That's not what the author means by "useful". If you spend a year describing a particular kind of plant, then you know that much more about it and that's a useful contribution to science if not society. If you keep coming up with zany theories and predictions that do not just turn out to be wrong (that's fine, "I've successfully found 10,000 ways that will not work") but where there's really no reason to even suppose they might have been right, then that's scientific busywork.