Monday Sep 30, 2013

How Java Helped Team USA Win the America's Cup

Congratulations ORACLE TEAM USA!

The team leveraged a lot of technology--including Java--to win the America's Cup. One member of the sailing team was on the team for both his sailing and programming skills. Here's a video with Gilberto Nobili, sailor and Java devloper.

While sailing, they "collect about 3,000 variables a second." The code provides real-time data while they are racing. When asked why he used Java, Nobili explained that he needed Java's "write once, run anywhere" ability. "I knew the system needed to be able to run on multiple devices and multiple operating systems," he explained, "Java provides the best way to do that." You can learn more in the article "Wind Powered. Data Powered" from this summer's Java Magazine.

Thursday Sep 26, 2013

The JavaOne 2013 Java Community Keynote

by Timothy Beneke

Geoff Lees, Senior Vice President and General Manager, Microcontrollers, at Freescale Semiconductor, got things rolling Thursday morning at the Java Community Keynote to a packed, standing-room-only crowd. Lees presented a precise and thoughtful vision of how the coming Internet of Things (IoT) might become a reality.

“The microcontroller community is rapidly moving to adopt Java and we need your help,” said Lees. He described a picture of the IoT that, for him, changes daily with ever-increasing speed. Current network developers often have a conception of edge node devices that consist of X86 processors with large scale OS, or multicore mobile processors with mobile OS. He spoke of an alternative picture in which edge node devices are everywhere in our environment, monitoring local climate conditions, monitoring mechanical factors like stress loading, and traffic data, and making smart roads happen. He sees node devices dealing with our busiest highways, offering road tolling and other measures to even out traffic flow. Or in agriculture, the IoT might monitor climate, humidity, weather conditions, and local microclimates. He spoke of smart homes where existing home networks are coming together to conjoin future personal networks and even body networks. He spoke of the IoT in aiding fitness and health tracking.

Lees elaborated a vision in which local autonomous data intelligence is gathered through local command control, and data sets are passed onto the cloud for higher level analytics. He explained that the IoT is changing the way the semiconductor industry is thinking about technology, in terms of processing node transitions, greater utilization of advanced sensor technologies, integration of those technologies, ever rapid adoption of low power technologies both from processing as well as design techniques. Advances in signal analog integration and the IoT are bringing these things closer.

“Instead of the next few years, we’re thinking about how to do all of this in the next few months,” said Lees.

In addition, a new class of products is arriving where connectivity underpins all product development. “Over the last decade,” commented Lees, “the microcontroller industry has shipped 150 billion devices into the market, the vast majority of which were not interconnected -- they were single, local points of intelligence. Today, the plans for those products are becoming ubiquitously part of another network. Networking technology needs to be built in with the capability to run advanced communication stacks present.” All of this will require more memory, more technology, and higher gate density.

He spoke of software becoming the big differentiator in the microcontroller community. “Today,” he explained, “in the embedded microcontroller community, our customers spend approximately 60% of their R&D on software as compared with hardware development. In the next 3 years that is forecast to become 70%. We need to see the software enablers as underpinning these system solution products. All of this comes together with a wide range of networking protocols, low-power communication protocols, and stacks. And the Holy Grail would be to have IPV 6 readily available from the cloud to networking, all the way to the edge node. Until that becomes enabled by cost, by process technology, by miniaturization and power reduction, we will have a wide range of sub-net protocols.”

He spoke of a move towards universal MEMS devices and technology for sensors, followed by the integration of universal MEMS within microcontrollers and embedded processors. “All of this,” he insisted “will come together as part of the ecosystem with all of these smart services. And these smart services depend on available local service providers and service models that we, the consumer, want to pay for, or that health insurers want to pay for. Providing this service and platform for secured services is a key for the adoption of the IoT.”

He expanded on his vision of a landscape for IoT, which would extend all the way from the cloud, down through processing nodes, local connectivity networks, and on to edge or sensor nodes, where the sensor, the MCU, and local connectivity come together. The challenge today, according to Lees, is that this diversity uses a huge range of technologies, different developmental environments, different software ecosystems, and different partner networks. There exist different security classes and considerations -- even down to the current edge node devices which offer no appreciable security. This clearly is not a landscape that will provide the kind of secured service delivery that the IoT needs.

