I do a lot of NLP stuff and the Python/Java divide is definitely a problem right now. I'm happy with both languages, but generally prefer Python. A lot of the state-of-the-art tools, especially the Stanford stuff, is in Java, while many of the more modern and experimental stuff is in Python. In theory it should be really easy to use both, but in practice I've always had to deal with bugs in the gap between the languages, whether Jython or OS level integration. So currently, yes, there are some new tools (including this one) that I would definitely try if they were written in Python and I could just do a pip install + import, but never get round to testing out because like this one they are more clunky to set up and require Java.