Nutrition research is really difficult; lack on immediate reactions (digestion and metabolism take time), highly variable results because on stimulus in other parts of the body (including stress) and its being ethically impossible to have a control group (locking people in a room and feed them one this isn't realistic anyways).
On the curve of "knowing how much we don't know" we're just realizing this is way harder than anyone expected even a few years ago.
Totally. I signed up for the Kickstarter of uBiome years ago who's market line evolved from "We'll figure out a 'normal' biome and help you get there" to "Turns out there is no normal, let's try grouping people" to now something around "OMG this stuff changes constantly! We're going to go science for a few years and get back to you latter with something".
An odd catch-22 of not investigating other possible solutions so they could get a legal precedent is that they couldn't tell anyone outright that's what they were doing. So someone trying to he helpful ended up ruining everything. No no good deed goes unpunished.
On the curve of "knowing how much we don't know" we're just realizing this is way harder than anyone expected even a few years ago.