I'm impressed with the speech synthesis, but I want to see a systematic evaluation of the conversational automation. A few demo calls do not cut it. All of the

I'm impressed with the speech synthesis, but I want to see a systematic evaluation of the conversational automation. A few demo calls do not cut it. All of the successful Google innovations (Tensorflow, Alphago, Alphazero, map-reduce, millwheel, TPU) seem to have arrived with a proper publication style writeup. The ones that were less successful like Glass, Wave... not so much.