JavaFX, why did it take so long?

I have been on a long holiday, and I missed the release of JavaFX 1.0 early December. At first glance, I didn't really understand why it took about 18 months since JavaFX was announced at JavaOne 2007 before a first package was released. But I think I understand it now.

The release of JavaFX 1.0 was accompanied by a video message from Jonathan Schwartz, Sun's CEO. That was a surprise to me, since Jonathan's blog does not mention Java developments too often. So this must be something really important to Sun.

One of the key assets of LodgON is that our developers combine Java on the back-end and Java on the front-end. We are very active users of GlassFish, and we also have projects where we write Midlets or MHP code.
It is my impression that Java on the back-end is "ok". Especially with the GlassFish v3 roadmap, including OSGi, I am very confident that Enterprise Java is going in the right direction.
In most of the projects we do, back-end development is the easy part. The difficulties start when one of the so many web development frameworks have to be used. Most development goes into the web tier. Not even into integrating the web tier with the business tier, but rather integrating an external design into JSP's or any other web technology.

I can imagine that the release of JavaFX 1.0 has been preceded by a lot of discussions that try to find a balance between the completely different environments of designers and developers, where the focus is clearly not on back-end developers.

The software is out, time now for the first real references. I think it is worth some R&D budget…