Re: Great idea (Not)
I think the basis is that proximity has to continue for some threshold period. OTOH it would need to add up the number of sub-threshold encounters. After all, it's some level of probability that should be the trigger.
But I think you're right. It will generate a lot of false positives, too many for those thus identified to be isolated. It needs to be the fornt-end for testing and a more capale testing system than currently exists.