The key to creating this secure IoT lies with Java developers. Edge nodes offered through Java and Java ME offer the potential to have secure encryption and authentication services throughout the network. Developing those in other environments will be locally difficult, will not be global, and will not reach the tipping point required for the IoT to develop. “Edge nodes is a category encompassing thousands and thousands of applications -- it’s billions of devices, most of which will include some form of MCU or embedded MPU technology,” Lees explained. “Sensors and actuators, integrated connectivity and an energy source, whether that’s energy harvesting or ultra low power battery technology and long life power generation – these nodes will need to be installed in remote places with battery life spans of 10 years or more – clearly out of the range of today’s processor technologies. And many will need to be industrial or automotive grade technologies.”

He spoke of a challenge to bring performance and cost requirements so as to harness and secure larger memory stacks which will require more software in the next 2-3 years.

After offering considerably more technical detail, he spoke of how IoT could benefit humans, focusing on home tele-health and the home health hub, something Freescale has invested in recently. The availability of a wide range of personal biometrics and health analytics in the home health hub could benefit our health enormously, according to Lees. Until now, the technology has been limited by the absence of open standard secured service across the network to complete the link onto the health provider, and medical insurance providers. The new model offers a huge adoption of preventive health as opposed to diagnostic clinical health. Consumers, health insurers and the medical industry all favor it.

“Measuring heart rate, blood pressure, and blood glucose continually every day through changing conditions offers an immense picture of our personal health with diagnostic insights into the development of progressive disease and chronic disease management,” observed Lees. “At the local hub level the installation of analytics will offer feedback and guidance on lifestyle, on the management of conditions and offer the personal feedback needed for us to all manage our own health. The data can also be sent in datasets for further analysis in the network and cloud so the next time you visit your health practitioner, all of that data is available.”

He reported that Freescale and Oracle have announced a new agreement collaborating on engineering to develop a platform for software and hardware models for both edge nodes and a wide variety of gateway solutions. “We’re working on optimizing Java together and bringing Java functionality further into the network. We’re even envisaging what Java might look like on a transmit-only Bluetooth low-energy node in the field for 25 years, a concept today that is unthinkable for a Java model. The key is to provide a secure service delivery for our customers and for service providers across the industry.”

The rest of the keynote consisted of a variety of informative and entertaining presentations.

Donald Smith, Senior Director, Product Management, at Oracle’s Java Platform Group, took the stage and commented on recent developments in recent JavaOnes. JavaOne 2011 was about moving Java forward and rebooting the infrastructure after the Java SE 7 launch. Last year, the theme was innovation and showing Java’s role in major tech segments like the cloud, big data, IoT and open source.

“This year,” he said, “we take one step beyond all that, and celebrate the 'End User' and 'Application Developers'. We want to go beyond the typical ‘ISV’ that appears at JavaOne keynotes and show some inspiring applications being built thanks to the hard work of the Java ecosystem.”

After some rousing appreciation of the Raspberry Pi Challenge, the Codegarten and other JavaOne matters by Tori Wieldt of the Oracle Technology Network, a series of creative contributors to Java took the stage, hosted by Henrik Stahl, Vice President, Product Management, Java Platform Group at Oracle.

-- Java Champion Stephan Janssen, founder of BeJUG, Devoxx(4kids), Parleys & playpass.be talked about “Devoxx for Kids,” a one-day event he organized in response to his 11-year-old son’s desire to learn to program.

-- Oracle Academy Vice President, Alison Derbenwick Miller, described the Academy’s important work impacting 2.5 million students in 102 countries.

-- Aditya Gupta, an impressive 10-year-old Minecraft hacker, entertained the audience with his hacking demo where he showed off some video explosions.

-- Two Duke Segway robots appeared on stage and strutted around under the guidance of Java Champion, Stephen Chin.

-- A video was shown of Java Champion Paul Perrone’s Java-powered cars.

-- Drew Hylbert, VP, Technology and Infrastructure at Opower, came onstage and shared how Java technology is used to enable consumers to save energy.

-- Mike Marzo, a Technology Fellow at Goldman Sachs, discussed the value of the 100 million lines of Java code that Goldman’s developers have written over the years.

--Sean Phillips, Senior Software Engineer at a.i. solutions, presented a video that explained uses of Java in NASA MMS mission operations software.

-- Finally “father of Java” and Chief Software Architect at Liquid Robotics, James Gosling appeared and remarked that Aditya Gupta made him feel that he too should be a Minecraft hacker. He showed the view from the water looking at Hawaii from one of the Liquid Robotics nodes and explained in detail how it all worked.

The Community Keynote offered a strong sense of renewal and pride at what Java has accomplished and where it may be headed.

