I would first like to say I’d very much support effort spent on the outlined “more ambitious alternative”, whether or not a “non-established” data quality assessment section is added to observations. That is, splitting off wild vs. not wild as a determinant of whether an observation is Needs ID or Research Grade. (Hi to 2018!)
As Scott mentioned, captive/cultivated (aka “not wild”) observations would still remain hidden to all users by default, but you could optionally select to view these observations if they’re a focus of interest. Or did you mean something else?
The differences between how this would apply to plants vs. animals seem significant and I’ll share more thoughts later, if others don’t beat me to it. :)