Good post describing the thought process involved in selecting a language. It is interesting to see that the choice boils down to having resident experts and the ability to steer/train the team quickly. I wonder what the decision would be if the entire team was open to using the best available language even without resident experts in that language. For APIs my biased decision would be node.js using CoffeeScript.
I think this misses the point a bit - the "best available language" is an assortment of criteria. If you take away some of those criteria, you're no longer asking the same question.
In that case this post tells us more about constraints present in their environment than about a process of selecting the right tool. We should gather many "best available language" definitions with their respective choices. Then we could extract some meaningful, general advice, otherwise it's anecdotal and not worth very much.
Thanks! I can't imagine you'd want to use Dropwizard with Play! since they both accomplish similar tasks. That said, I'm not really familiar with Play, but to my understanding its a full web application framework. Dropwizard focuses more around creating web APIs but it does include a templating library in case you want to create applications beyond APIs.
