the platform completly. It is a shame, since JVM is top notch, but now, Java is not more then MS offerings :(

For the record, I don't work for Oracle. All of this non-sense about Sun vs. Oracle is silly to me. If Sun didn't go out of business in the first place, we wouldn't even be having these silly discussions. If IBM bought Sun, we'd likely see the same kinds of silly discussion. Being an open standard doesn't really mean anything in my mind. It's how the platform/language is managed day-in-day-out and year-to-year that makes the difference. There are plenty of useless open standards that do very little.

I understand some people have irrational fears of Oracle, IBM and Microsoft, that's normal for humans. At some point though, it's "nice" to get some sanity back and "try" to be a little bit objective. I'm just as guilty as anyone else of being irrational and emotional, but is there really a need to switch platforms?

On a daily basis I use both .Net and Java. There are things each platform is good at and suck at. No single platform will ever be the final solution. Only time will tell if Oracle does a good job of guiding java platform.

You miss the point a bit. we are not talking about technically which platform is good or sucks. 

We are talking about controlling the language/platform and keeping community in the loop. 

MSFT whatever you say dioes a good job of maintaining MSDN, visual studio, IIS server etc.

Again - We are not comparing java to .Net ( as you tend to at the end of your post ) , we are talking about Oracle vs. Microsoft.

Too bad Java is not open standard and Oracle is in full control over its development adn execution.

Seems like we will not have othe choice but to leave Java the platform completly. It is a shame, since JVM is top notch, but now, Java is not more then MS offerings :(

For the record, I don't work for Oracle. All of this non-sense about Sun vs. Oracle is silly to me. If Sun didn't go out of business in the first place, we wouldn't even be having these silly discussions. If IBM bought Sun, we'd likely see the same kinds of silly discussion. Being an open standard doesn't really mean anything in my mind. It's how the platform/language is managed day-in-day-out and year-to-year that makes the difference. There are plenty of useless open standards that do very little.

I understand some people have irrational fears of Oracle, IBM and Microsoft, that's normal for humans. At some point though, it's "nice" to get some sanity back and "try" to be a little bit objective. I'm just as guilty as anyone else of being irrational and emotional, but is there really a need to switch platforms?

On a daily basis I use both .Net and Java. There are things each platform is good at and suck at. No single platform will ever be the final solution. Only time will tell if Oracle does a good job of guiding java platform.

You miss the point a bit. we are not talking about technically which platform is good or sucks. 

We are talking about controlling the language/platform and keeping community in the loop. 

MSFT whatever you say dioes a good job of maintaining MSDN, visual studio, IIS server etc.

Again - We are not comparing java to .Net ( as you tend to at the end of your post ) , we are talking about Oracle vs. Microsoft.

 

Clearly I wasn't very clear. I wasn't comparing just the technology. I speakin in terms of general guidence and stewardship of Java platform vs .Net platform. Take for example ORM technology. How many different ORM products has Microsoft come up with only to throw it away, leaving developers high and dry? I don't necessarily agree Microsoft has done a "good" job of maintaining .Net platform. I could nick pick about things, but over all I "feel" they've done a fair job. The same is true of Sun. There have been jsr's that were left dangling or abandoned for a long time in the JCP in the past.

Honestly, i don't see how Oracle, Sun, IBM or Microsoft differ that much in terms of stewardship of the products they sell. People just seem to have irrational fears about Oracle and feel they will ruin Java. That doesn't make their feeling invalid, but it doesn't necessarily match reality. I have irrational fears about microsoft, but it doesn't make it real either.

The only solution is the Apache Harmony project to replace and fork OpenJDK as a platform of choice for Java development and ask to IBM to protect the Apache Software Foundation with their massive patent stock as they did with Linux against SCO. Apache Harmony could be the solution to a free Java like platform. Google, IBM, SAP and everybody could win with this and let Oracle in the dust with their Java premium overpriced crap.

The only solution is the Apache Harmony project to replace and fork OpenJDK as a platform of choice for Java development and ask to IBM to protect the Apache Software Foundation with their massive patent stock as they did with Linux against SCO. Apache Harmony could be the solution to a free Java like platform. Google, IBM, SAP and everybody could win with this and let Oracle in the dust with their Java premium overpriced crap.

So let me get this straight. The same exact code that was owned by Sun is now crap because Oracle owns it? For the record, I'm an apache committer.

Not everything from Apache is beautiful or spotless. Apache developers are often volunteers, with families and real lives. Although apache developers do their best to support the projects, often we don't have the time or cycles to respond immediately. There's plenty of room for commercial support and open source.

The big issue is openJDK is GPL and apache officially prohibits GPL code. Go look at the license page on apache.org. Forking OpenJDK within apache is not an option.

The main question here, I think, should be what would Oracle gain by creating a premium JVM?

Well, personally I think they could gain quite a lot. What's in it for Oracle at the moment with regards to the massive amount of Open Source and competetive commercial products on the market, built on Java technologies, that Oracle are not getting a single dime for? The answer here is obviously nothing.

Oracle only realy needs the JVM for their own product range which is heavily dependent on the JVM. Possibly this is one of the reasons they couldn't afford to let it fall into anothers companies hands.

Now that they own it they are free to drive the direction of future enhancements specifically with the intent of improving their own product offerings and if that means nobody else gets the benifit of these enhancements or optimisations then perhaps thats OK by them.

Even if the rest of the world stopped using Java over night, would it realy matter that much to Oracle? I don't know the answer to this question but I suspect a few Oracle people might.

Steve

http://blogs.oracle.com/henrik/2010/11/oracles_jvm_strategy.html

http://blogs.oracle.com/henrik/2010/11/oracles_jvm_strategy.html

Now this is the type of communication I expect to see come out of Oracle on a consistent basis. Clear, to the point, and having little or no gaps to cause nerve wracking speculation.

We are mostly a technical community, probably about 90% developers to about 10% managers/CIOs, so the marketing jive doesn't mean much to us. :)

As one who is not inclined to applaud annoucements over results, I continue to await the tangibles.

Java Forever!

It is nice and clear.  Still, I'm ticked about Google and Apple platform uncertainty. And reading about the tiering of InnoDB support in MySql just seemed like a bad trend.

Java For...now!

It is nice and clear.  Still, I'm ticked about Google and Apple platform uncertainty. And reading about the tiering of InnoDB support in MySql just seemed like a bad trend.

Java For...now!

Not to get too off topic, but I actually saw the tiered pricing as a positive sign. To me it means Oracle is serious about pushing mysql forward and isn't going to just let it die slowly. To make sure mysql continues to grow and flourish, it takes money. To me charging more seems like a logical thing to do if you want make sure you can keep developing it and provide good support.

Sorry for staying off topic, but this blog was helpful to me, related to the MySQL announcement:

http://blogs.oracle.com/mysql/2010/11/get_the_facts_mysql_licensing_and_pricing.html

First IBM, now Apple have joined OpenJDK. Mac OSX will continue to support Java :-)

http://www.theserverside.com/news/thread.tss?thread_id=61297

Peace,

Cameron Purdy | Oracle Coherence

http://coherence.oracle.com/