Ian is the Director of Marketing for the Eclipse Foundation and live in Ottawa, Canada. Ian is a DZone MVB and is not an employee of DZone and has posted 163 posts at DZone. You can read more from them at their website. View Full User Profile

My JavaOne Wish List

09.17.2010
| 6969 views |
  • submit to reddit

Next week I will be attending JavaOne, the first organized by Oracle, and I am really looking forward to it.  The last couple of JavaOne conferences were pretty lame in terms of key Java announcements.  This year will be Oracle’s chance to set the vision for the future of Java.

I have high expectations Oracle will do right for Java.  Oracle is professionally managed company that knows how to make money.  They are going to do things differenly from Sun but that will/can be goodness.  Although, I admit their lawsuit against Google is not a great signal to the community.

Next week, here is what I would like to hear from Oracle.  [Note: FWIW, these are my own views and don't represent any official view of the Eclipse Foundation.]

Stop JavaFX, please. I have never understood JavaFX.  Sun had no hope and so far no momentum for this technology.  Oracle should just politely say it was a bad idea and move on.   Use the resources to provide better technology for Java programmers.  Trust me, no one will miss it.

Announce Java 7 Plans. Java 7 has been this lost puppy with no home for a long time.  Publish the plan and keep to it.  Show the future, lead by example.  My vote would be to announce Plan B next week.

Make Apache Happy.   The long standing issue over the field of use restriction for Apache Harmony’s use of the TCK needs to be resolved.   Resolving this one issue would be a HUGE win with the open source community.

Promote OpenJDK.  There is a lot of great stuff going on in the OpenJDK community.  Captialize on it, tell people about, invest in it, promote the hell out of it.  You also need to fix the governance for it to really thrive.  Establish a community board and governance charter that makes it truly open.  Sun was never really open with OpenJDK but it was still succeeding.  Do better than Sun and make it open and great.

Merge Java ME and Java SE. Java in mobile needs to be re-energized.  Please don’t announce at JavaOne that their are billions of devices using Java ME and all is well.  All is not well.  Bring the two code bases together and make it easier for a wider group of developers to write applications for mobile and enterprise.  Regardless, shake things up for Java ME

That is it.  Solve these points with concrete solutions and Java will be on the path to a great future.

btw, curious to know what else Oracle needs to announce?

From http://ianskerrett.wordpress.com/2010/09/17/my-javaone-wish-list/

Published at DZone with permission of Ian Skerrett, author and DZone MVB.

(Note: Opinions expressed in this article and its replies are the opinions of their respective authors and not those of DZone, Inc.)

Tags:

Comments

Fabrizio Giudici replied on Fri, 2010/09/17 - 1:55pm

Trust me, no one will miss it.

I think we can't trust you, because there's a number of people who are using it. Bad way to start a wish list hoping that some technology, that you "don't understand" but others do, goes away. Furthermore, there's no logic in your wish: the times of Sun with no money in which working on a project meant stealing resources from another are over. If Oracle works on something, it's committed on it.

Manuel Jordan replied on Fri, 2010/09/17 - 4:37pm

>>Merge Java ME and Java SE 

How could be possible this? , both handle different hardware resources

Ryan Developer replied on Fri, 2010/09/17 - 4:43pm

I for one think dumping Java FX is a very bad idea and seriously doubt Oracle is even considering that.  Lets get JavaFX designer out the door, and a next generation Java ME that is very much like Java 6 and that includes Java FX.  Lets get it on iPhone, Android and BlackBerry.

Otengi Miloskov replied on Fri, 2010/09/17 - 5:29pm

Sun aficionado's as Ryan and Fabrizio will tell you no to dump JavaFX but the rest of Java developers will tell you same as you do, we dont need JavaFX, better Oracle should put effort on Java/Java2D/Swing and OpenJDK, JCP, jsr's and TCK isues, Java 7, etc. Apache Pivot is better bet than JavaFX really. JavaFX it is dead since day 1.

Bob Smith replied on Fri, 2010/09/17 - 6:03pm

JavaFX doesn't have the legacy baggage that Java does.  For instance, JavaFX has closures and Java doesn't.  JavaFX is designed for GUI programming.   Dumping JavaFX makes no sense.

Richard Osbaldeston replied on Sat, 2010/09/18 - 4:07pm

Drop JavaFX are you kidding? have you seen how many JavaFX events are being held? looking at the schedules it looks more like JavaFXOne than JavaOne. Reading Amy's comments http://amyfowlersblog.wordpress.com/2010/09/15/back-to-our-roots-for-javaone about how its a return to the core I did wonder since when was JavaFX part of the JDK? source? licence? (barring any MAJOR announcement on Monday that might change this of course ;) ). There's also many alternate JVM language sessions, is this the new core? the JVM not the language, not saying thats bad, just not a refocus on the core.

