From a computer science point of view it is perfectly possible to filter out records that ae completely impossible eg a census record for someone who would be 150 years old. No matter how many people may have added it to their tree. I think Ancestry should do this. It would greatly reduce the hints people have to sort through. I have over 1000 hints and it's likely a good 500 of these at least will be impossible, not just wrong. I don't mind looking at a hint and deciding it doesn't apply to the person in my tree but looking at a hint that is 100% biologically impossible really annoys me.
This is entirely true, but I think the problem isn’t one of computer science, but of economics.
From the perspective of any of the subscription-based genealogy websites, a dollar from a patient, thoughtful, meticulous researcher counts just the same as a dollar from someone who is just chaining surnames together back to the Middle Ages. What’s more, there’s probably considerably more of the latter than there are the former.
I’ve always felt the purpose of these hints is to generate that thrill of discovery
even where there are no discoveries to be made. From a cold-hearted business perspective it’s far better to have happy customers ‘discovering’ impossible new people for make-believe trees and who keep subscribing than have those same people get discouraged by the lack of ‘easy’ progress and stop paying money.
Maybe I’m just a cynical sod, but that’s how I’ve always seen it. The best thing to do is trust in your own critical thinking skills and build up your own logic muscles.