Is it reasonable to learn D

[Trass3r]

> - The D compiler has only bad code optimization
Yep, but there is LDC and GDC which use LLVM and GCC as backends  
respectively.

> - There are no maintained GUI libraries
I wouldn't agree with that. Some people are still working on GtkD, QtD and  
DWT.

> - The development of the compiler is very slow
More and more people are contributing patches so development has  
definitely become faster.
Also Don has more or less taken over development of the CTFE  
functionality. Nice trend.

> - Only a small community
>   => no real German community
There is no separate German community but there are plenty of Germans here.
Manchmal sieht man sie nur nicht sofort ^^

> So I ask you - Is it reasonable to learn D?
Yeah, learning D is definitely worth it, if only to know why C++ and  
especially C++0x sucks ass ;)
The language and its potential is just great.

In my experience you shouldn't use D for a particular project though if you
- have pressing deadlines, cause compiler bugs might become very painful.
- depend on C++ libraries or have to code a module for a big C++  
framework. I tried using SFML but wasted a lot of time writing the wrapper  
code instead of actual application code.

Apart from that it's perfectly possible to accomplish non-trivial projects  
with D, be it against all odds like
http://h3.gd/code/nucleus/