We have such libraries for JavaScript, PHP, Ruby, Python,.NET and Go. Implementation in Java won't be that different and difficult. It would be great to have one for Java of course but the proposed cost is more than 20x too much. It's just a simple RPC client with few serializers (boolean, string, varint etc) and ecdsa. All transaction types use the same set of serializers so adding support for new transactions is trivial if you already have at least some of them. You also overestimate the impact it will make. I would vote for an 8-10k proposal but 165k is a joke.
RE: HIVE JUMP PROPOSAL - JAVA API FOR HIVE