Internally things seem to be have been a bit of a mess post acquisition, how much of the delays to JDK7 are due to this? I'd like to see Oracle considerably more developer friendly and far less of this http://forums.sun.com/thread.jspa?threadID=5444316&start=15 (broken bugparade/forums/search) or lossy forums/java.net/OTN type mergers. Personally I think one of the major issues Java faces is attracting new developers, Java has a horrible reputation outside of the cliquey Java community. Perceived complexity (if not overwhelming complexity & choice) if a major part of this merging with 100s of Oracle forums isnt a great help.

The lack of much Swing content at JavaOne bothered me, esp. the official Swing team lead not being present to answer the thorny questions on whether the Java desktop is dead? over on java.net.. the conspiracy theorist in me wonders how much of the continued undermining of Swing (early death of its JDK7 JSRs etc..) is deliberate weening process to push developers towards JavaFX whether they like it or not. It has echos of henry ford in the javaone message - you can use whatever jvm language you like just so long as its ui is javafx.

In terms of merging JSE & J2ME I believe Jigsaw was the key part to that. By modularizing the Java platform they could then have several profiles. A server one, sans user interface stuff (java2d, awt, swing, javafx) but possibly including more jee apis, a slimline mobile one (again sans java2d,awt,swing but with javafx I suspect) and desktop profile with java2d,swing & javafx but sans jee.

Bob Smith replied on Sat, 2010/09/18 - 7:59pm

I imagine Sun gave up on Spring and started investing in JavaFX due to the backward compatibility concerns associated with the Java language and Spring itself.   This seems like the logical course of action, since JavaFX can evolve separately as a GUI language and isn't limited by language decisions made in the mid 90's as the Java language is.   For example, the lack of closures in Java 1.0 (which for all intents and purposes they should have introduced in that version) has limited the evolution of the language, as it won't even get watered-down closures until mid-2012 at the earliest.

Michael Urban replied on Sun, 2010/09/19 - 10:04am in response to: Otengi Miloskov

Apache Pivot is better bet than JavaFX really. JavaFX it is dead since day 1.

 

The big problem with Apache Pivot and the main reason it will have trouble getting any kind of market penetration, , is that for some reason, the developers decided to re-invent the wheel rather than wrap Swing. The result is that you cannot embed any of the thousands of existing advanced third party Swing widgets in a Pivot application. Combine that with the fact that Pivot's stock widget selection is rather limitted, and it's currently not possible to write anything much more advanced than relatively simple data entry and display applications in Pivot.

The one thing that JavaFX got right was that they used Swing. So embedding any of the existing third party advanced Swing widgets in a JavaFX app is trivial.

Greg Brown replied on Mon, 2010/09/20 - 7:39am

for some reason, the developers decided to re-invent the wheel rather than wrap Swing.

Sometimes you need to re-invent the wheel in order to move on. See the Pivot FAQ for more information:

http://pivot.apache.org/faq.html

The result is that you cannot embed any of the thousands of existing advanced third party Swing widgets in a Pivot application.

That's not entirely accurate. It's not impossible - it's just not something we have focused on. And "thousands" may be a bit of an inflated estimate, no? :-)

While it is true that you can't currently drop an existing Swing component into a Pivot app, you can use Pivot within a Swing app, which may provide a better migration path as it allows you to build new functionality in Pivot and retain any existing Swing-based infrastructure until it is no longer needed.

Also, because Pivot is based on Java2D, you can easily take advantage of existing 3rd-party drawing APIs in a Pivot application - see this example that uses JFreeChart:

http://ixnay.biz/pivot-jfree/charts_demo.html

Pivot's stock widget selection is rather limitted

Here is a class diagram that shows Pivot's current component set. I personally think it is pretty comprehensive. What specifically do you think is missing?

http://pivot.apache.org/tutorials/platform_overview/wtk_classes-large.png

it's currently not possible to write anything much more advanced than relatively simple data entry and display applications in Pivot.

Having participated in the development of several non-trivial Pivot applications, I can assure you that this is not true.

The one thing that JavaFX got right was that they used Swing. So embedding any of the existing third party advanced Swing widgets in a JavaFX app is trivial.

I haven't verified this myself, but my understanding is that Swing support has actually been dropped from newer releases of JavaFX. If true, that's not much of a selling point for JavaFX.

Greg

Mike P(Okidoky) replied on Mon, 2010/09/20 - 1:27pm

My experiences are that J2ME is indeed dead. It simply does not make any business sense to invent any solutions using J2ME. Java development on Blackberries is insanely horrible. Actual Java is not available on most big platforms. Android is a Java dialect. Either Java full SE with Java2D and everything makes a come back immediately, and on platforms like iPhone, Palm Pre, etc. like pronto, or else the only Java like technology we'll ever see on mobiles is Android.

Or will we.... everything is going to change in the mobile world and I've been saying this for years ever since the first netbook came out (EeePC was it?). Eventually, actual PCs will be miniaturized small enough where they fit in a mobile form factor. Intel just came out with their new Moorestown chip, which is a low power Intel instruction set Atom system-on-a-chip. This *will* have a dramatic impact on *everything*. No more shoehorning everything inside J2ME, Android, or whatnots. You'll simply will be able to run todays full Java 6 on your near-future mobile. Simple as that!

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.