The Spring Blog

Engineering
Releases
News and Events

Why Open Source Businesses are not like Wal-Mart

Hopefully one or more open source businesses will be among the standout successes of early 21st century capitalism. However, it’s interesting to look back at one of the standout successes of late 20th century capitalism for an instructive example of one of the unusual challenges facing open source businesses.

Wal-Mart’s history is well known. The first Wal-Mart opened in Rogers, Arkansas in 1962. Five years later there were 24 stores across Arkansas. In 1968, Wal-Mart opened its first stores outside Arkansas, in Missouri and Oklahoma. Both those states, of course, neighbour Arkansas. Wal-Mart’s expansion thereafter continued to be in concentric circles around its home base, enabling logistical and cultural challenges to be overcome gradually and execution to continue to be highly efficient. Despite enormous success in Wal-Mart’s home market, wider geographical expansion has not always proven successful–as in Germany.

Read more...

CarPlant not accepting null CarModels

Last Friday I finished a training session at a client of ours. Because I had some time to kill in the hotel I was staying in, I polished the sample application I coded up during the training to post it online for the guys of the training. Usually I try to find a little sample application specific to the client’s domain to use during the training. This makes it a bit more lively instead of some of the HelloWorld examples.

This client is a big car brand, that have adopted Spring widely throughout their organization. That’s why I created a CarPlant system capable of producing cars. Below you can find a little UML diagram displaying the (rather tiny) domain model and services in the system.

Read more...

New on the Spring forum: regular polls

A while ago, we were thinking about doing a big survey among Spring users to render feedback about parts of the framework. We’d include questions like ‘which Java version are you using’, but also ask you which Spring feature you like, dislike, et cetera.

Partially because of lack fo this, but also because for now, we don’t want to force Spring users to fill out huge surveys, we never got to actually doing this.

In the past half hour, I’ve set up a nice alternative on the Spring Framework forum. From now on, we’ll be hosting a series of polls on the forum. The polls are basically very simple multiple choice questions and the results will be available to everybody. In addition to the poll, we’ll also give some background information on each of the options.

Read more...

Maven Artifacts

At long last I can finally say that SPR-1484 is resolved. Opened on 20 November 2005 with 121 votes, 63 watchers, and even its own anti-ticket this issue rates as one of the all time biggies. In the last 10 minutes I’ve uploaded the maven artifacts for Spring 1.2.9 to our local repo and you should be seeing them replicated onto the central maven repo in the next 6 hours or so.

You will notice on the Spring JIRA issues such as SPR-2704, SPR-1383, and SPR-3198. What this means is that we’re not done improving our builds and and we will continue to respond to the community. In fact, we’ve got some really cool new tools that I can’t wait to show off in upcoming posts.

Read more...

Oracle Contributing Oracle Application Server Integration Code to Spring Framework

On the theme of application servers embracing Spring, another update. Oracle have been working on value added integration with their application server.

This is similar to what we have with WebLogic 8.1 and above in WebLogicJtaTransactionManager. The OC4JJtaTransactionManager should be used in place of the generic JtaTransactionManager in an OC4J environment, and provides the following benefits:


  • Direct access to the transaction manager and helper classes without having to use
    JNDI lookups

  • Auto-detection of server version to get the most out of the different transaction manager implementations in different versions

  • Control over transaction isolation level: a very useful feature not available in JTA

Read more...

Sun's GlassFish Embracing Spring

Sun take open source seriously these days, and users seem to be starting to take Sun open source seriously too.

GlassFish was late to the party in open source application servers, but it seems to be gaining traction. And, more importantly, it actually seems to be pretty good. Various Interface21ers, including Costin and Juergen, have taken a look at GlassFish and given it the thumbs up (although we haven’t yet worked with it in production). From what I’ve heard, performance is excellent–probably substantially due to the reworked servlet engine based on NIO. The JPA implementation–TopLink Essentials–should perform well too, being based on the mature and performant TopLink engine.

Read more...

WebLogic 10 Tech Preview Ships - Builds on Spring Framework

Congratulations to the WebLogic team on shipping a preview of WebLogic 10, which passes the Java EE 5 CTS. It’s good to see BEA getting back to their tradition of being quick off the mark, after their aberration with J2EE 1.4.

This is interesting news for the Spring community, because WebLogic 10 uses Pitchfork internally. The Pitchfork Project is an open source project led by Interface21 and collaboratively developed with BEA that implements EJB 3.0 interception and JSR-250 injection on top of Spring. It is used inside the WebLogic 10 EJB container and in other parts of the server to meet new spec requirements, but is not tied to WebLogic. Pitchfork can be used standalone (although that’s not its primary goal) or could be adopted by other application servers.