Freescale Semiconductor

Liquid Robotics

Perrone Robotics

Watch Keynote and Session Highlights on Demand

At the Java Demogrounds: The Internet of Things at JavaOne 2013

Throughout JavaOne (and Oracle OpenWorld) the movements of attendees were being tracked in an impressive demonstration that powerful applications of the Internet of Things can be rapidly put together in order to gather a host of data with relatively little expense or effort. Throughout the conference, IoT in Motion has been efficiently counting and tracking conference attendees in various locations to reveal the power and utility of end-to-end data collection and management technologies. IoT in Motion is a collaboration among Oracle, Eurotech, Hitachi Communication Technologies America (CTA), and Hitachi Consulting.

Oracle’s Jennifer Douglas provided a concise overview of the technology: “We have Hitachi Consulting, who helped build the actual application that is running the data, using an Oracle Exalytic box over at Open World and the Oracle BI (Business Intelligence) dashboard. People from the Oracle BI team also contributed to this. Hitachi CTA has their SuperJ running in conjunction with Oracle’s Java SE embedded through a gateway to the Eurotech Everyware Cloud, which collects the raw data. Then the Exalytic box compiles the data and converts it into something we can actually utilize.” All of the technology is running on Java.  

IoT in Motion is not to be confused with security tracking using face recognition software which can recognize and identify the movements of individual people. While it can distinguish a dog or a vehicle from a person, the stereoscopic camera merely registers and counts people going in and out of the spaces without monitoring any features of individual people. No one’s privacy is violated in the process of tracking.

Oleg Kostukovsky of Oracle’s Java Embedded Global Business Unit, articulated the importance of IoT in Motion for the Java developer. “This solution from end to end is built on Java,” explained Kostukovsky, “all the way from the embedded device to the back end. So a Java developer can leverage existing Java skills to develop the application. All of the underlying pieces and blocks to enable application development are already in place, so if you are customizing an application running on a gateway, there is a Java framework available for that. It’s the type of environment your typical java programmer is used to so they don’t need to know about any specific embedded stuff or connectivity with sensors. On the back end, we are leveraging Oracle middleware products – pure Java based. You can develop Java code and connect Java adapters to different sources of data. So there is nothing you need to know except your basic Java development skills – it’s very similar to a Java enterprise scenario. So the learning curve is very low.”

Oracle’s Internet of Things Platform

Session Report: DataFX: The Best Way to Get Real-World Data into Your JavaFX Application

Java Champion Johan Vos of Lodgon and GUI expert Hendrik Ebbers of Materna GmbH gave a session on Vos’ labor of love, Project DataFX, a 2-year-old free and open source project intended to make retrieving, massaging, populating, viewing, and editing data in JavaFX UI controls easier. In so doing, they extend the functionality of JavaFX.

The goal of the session was to teach attendees how to develop enterprise applications in JavaFX with real-world services and data. They were focused on showing how the DataFX framework facilitates the process of data retrieval and rendering, and how it allows developers to focus on their application-specific logic.

According to Vos and Ebbers, the real value in most client-oriented business applications is the data sitting on remote servers and cloud systems. Retrieving and displaying this data must be done correctly before end users can interact with it. The open source DataFX framework aims to simplify this by enabling JavaFX developers to easily retrieve data from a variety of sources in several formats and rapidly integrate it with JavaFX components (such as TableView), using typical JavaFX patterns. The session introduced the DataFX project, gave practical advice for using it, and provided insight into future applications.

The session explored a variety of tools used in DataFX: the Concurrency API, DataFX Executor, lambda expressions, DataReader, RestReader, DataProvider, Controller API, Flow API, and many more.

The DataFX Project
Java Champion Johan Vos
Hendrik Ebbers
Be sure and go to Parleys.com to view the session in early October.

Session Report: 50 New Features of Java EE 7 in 50 minutes

 by Timothy Beneke

On Tuesday afternoon, noted Java EE authors Arun Gupta and Antonio Goncalves offered a whirlwind tour of new features in “Java EE 7: Fifty New Features of Java EE 7 in 50 Minutes”. Gupta is legendary at Oracle for his hard work and astute grasp of the Java EE platform. His blog offers a wealth of insight into Java EE and other Java matters. He is the author, most recently, of Java EE 7 Essentials published by O’Reilly. Goncalves is one of the most highly regarded writers on EE anywhere and the author of Beginning Java EE 7, published by Apress.

