With JavaFX 1.0 out of the way, it seems like the Java 7 train has started moving again.
Concurrency plans posted by Doug Lea. Two points of interest are that the parallel collections API is probably not going to get included (but the core fork-join framework will) and that a low-level fences API might get included. Without closures, it’s hard to include the parallel collections API with good performance for primitives while avoiding interface explosion. The positive aspect is that it will continue to be developed outside the JDK giving it more flexibility in terms of API evolution.
Devoxx summary by Danny Coward containing various links covering JavaFX, JDK7 and Languages in the JVM.
Summary of Java 7 update given by Mark Reinhold written by Hamlet D’Arcy. Small language changes and Project Jigsaw had already been announced and many of the items described there had been known for a while, but some new ones are Null dereference expressions, XRender pipeline for Java 2D and the chance of a MVM-lite. The XRender pipeline is interesting because it was developed by Clemens Eisserer as part of the OpenJDK challenge and would be a nice example of an external contribution to OpenJDK.
One thing that is clear from the various announcements is that Sun intends to use OpenJDK as the place where work for Java 7 will take place (even if JSRs are also planned). Some people expressed concern that this is bad because it circumvents the JCP and that OpenJDK is a project controlled by Sun since the vast majority of the committers are from Sun. Given how opaque the JCP is, I don’t see how OpenJDK could be worse, but my hope is that Sun will run these projects transparently as promised where the end-result is diversification of committers due to more external participation.
Finally, for the people who still use the proprietary JDK from Sun, there’s an early access release of JDK6u12 that includes a 64-bit Java Plugin.