Pure Danger Tech


Java 7 Roundup (July 19th)

19 Jul 2007

This is a weekly update on new Java 7 information. All current and previous information is maintained on my Java 7 page.

In general Java 7 news, Frederic Simon recently gave an overview of Java 7 at Java Tech Day in Israel.


Java 7 modularity, specifically JSR 277/294 vs JSR 291 OSGi, seems to be a hot topic these days. There is probably no better place to look than Glyn Normington’s summary of where we are and where things are going. Glyn has a foot in both camps and is an excellent writer – I am glad he’s involved and helping to translate for the rest of us. The Java Posse Episode #133 also has some discussion of JSR 277/294 and OSGi. InfoQ interviewed Eric Newcomer about the future of OSGi and the discussion turned to JSR 277 near the end.

These days I’m not sure what to think about it all. I used to be pretty excited about JSR 277/294 because I thought having it built into the JVM would make using it in some scenarios easier to approach than OSGi. But I’ve come to really appreciate the depth of the OSGi approach and the undeniable fact that it works now and people are adopting it at a good pace. As I read the various mailing lists, it’s starting to sink in that there is even more complexity than I originally thought involved in putting 277/294 support into Java. The impacts on classloaders, packaging, etc are going to be a big change. I’m not sure at this point how it’s going to fall out. I’m still hopeful that at the end of the day we will have a usable modularity system built into Java that complements OSGi.

More info: JSR 277, JSR 294


John O’Conner has expanded his series of blog posts on JSR 296 into a full article on Using the Swing Application Framework.

There was a very interesting message from spec lead Shannon Hickey regarding a proposal for addressing EL lock-in and other issues that JSR 295 has been criticized for by the community. Mikael Grev posted some follow up comments.

More info: JSR 296 JSR 295

Types and Generics

JavaPosse episode #132 has a really fantastic interviews with Bill Pugh and Brian Goetz, which covers a variety of topics, but also talks at length about JSR 305 annotations, which are based on JSR 308 support for annotations in more locations.

More info: JSR 308

Language Proposals

Stephen Colebourne posted on BigDecimal literals and other big thoughts.

John Rose posted some thoughts about tuples in the JVM which is something that has been kicked around for a long time. John also posted some interesting ideas about a tail call bytecode in the JVM to better support dynamic language optimization.

More info: BigDecimal operator support