Java EE 7’s new features enhance HTML5 support, increase developer productivity, and further improve how enterprise demands can be met. Developers will write significantly less boilerplate code, have better support for the latest Web applications, and gain access to enhanced scalability and richer, simpler functionality. The session did a stellar job of spelling out the details to a packed house.

With four new components (WebSocket, JSON-P, batch, and concurrency), and three old ones significantly updated (JAX-RS, JMS, and EL), along with other significant changes to the platform, a lot of new functionality has been added.

They divided the new Java EE 7 features into 19 categories and explained an average of two to three features in each category.  Here were the categories:

CDI 1.1 (JSR 346)
Bean Validation 1.1 (JSR 349)
Interceptors 1.2 (JSR 318)
Concurrency utilities 1.0 (JSR 236)
JPA 2.1 (JSR 338)
JTA 1.2 (JSR 907)
EJB 3.2 (JSR 345)
JMS 2.0 (JSR 343)
Servlet 3.1 (JSR 340)
Web Socket 1.0 (JSR 356)
Expression Language 3.0 (JSR 341)
JSF 2.2 (JSR 344)
JAX-RS 2.0 (JSR 339)
JSON-P 1.0 (JSR 353)
Batch 1.0 (JSR 352)
JavaMail 1.5 (JSR 919)
JCA 1.7 (JSR 322)
Java Connector Architecture
Default Resources

Here are just a few of the high points:

CDI 1.1 (JSR 346) enables finer scanning control and the ability to veto the processing of a class or package. Bean Validation 1.1 (JSR 349) allows for method validation and the ability to pre/post conditions on method and constructors. Interceptors 1.2 (JSR 318) focused on the ability to associate an Interceptor associated with a constructor and the ability to prioritize interceptor bindings.

For Concurrency utilities 1.0 (JSR 236), the emphasis was on ManagedExecutor with a focus on:
* User threads in Java EE applications
* The ability to support simple and advance concurrency design patterns
* And to extend Concurrency Utilities API from Java SE (JSR 166y)

Further emphasis in concurrency was on ManagedThreadFactory and DynamicProxy.

Dynamic Proxy:
* Creates dynamic proxy objects, and adds contextual information available for applications running in Java EE environment
* It supports Classloading, JNDI, Security, …

Also covered as part of concurrency: ManagedExecutor
* User threads in Java EE applications
* Support simple and advance concurrency design patterns
* Extend Concurrency Utilities API from Java SE (JSR 166y)
– java.util.concurrent package

In addition: ManagedScheduledExecutor
* Managed version of ScheduledExecutorService
* Submit delayed or periodic tasks

For JPA 2.1 (JSR 338), standardized database schema generation and the ability to define additional indexes in schema generation were emphasized. JTA 1.2 (JSR 907) was praised for its capacity for transaction management on Managed Beans as a CDI interceptor binding; in addition, it offers CDI scope whose lifecycle is scoped to the currently active JTA transaction.

They discussed WebSocket and annotated server endpoint which enables full-duplex bi-directional communication over a single TCP connection.

JSON Builder creates an object model (or an array) in memory by adding elements. JsonParser is an event-based parser that can read JSON data from a stream.

All in all, it was an impressive display of Java SE 7 expertise.

Java EE 7 Essentials by Arun Gupta

Beginning Java EE 7 by Antonio Goncalves

Be sure to check out Parleys.com in early October to listen to the entire session. It's well worth it.

Wednesday Sep 25, 2013

Session Report: Is It a Car? Is It a Computer? No, It’s a Raspberry Pi JavaFX Informatics System

by Timothy Beneke

On Tuesday, Oracle Technology Evangelist Simon Ritter presented a session that demonstrated how a simple Raspberry Pi computer could be connected to an Audi to access far more information about the inner workings of the Audi than is currently available. Ritter, a great showman who has entertained thousands of JavaOne attendees over the years with his Java tricks, communicated his contagious passion for gadgets – and for sharing them with others.


“I love computers and all things electronic, and I also love cars – I’m a petrolhead -- so I’ve combined these 2 passions into one thing,” said Ritter. 

His talk was divided into several sections. He began by talking about the reliance of cars upon computers and moved to a discussion of the Raspberry Pi and why it’s a good choice for the kinds of systems he built into his car. He went on to briefly provide some background about embedded Java and JavaFX, and why they are good choices for the Raspberry Pi. He then spent much of the session describing the software system he built in his car, its functionality, how it works, and so on. He speculated on possible future enhancements to his “carputer,” and, finally, showed a video that provided a sense of what it is like to ride in the car with the new data accessible.

