I know we are well into 2019 now and that I did sum up 2018 in early January. In that post, I mentioned that Christian and I received the JCP Outstanding Spec Lead Award at the JCP party for our work on JSR 371. At the same ceremony, Daniel Dias Dos Santos received the Outstanding Adopt-a-JSR Participant Award, also for his work on JSR 371.
The SPRING INITIALIZR at https://start.spring.io has been around for a while and is the best way to bootstrap a new Spring Boot application.
So far, there hasn’t been a similar way to bootstrap a new MicroProfile project even if the different vendors have provided starters for their implementations. But the wait is over! The MicroProfile Starter is currently in “Beta”, but works like a charm. Just navigate to https://start.microprofile.io and start generating.
Based on which version of MicroProfile you select, you will get the available implementations that supports that particular version. You have the option of generating examples for the specifications included in the selected version. This is an excellent way to learn how the different technologies work.
The Java™ Community Process has been updated through JSR 387: Streamline the JCP Program. The most significant change is to open up for Iterative JSRs, i.e. JSRs that intend to deliver multiple releases of a technology on a time-based cadence. The driving force for this change is the 6 month release cadence for Java™ SE.
The first JSR using the option for being iterable is JSR 388: Java™ SE 13 (I guess the JSR name will have to be changed to not include the version number…).
Another change made in this update where changes to reviews and ballots in order to make the process even lighter to accommodate shorter release cycles.
OpenJDK64-Bit Server VM Corretto-8.202.08.1(build25.202-b08,mixed mode)
Amazon Corretto is licensed under the same license as OpenJDK (GPLv2 with CPE) and there are no costs associated with using it. Amazon will provide quarterly security updates to Corretto 8 at least until June 2023.
But Java 8 is sooo old!
Relax, Amazon plans to make Corretto 11 available during the first half of 2019. Corretto 11 will be based on OpenJDK 11.
I am super happy to have gotten the opportunity to test out Oracle Functions through theCloud Native Limited Availability Program. When I last tried out running serverless functions in Oracle Cloud during the Oracle Groundbreaker APAC Tour last year, there were two options available. Either run my own Fn server in a virtual machine or set it up in a managed Kubernetes cluster. Now, a third option is available!
Oracle Functions is built on Oracle Cloud Infrastructure (OCI) and offer a managed environment for the Fn project. This means that you don’t have to manually manage an Fn cluster yourself. It also means that any function that runs on Oracle Functions will also run on any Fn server, something that offers you full flexibility.
The Fn project supports functions written in Go, Java, Node.js, Python or Ruby. The fn-duke function that I am using in this test is, of course, written in Java.
Oracle has made a good choice when investing in the Fn project and use it as a basis for the Oracle Functions platform. It integrates extremely well with Fn and no extra tooling is needed to get started.
It’s this time of the year again. Time for the yearly summary of conferences, travels, community activities, open source projects, amazing people!
Like most recent years, I have been speaking at quite a few conferences around the World. The countries I visited as a speaker in 2018 were Sweden, Germany, USA, England, Denmark, France, Belgium, South Africa, Australia, and New Zealand.
Another acknowledgement by the community was to be re-elected for an associate seat in the JCP Executive Committee.
Besides speaking at conferences, a great deal of my time in 2018 was dedicated Jakarta EE at the Eclipse Foundation where I act as the PMC Lead of EE4J a well as being a member of the Steering-, Specification-, and Marketing Committees in the Jakarta EE Working Group.
All in all 2018 was an eventful year and I expect no less of 2019!
This is all very good, excellent actually! When you think about the size of it all, it is actually quite an achievement. We are talking about 7.7 million lines of code! More than 60.000 files and a total of 38 new projects that have been set up at the Eclipse Foundation.
But, as everyone knows, developers are impatient and eager to try out everything new, so there are still a couple of questions that I always get when talking about Jakarta EE:
When can I start developing Jakarta EE applications?
The answer to the first question is: “not yet”. Until the Jakarta EE specification process is finalized, the technologies are still Java™ EE.
The answer to the second question differs a little depending on who you ask, but usually is something in the lines of “I am pretty sure that some of the MicroProfile specs will be integrated into Jakarta EE when they have proven to be useful”.
NO, here is what you should do: Use the power of Java™ EE 8 and combine it with Eclipse MicroProfile.
Many of the application server vendors have added MicroProfile features to their Java™ EE 8 compliant or certified application servers. Examples are Open Liberty, WildFly, Payara and Apache TomEE. See the respective vendor’s documentation for which versions they have included.
I have put together a simple application called Jakarta EE Duke to demonstrate how to do this. The application uses the @ConfigProperty annotation from MicroProfile Config to configure a message as well as the new @Email annotation from Bean Validation 2.0, which came with Java™ EE 8 to validate input.
While this example is extremely simple, it does indicate how you can combine the full power of Java™ EE 8 with the lightweight APIs of MicroProfile to implement cloud-native microservices using Java™ technology.
One of the requirements for Eclipse Projects is that the name is not associated with any trademarks or potential trademarks. When we created the project proposal for Ozark, this turned out to be the case here. The name Ozark is simply used too many places for it to be a valid Eclipse project name.
So, how did we come up with the new name? First of all, we asked for input on the Ozark developer mailing list. We also wrote a small program that generated all permutations of ‘ozark’ to see if something cool came out of that.
Then we started filtering, discussion and voting until we ended up with Krazo, which turns out to be Ozark spelled backward. We are really excited about the new name and hope you all will join us in spreading the word that the reference implementation of MVC 1.0 that was previously known as Ozark is now called Eclipse Krazo.