Connecting Java to the Open Web
+1 JavaFX
When Sun introduced JavaFX at the 2008 Web 2.0 Conference, it was pretty clear that Adobe Flash/Flex/AiR and Microsoft Silverlight had crowded out the previous Web 2.0 darling, Ajax. RiA was all the rage, with the highly interactive and graphical vector interfaces challenging the boxy clunkers everywhere you looked.
Here's the thing about JavaFX that makes it important. JavaFX embeds the WebKit layout engine and document model. This is the same HTML+ WebKit engine used by Apple iPhone/Safari, Google Android/Chrome, Palm Pre/WebOS, Nokia S60/S40/QT, RiMM Blackberry Storm, IRiS, Midori, and so many others. The WebKit layout engine is also used by Adobe AiR!
The WebKit document model has also contributed to the advance of HTML+, supporting HTML5, CSS3, CSS4, SVG/Canvas, JavaScript, and the emerging JS libraries such as jQuery. In this HTML+ push, Webkit is joined by Opera and Firefox. The results are a highly interoperable, very graphical and interactive, dancing document model capable of covering everything connecting desktops, devices and servers across the Open Web.
Running on the WebKit layout engine, JavaFX could connect the vast Java developer community to the Open Web. WebKit plug-ins have also hit the Eclipse developer community, with the Genuitec "WebKit for SWT" leading the way. This is good news for the future of the Open Web.
Mr. Ellison needs to march into the Hamburg OpenOffice/StarOffice, with his trusted lieutenant Phil Boutros by his side, and kick some ass. It's way past time that OpenOffice directly supported the Open Web with native read/write/collaborate capabilities. The world needs powerful desktop editors capable of speaking Open Web, and JavaFX might be the best way of doing it. Make no mistake though. This would be major surgery for OOo. I also think it would be a good thing to have seamless transparency between OpenOffice ODF and OpenWeb HTML+.
Finally there is Phil Boutros, the ace up Ellison's sleeve. Phil came to Oracle by way of the Stellent acquisition. While at Stellent, Phil was one of the founding members of the OASIS ODF TC (then Open Office XML). During his time working on ODF, Phil pushed the important issue of compatibility with the billions of legacy binary documents, and the applications that created them. His reasoning centered around the rather obvious logic that if the world was unable to convert these legacy documents to ODF (with an acceptable lossiness), there wasn't much point to developing a new format. For new formats to be successful, they would have to be compatible with legacy information and business processing systems.
In spite of Phil's efforts, (as well as the efforts of many others), ODF was not developed to be compatible with the legacy corpus of documents and document generating systems. Compatibility work was deemed by Sun to be outside the ODF charter, and outside the scope of the specification. An argument that had a lot of substance back then primarily because Microsoft was not a participant.
Of course if Microsoft had been a participant, backwards compatibility with the existing MSOffice desktop productivity environments would have been a primary concern for the ODF TC. Many argued however that the TC should give compatibility it's best shot regardless of Microsoft's abscence. The world needed us to do this. Besides, the ODF TC had the participation of some of the world's foremost binary reverse engineering experts, who spent their lives studying the internal MSOffice layout model and in-memory-binary-document representation. With renown experts like Phil Boutros, Paul Langelle (Corel), Daniel Vogleheim and Florian Reuter working on ODF, i often thought that it was unlikely that Microsoft could have had better representation of their MSOffice technical issues.
Now that Oracle is in charge of OpenOffice and ODF, i'm hopeful that ODF can get back on track. JavaFX is an interesting way of doing this. It works for all kinds of Oracle client/server/cloud/ria initiatives. And it works well for the future of the Open Web.
What's not to like?
~ge~