As I have already stated, I am for experimentation. It is from experimentation that we learn what works best and why. It may be that the ultimately best format for Hive or Hive communities is one that we haven't thought of yet - but it might not be apparent until experiments are completed.
Hive is quite a complicated system that reaches into human psychology, economics, tech and just about everywhere else. Since each person is unique, it is not really possible for me to know whether a community that doesn't have downvotes will thrive or not on Hive.. I can see how it could and I also see the concerns about why it might not. Ultimately, the point of layer 2 for me is that people are free to experiment and create voluntarily as they prefer. Provided they are not committing fraud or harming people in other ways, I only see benefits from the experiments in the long term.