He initially made the point that computers and cars are now inexorably tied together. His first car, a 1971 Mini Clubman 1000 had, aside from a radio, no electronics. In contrast, his most recent car, a 2011 Audi S3, has lots of electronic devices, like all modern cars: an engine control unit, fuel injection/electronic timing, “Fly-by-wire” throttle, an anti-lock braking system, Satellite navigation, auto-sensing wipers and lights, and much more, some of which are mandated by law. The key point is that cars are already heavily computerized.

He described how the bus architecture works in the Audi, a communication system that transfers data between components inside a computer, and presented basic information about embedded Java, Java ME and JavaFX.

The Advantages of the Raspberry Pi for Car Computing
Ritter described the history of the remarkable Raspberry Pi, a project begun in 2006 that was initially created to inspire children to learn about computers. It was officially launched on Feb 29th, 2012 and currently, nearly 2 million have been shipped at a cost of around $25. It is ideal for children because of its cost and ease of use.

Its core features are:

* CPU: ARM 11 (v6) core running at 700MHz
– Broadcom SoC package
– Can now be overclocked to 1GHz (without breaking the warranty!)
* Memory: 512Mb
* I/O:
– HDMI and composite video
– Audio out (3.5mm plug)
– 2 x USB ports
– Ethernet
– Header pins for GPIO, UART, SPI and I2C

Ritter pointed out that adult computer geeks have been playing with the Raspberry Pi in countless ways. He then summarized why it is ideal for car computing:
* It has plenty of computing power with low electrical power consumption (< 1 Amp at 5V).
* Persistent storage is provided by the SD card.
– Disk drives are not ideal in hot places with lots of vibration.
– It’s a supported device for embedded Java.
– It is configured for floating point acceleration.
– It works with Java SE Embedded and Java ME Embedded.
                  --   A JavaFX Prism graphics engine is ported.

His goal was to gain new information from his car in real-time and have it available for analysis to ultimately improve his driving style:
* Display realtime data
– Engine performance (Power, Torque, Load)
– Driver data (Throttle position, steering angle, braking force, etc)
– G-Forces on car
* Record data for later analysis
– Produce graphs to display changes over time
– Play at Formula 1
– Improve driving style

He went on to describe the creation of the accelerometer, touch screen, measures of torque, among other things, and closed with a 3-minute video showing the box in action in the car from the perspective of someone riding in the car with various measures visible. Then he played back the information recorded on the drive.

All in all, a super entertaining, informative session.

Ritter will be posting more info about his session here: blogs.oracle.com/speakjava

Audi
The Raspberry Pi
Check out Parleys.com where you can listen to the session in early October.

Session Report: Demystifying Java EE

Adam Bien, who is not only a Java Champion and JavaOne Rock Star, but was named in 2010 as Oracle Magazine’s Java Developer of the Year, spoke to an enthusiastic crowd where he addressed some core issues about Java EE. He encouraged questions – “The more heretical or offensive the better.” It was obvious that Bien loves to think about and code in Java. He remarked, “The more I code the happier I am”. Spoken like a hard-core Java developer!

First, he asked, “What is Java EE? Innovation vs. Standardization”?  For Bien, Java EE is nothing but a release of co-existing APIs. Before Java EE, there was a mess with lots of application servers, with absolutely no chance of finding two application servers with similar APIs. Java EE resulted in a huge simplification. Now with Java EE 7 a wealth of are applications available. Java EE, insisted Bien, was never about innovation because building a standard precludes innovation. “Java EE will always lag behind,” he observed. “For instance, Hibernate will always have more features than JPA. Spring will always have more features than CDI. Java EE is the 80% that makes products work. It was never about innovation.”

He boiled down the whole point of Java EE: “What matters are small WARs – the smaller the WAR, the faster the build and deployment. The faster the build and deployment, the more productive you become,” he insisted. He explained that Java EE enables you to not put everything into the WAR and place as much as possible on the application server and less on the WAR. He explained that most of his WARs in Java EE 6 or Java EE 7 projects are very small

Bien asked, “Are EJBs bloated?” He explained that the question implies some voodoo stuff behind the scenes making EJBs bloated. He offered a means to answer this question.

He went on to answer a wealth of questions in a way that was thoughtful, incisive, witty and, at times, a bit provocative.

