The Source for Java Technology Collaboration
User: Password:
Register | Login help    

Search

Online Books:
java.net on MarkMail:

Bedbugs and Ballyhoo

Posted by editor on March 10, 2009 at 9:08 AM PDT

Browser plug-in stats aren't what you might expect

An InfoQ article Flash is Dominating the Landscape, but Silverlight is Growing uses statistics from the RIAStats page to size up the competition between RIA browser plug-in platforms. The stats (as surveyed from 1.4 million browsers across 41 sites) seem to refresh periodically, and some are dubious (what is this "Version 1.8" that occasionally shows up in the Java pie chart?), but they're interesting nonetheless, particularly in the ways they defy your expectations.

To wit: you would think that the now-GPL OpenJDK, available in an easy-to-use package for the major distros, would be more popular on Linux than the closed-source and proprietary Flash plug-in. And you would be wrong. Filtering to only show Linux results (from 15,350 browsers across 28 sites over the past 30 days), RIAStats shows Flash present on 81% of those Linux browsers, while Java is only on 28%. Conversely, over on the Mac, where Steve Jobs earned the endless enmity of the Java community for calling Java a "big ball and chain" when asked about putting it on the iPhone, Java is present on 96% of the 73,909 surveyed Mac browsers... surely a result of the fact that Java is included with Mac OS X. Still, that makes the Mac the platform with the most consistently-observed Java availability.

Over all browsers, platforms, and countries, version 1.6 (i.e., Java SE 6) appears in about half the logged visits. It's too bad the report isn't granular enough to show how many of those Java 6 users have the "consumer JRE" (Java 6 Update 10 and later), which would be highly useful for those planning to deploy JavaFX, given that some of 6u10's improvements are highly valuable to graphics-rich applications like JavaFX.

Also in Java Today, Danny Coward has posted a collection of links describing the use of LWUIT on Interactive TV platforms, in Java ME: LWUIT, as seen on TV. "You can see LWUIT on TV in this short video. A little like the JavaFX runtime, which runs over an underlying Java SE runtime (suitably consumerized), or a Java ME CLDC runtime, LWUIT runs atop today's Java ME CLDC runtime (like, for example, on this touch-enabled Samsung F480), and now the CDC based TV platform. For developing on TV, you can get the SDK and emulator here. All works with NetBeans 6.1 or later."

Rakesh Menon has posted an example JavaFX Slide Show to the JFX Studio site. The slide show player dissolves between slides from the JavaFX presentation at Sun Tech Days Hyderabad, and can be viewed as an applet. Source for the SlideShow.fx and ImageButton.fx classes are also available.

Today's Weblogs begin with Jean-Francois Arcand's GlassFish v3 Extreme Makeover using GrizzlyAdapter part 1: Hello World. "GlassFish v3 offer a lot of extensibility point, and one of them is quite interesting: GrizzlyAdapter. Any applications developed the GrizzlyAdapter API can be deployed inside v3 and transform a fish into a monster...an extreme makeover!"

Ramesh Parthasarathy looks at STUN server in SailFin "SailFin can be extended to provide STUN [Simple Traversal of User Datagram Protocol (UDP) Through Network Address Translators (NATs)] service, BINDING requests primarily. A simple STUN server is now available in Project SailFin and is implemented as a lifecycle module."

Arun Gupta continues a recent set of tips with TOTD # 72: JRuby and GlassFish Integration Test #3: JRuby 1.2.0 RC2 + Rails 2.2.x + GlassFish v3. "The third test (explained in this blog) ensures that the same application can be easily deployed using GlassFish v3. GlassFish v3 will be available later this year with full support for Java EE 6."

In today's Forums, fenixx would like to List the JNDI of all running Beans. "Is there any possibility to list all the running Beans? Of course it's possible to list it at the Glassfish Admin Console, but it doesn't list it all. I would suggest the remote Beans are listed only. For example: I want to login at my application. Therefore my AuthenticatorBean is called correct, but I don't see it in the listing."

In another request, vrakesh_techie wants to Open a new Java Fx Stage from existing Stage. "I'm working on Java fx. In my Application, when i click on certain tabs, then a new window/Stage should open up . For this purpose what i'm currently doing is that, in my java fx file, i'm declaring all the necessary stages and setting the opacity for all other stages to 0 and the opacity of primary stage to 1 . So when i launch the Application, so many windows open up but only one of them is visible . When i click on Tabs like configure Settings or Configure Mail Server sub tabs in my main stage, then i will set opacity of the other stages to 1 so that they are visible But when the Application launches, several windows are opening up in the task bar and this is looking ackward. So kindly let me know if there is a way to open a new Java Fx stage from an Existing Stage."

