I was very annoyed by little quirks in languages. C++ didn't have proper modules, Java was too intrusive and verbose, Python was too slow and forced indentation, etc... However, I have realized that, while my projects actually are my code, the language doesn't have to, and cannot suit my needs completely (until, of course, Kava is implemented). So I'll rephrase that saying into "I am not my language".
And now I'm back to C++. Not that I was ever away, though, because I intended to implement the KVM in it, and had some other side projects for which I was already using C++.
There are, of course, technical reasons for which I chose C++ too. I thought about what I would actually like to achieve with Kava and then realized that, while as a standalone it would be a magnificent and fun project, the things I actually required from it swayed it a bit out of its scope. So, with the initial incentive to create Kava gone, had to push it back onto the project stack.
Hopefully, it will be popped back again soon, as it really is a fun project, and my design notebook is filled with notes, sketches and diagrams, which I cannot let simply go to waste.
Other reasons for delaying the development include the tightening of my schedule concerning the project that I needed a language for initially. I still might be able to recreate my old evolution simulator in time for another biology competition.