Other partners are more strident. Vishal Sikka, the CTO for SAP, pushed for Sun to release the Java Community Process and pointed to a proposal made by Oracle in 2007 to vest control in an "open independent vendor-neutral standards organization where all members participate on a level playing field." Although Oracle has said little about these details, it's fair to assume that the company will find new respect for the opinions it may acquire from Sun.
But while the other partners will probably engage in some saber-rattling, there's a limit to what they might consider because everyone is locked in a relationship bound by mutual-assured destruction. Any political fractures, no matter how tempting, may be prevented because everyone recognizes the power of the write-once, run-everywhere philosophy. If code written for an Android phone doesn't port easily to a Sony Blu-ray player or the Glassfish server, the opportunity opens for another language to emerge. There are already companies that write toolkits that make it simple to run Ruby on Rails or Python projects on your iPhone.
Many of these newer languages can be serious competitors, and too much in-fighting would only help them gain ground. There are dozens of very good, if a bit obscure, open source packages, and they are usually ported to every platform with just a bit of conversion by their devotees. Python, for instance, is already the dominant language in some branches of science, and Google chose it, not Java, for the first language supported by its App Engine. Although IBM, Oracle, and Google may throw discrete elbows in a battle for dominance, they must realize that all of them benefit more from the standardization than from dominance. If one pushes too hard to gain control of Java, the others may flee to a rival.
Java for the masses
The fact that the Java code base remains fairly cohesive and relatively standard leads many to speculate on just how Sun and, if the merger closes, Oracle might monetize the ubiquity without jacking up the licensing fees and squeezing everyone. What if someone created a kind of pan-Java App Store that made it possible to sell little Java-coded widgets just like the apps on the iPhone, but ones that would run on desktops, cell phones, Blu-ray players, or anywhere else that Java might be found?
There's already a decent collection of apps available for Android phones, and Sun is working on a Java Store that will let people just drag an application onto their desktop. It would take only a bit more code to tie together the applications and let them run in almost every corner of the Java world. The cosmetic issues of dealing with the vastly different screens and user interfaces would certainly be a challenge, but they would probably be pretty manageable.
Imagine how attractive it would be for the developers to write one app that can start the day on an Android phone, live on the desktop at work, and come home to run cleanly on the Blu-ray players in the living room.
Bringing Java apps to the couch dwellers would even be a bit ironic because Java the language was intended to run on set-top boxes in its early days when it was still known as Oak. Such a vision would build on the greatest strengths of the platform -- its ubiquity and cross-platform stability -- and perhaps help it evolve into a modern mechanism for bringing cash into the hands of the creators.