Wolfram Rittmeyer explains GlassFish's concept of pinging in the follow-up Re: unable to ping "GlassFish's ping-connection-pool command (or it's web based equivalent) simply tries to get a connection for the given configuration (datasource class, databsename, user, pwd and so on). If it gets a connection the ping command is considered to be successful."

Finally, turbogeek considers the nature and liabilities of agents in Re: JXTA for Multi Agent Systems. "One of the most important insights I ever heard in regards to agents was from Graham Glass. He said the essentially the one primary goal of an agent was to conserve bandwidth. This was said over ten years ago, but it is still true. If the duty of an agent is to perform actions based on instructions, then the agent is said to be successful. Having said that, the problem with agents in general is that they are hard to manage if you believe agents roam the internet. You have trouble with CPU/memory consumption and security. If you can control all three of these, there is a fourth issue that the agent could talk to anyone. But the last little hobgoblin of agents is reliability. How do we know the agent is succeeding? Can it crash the system or be managed when it does crash? So, agents sound really really cool, but they are very very ugly. This is why their are few agent systems in common use."

Current and upcoming Java Events :

Registered users can submit event listings for the java.net Events Page using our events submission form. All submissions go through an editorial review before being posted to the site.

Archives and Subscriptions: This blog is delivered weekdays as the Java Today RSS feed. Also, once this page is no longer featured as the front page of java.net it will be archived along with other past issues in the java.net Archive.



Browser plug-in stats aren't what you might expect
»
Comments
Comments are listed in date ascending order (oldest first)

Submitted by mrmorris on Tue, 2009-03-10 17:31.
"You would think that the now-GPL OpenJDK, available in an easy-to-use package for the major distros, would be more popular on Linux than the closed-source and proprietary Flash plug-in." Not really. However you would think that a language with such a *nix background as Java, would be able to deliver a Linux JavaFX plugin before i.e. Silverlight. Not so however: http://www.go-mono.com/moonlight/ (And notice the one-click installation)

Submitted by osbald on Tue, 2009-03-10 18:27.
If Java comes pre-installed with OSX then does a figure of 96% mean 4% took the trouble to remove Java from their machine?

Submitted by osbald on Tue, 2009-03-10 18:31.
> It's too bad the report isn't granular enough to show how many of > those Java 6 users have the "consumer JRE" Well you guy's own the servers that provide these resources to the kernel installer, you could just process the logs and publish that information yourselves. Last time we heard about runtime detection (via google analytics) the script wasn't up to distinguishing between runtime versions/vendors.

Submitted by fabriziogiudici on Tue, 2009-03-10 18:56.
The phantom "version 1.8" has been reported also by others and it's a mystery - maybe it's a 1.1.8 erroneously reported?

Submitted by opinali on Tue, 2009-03-10 21:04.
@fabriziogiudici: This is obvious evidence of E.T.s visiting Earth! Civilizations that mastered space travel are more advanced than ours, so I'd expect them to already have Java SE 8. ;-)

Submitted by prunge on Tue, 2009-03-10 23:56.
I wonder if the previous lack of 64-bit support is affecting usage on Linux? I don't think there is an option on my 64-bit Ubuntu installation in the package manager to get Java working in Firefox (though Java works fine from command line, no webstart/applets). The release of 6u12 with 64-bit plugin and its inclusion in repositories might improve this.

Submitted by robilad on Wed, 2009-03-11 02:34.
Afaict, at least the Linux statistics are somewhat questionable: the pie chart says 18% for Java 1.6, while the line chart says 50%. They can't be simultaneously true. ;)

Submitted by cowwoc on Wed, 2009-03-11 08:03.
1.8 is a bug detecting is a Safari on Mac. See http://forums.java.net/jive/thread.jspa?threadID=53140&tstart=0 for more information. What is very worrisome about these statistics is that Java6 usage has actually gone *down* over the past 3 months. We need to figure out why.




 

Feedback  | FAQ  | Terms of Use
Privacy  |  Trademarks | Site Map
 
Your use of this web site or any of its content or software indicates your agreement to be bound by these Terms of Participation.
 
Copyright © 1995-2008 Sun Microsystems, Inc.
Collabnet Logo Cognisync Communities
Powered by Sun Microsystems, Inc.,
CollabNet and Cognisync