Read more...

Why the name Acegi?

Between Rod’s recent blog on the origins of the Interface name, a recent thread querying the renaming of Acegi Security, and a suggestion late last year from my colleague Ben Hale to blog about the origins of the “Acegi” name, I’ve decided that it’s probably time to do so!

First of all, the pronunciation: it’s “ah-see-gee”. Now that we’ve got that out of the way, let’s go through where it came from:


  • Back in the early 1990s, I started a bulletin board system. I still remember using my first 300 baud modem, and actually watching the characters appearing at the same speed as I could read them! Anyway, a BBS required a name, and I quite liked “Midnight BBS” (or was it “Lightning BBS”?). After attempting to register my preferred name with the Australian BBS Registry, I received a phone call to be informed that the name was taken. I searched for another name, found no conflicts, and changed all of my screens (anyone remember Avatar?). I then went to register the BBS with its new name, although someone else had just beaten me to the same name. There were about 800 BBSes in Australia around the time - this was competitive stuff! So, I decided to go with something that was guaranteed to be original: characters 1, 3, 5, 7 and 9 from the English alphabet. Acegi BBS was then born.

  • This was the BBS heyday. It had callers from all across Australia, and it was the first BBS in our area code to acquire the seemingly limitless capacity of a CD-ROM. I became the 3:624 network coordinator for Fidonet and helped distributed “echomail” throughout our region. Before long, people started sending cheques payable to “Acegi BBS”, as they wanted greater file download limits and access to Fidonet’s netmail.

  • Because I needed to cash cheques with “Acegi” in the name, in early 1993 we decided to register a business name. Acegi Computer Technology was selected, and people happily made out their cheques accordingly.

  • In about 1995, I moved to Sydney with work. Dial-up Internet was just starting to become publicly available in Australia (it still cost around $5 per hour for modem access!). Because a BBS draws most of its users from a particular local area and I was leaving that area, I gave Acegi BBS to a friend to continue running. He too subsequently moved, and passed the BBS to another friend. I lost track after that. I did a Google for “Acegi BBS” and even found an old 1995 record of it.

  • By 1996 I needed a company for my IT projects, so Acegi Technology Pty Limited was born. It has remained operational since.

  • Sometime around November 2003, I wrote what eventually became Acegi Security. I put it into a ZIP file and shot it across to Rod and Juergen. I proposed the new project be called “Spring Security”. However, they didn’t have time to fully review it at that stage, so suggested I simply get it out there as “_____ Security System for Spring”. As such, I prefixed “Acegi” into the name. As of today there are about 660,000 hits on Google for a search of “Acegi”, and nearly all of it relates to the security framework.

Read more...

Dynamic DataSource Routing

Spring 2.0.1 introduced an AbstractRoutingDataSource. I believe that it deserves attention, since (based on frequent questions from clients) I have a hunch that there are quite a few ‘home-grown’ solutions to this problem floating around. That combined with the fact that it is trivial to implement yet easy to overlook, and now I have several reasons to dust off my corner of the team blog.

The general idea is that a routing DataSource acts as an intermediary - while the ‘real’ DataSource can be determined dynamically at runtime based upon a lookup key. One potential use-case is for ensuring transaction-specific isolation levels which are not supported by standard JTA. For that, Spring provides an implementation: IsolationLevelDataSourceRouter. Consult its JavaDoc for a detailed description including configuration examples.

Read more...

Java to JavaScript Compilation with AJAX and Spring Integration

For some time I have been interested in client-centric, web-based user interfaces. These Generation IV frameworks are characterised by their component-based, event-driven programming model, and focus on the presentation logic residing entirely on the client. Targeting a web browser in this manner typically necessitates the use of JavaScript or Flash, which in itself imposes a number of unique challenges.

It is possible to address many of these challenges if we can program in Java and automatically produce a JavaScript or Flash runtime module. Two well-known products for achieving this today are Google Web Toolkit (GWT) and Open Laszlo respectively. Both are available under OSI-approved licenses and have active communities, together with their own unique complexity. One consideration is to what extent they fulfil an objective of providing a transparent Java-based development environment that targets web browser deployment. This consideration has several facets, including IDE support, debugging integration, reflective capabilities, runtime widget binding and alike. All of these are normal considerations when developing rich clients using traditional Java technologies such as Swing and Standard Widget Toolkit (SWT).

Read more...