I'm not sure that is entirely true. There have certainly been a few people who have attempted to study what did or didn't work in industrial settings - either pure academics or people working in industrial research labs. But I agree that currently we have nowhere near enough data to form robust conclusions about almost anything in this field and I think this is the strongest argument that the industry is not ready for any kind of licensing and regulation regime.