In a fish tank with plants, nitrites are dead simple to keep in check - and that's in a very small body of water, whereas this type of system would have a much larger volume, and be much easier to manage. Bacteria consume nitrite and convert it into nitrate. This process is relatively short, and once the bacterial colony is established, it can accommodate relatively large increases in ammonia input (like a dead, decaying fish) fairly quickly. Plants (and algae) consume nitrates extremely quickly. Anyone who has trouble handling nitrites in their fish tank is clueless, since the bacterial growth process happens entirely on its' own once they're present in the water.
I know this because I've had many fish tanks, the most recent of which was a saltwater reef. Dealing with nitrates in a reef is "hard", in the sense that you need something to consume them, as most corals don't deal with even nitrates (the end product, not the middlemen nitrites which are deadly in minuscule quantities) well, and some have trouble with algae growth near/on them. Once you have a separate tank with plants or algae, it's next to impossible to fuck up that aspect of the system.
The big problem that people have comes from commercial filter design and recommendations (far too small for the tank size), which largely don't contain enough surface area (Penguin Biowheels are one of the few power filters that even have a design specifically for surface area, and it's still not sufficient) to process the waste their fish create in the first place, then they overfeed and make the problem worse, then they add fish before the colonies are settled, and then they wonder why fish keep dying but add another one anyway, and then they don't do large enough water changes to remove the nitrates (on typical systems, there's no plants/algae there specifically to consume it). Done by someone with even a moderate amount of knowledge and experience, which you'd expect from the early implementations, it's a great idea.
