The problem of Qt Jambi
I like desktop UI development. Especially I like architecture of Swing, but not so long ago I felt the power of Qt, thanx to Qt Jambi. Now I know that while Qt have some architecture problems it has superior Graphics, more usable widgets, and saves a lot of time on borders and pixels manipulations thanx to it's advanced widgets.
But Nokia decided to stop supporting Qt Jambi, and gave it away to "community". While there are some smart guys on Qt Jambi mailing lists - it is not community yet. And of course I thought about ways to help Jambi survive.
The problem is, I am Java developer, and Jambi was created for Java developers. Most of people on Jambi mailing list are Java developers. It is not very popular among C++ developers because they will not be able to use the result of their work. While Jambi generator is written in C++, and to be able to update Jambi according to Qt changes you should know C++.
Ok, let's assume I am learning C++, to help Qt Jambi to write Java apps using Qt. Well, not Qt itself but bindings of Qt to Java. Do you see the problem? Why should I use Jambi if I already know C++ and I can use Qt itself, without bindings and additional abstraction levels. Then, why should I help Jambi if I will not use it?...
As for me Jambi-like project will not survive without support of some company, and stopping Jambi support Trolltech/Nokia almost killed the project. Now it most likely some parts of Jambi will be salvaged, for example it's extremely cool generator. Maybe in some time I will learn C++ a bit better and will dive a bit inside of Generator sources, and maybe even write something similar in Java if I'll have nothing better to do...