Here's some of the topics/issues (pulled directly from his slides) that Adam touched on in this fast-paced session:

*Do we need transactions?
*Is Dependency Injection Black Magic, VooDoo, or both?
*Is EJB pooling needed? Are EJBs bloated? What happens, if you violate the EJB programming restrictions?
*Why AOP didn't take off in Java EE?
*Stateless vs. Stateful programming model?
*HA without a Cluster?
*Are there any POJOs out there? What happens during deployment?
*Is Java EE faster than J2EE? Does JMS 2.0 scale and perform well? Is Java EE only suitable for the "big" enterprise?
*Is JSF the silver bullet? What is the deal with CORBA and RMI?
*How to unit test Java EE applications? Why we don't build a best of breed server from scratch?

This was a lively, entertaining and information-packed session. Just what you would expect from a pro developer as Adam Bien. I highly recommend viewing this session.

Adam Bien’s Blog
Check out Parleys.com where you can listen to the session in early October.

Brazilian Java Man at JavaOne

In this video, Bruno Souza and Java Community Manager Tori Wieldt discuss JavaOne, the Java Community Process (JCP), cloud computing and mad scientists.

Java in Action at JavaOne: The People Counter

There's a lot of action at JavaOne, lots of movement. We've got robots, cars, door sensors, apps in the cloud and more. In this video, Architect Oleg Kostukovsky describes the People Counter, a combination of sensors, gateways, a database in the cloud, and a JavaFX UI which provides real time data on people movement.

Tuesday Sep 24, 2013

Session Report: JSR 341: Expression Language 3.0

Ed Burns, Consulting Member of Technical Staff, and Kinman Chung, Principle Member of Technical Staff, both at Oracle, presented a session on Monday in which they described new features in JSR 341, Expression Language (EL) 3.0. They discussed the APIs for the use of EL in standalone environments and EL syntax for new operators, plus lambda expressions, and support for collection objects, all the while offering copious code illustrations.

Burns remarked that he was pleased that Java Champion and JavaOne Rock Star Adam Bien had referred to EL 3.0 as “the hidden gem of Java EE”. “I don’t know how hidden it is,” said Burns, “but I think it’s a gem.”

He discussed the origins of EL, which has a long and active history in the Java platform. EL began in 2004 as part of the Java Standard Tag Library (JSTL 1.0), moved to JSP 2.0 in 2006, and became an independent specification with JSR 341 in 2011. It is used in JSF, CDI, and Avatar. Now, 9 years after its inception, it is an independent specification that is heavily used in JSF.   

Burns observed that the presence of EL is the key differentiator between Java server and non-java server stacks. “Java server-based web frameworks are likely to use EL,” said Burns. “When you show someone who is not familiar with EL how easy it is to move things together from disparate parts of your application, it’s very compelling.”

The most important feature that EL 3 brings is lambda expressions – developers do not have to wait until Java SE 8 is released. It all runs on Java EE 7, which requires Java SE 7 -- which means that it is currently available. Burns gave a brief discussion of lambda expressions, which basically behave like an anonymous function -- lambdas in EL are EL expressions. They offer full access to the EL environment within the body of the EL lambda expression, something not available from Java SE lambdas. “You won’t be able to refer to other EL things from a plain old SE lambda expression,” said Burns.

The goal of EL 3 is to provide greater expressive power for applications and to use it outside of Java EE. Burns and Chung provided an overview of collection operations and explained EL’s support for stand-alone environments. EL 3 is easy to use outside of Java EE and provides standard data structures: ELContext; ELResolvers; and local variable and function repositories.

They explained that it enables direct EL operations and has: EL expression evaluation; Bean definition; and Function definition. They emphasized that other key parts of Java EE can also be used standalone, such as: Bean Validation; Persistence (JPA); and Contexts and Dependency Injection (CDI). They encouraged developers to consider the possibilities for cloud deployment in: Defining functions and variables and defining beans.

They spent the rest of the session illustrating their key points with a healthy dose of code.

Links and Downloads:
* JSR 341: http://www.jcp.org/en/jsr/detail?id=341
    Download spec and API javadocs

* Project home: https://java.net/projects/el-spec/
   Report spec bugs or RFE for el.next

* RI: https://java.net/projects/uel/
   Maven artifacts available from Maven Central
   Download source and report RI bugs

* Integrated in Glassfish 4.0: https://glassfish.java.net/

You can listen to this session in early October on Parleys.com.

About

Insider News from the Java Team at Oracle!

duke
javeone logo
Links


Search

Archives
« September 2013 »
SunMonTueWedThuFriSat
1
2
4
7
8
9
14
15
16
18
20
21
27
28
29
     
       
Today