So, we can either run with the assumption that people at Apple are idiots which seems massively unlikely, or the assumption that people at Apple are reasonably

So, we can either run with the assumption that people at Apple are idiots which seems massively unlikely, or the assumption that people at Apple are reasonably smart, which seems more likely. When Google Maps was getting spun up, they relied on people adding POI, reporting issues and so on. This was a long, long while ago but it's still possible to report concerns to Google. Crowdsourcing will not solve it, but it will become a big factor in determining what they need to do. Lets say 10% of people report an issue, and of those 10% there's 2% reporting really specific issues in one specific area of the globe, that allows Apple to then understand a key facet of the dataset and where it can be improved, that gives them knowledge of what issues exist in the current dataset and gives them the opportunity to see whether they can apply that on a larger scale. More data, more input, faster the problem is solved. Everything I've reported has been sorted, things I've not reported but are in the same generalish area have been fixed. I'd assume Apple is running essentially a massive data mining operation to work out where needs work. If everyone jumps ship I'm sure Apple will continue to improve it but it'll be much, much slower. In short: if you think crowdsourcing is useless, you are probably wrong.