Geofencing in this case directly led to the proper murder suspect. It was the technique that caught the right guy. The right guy was driving his relative's car and using the relative's old phone that he didn't sign out of, which led police to (entirely 100% legitimately) suspect the actual owner of both the car and the Google account. That was cleared up quite quickly, the right guy was arrested, all thanks to Good Guy Geofencing.
Where it went wrong is a completely orthogonal matter. He was kept in jail for days after it was crystal clear he's not the murderer, and his name was dragged through the mud also after that point. He's completely legitimately suing for that, and hopefully wins. But what does that have to do with a legitimate and effective use of his geolocation data? Again, a murderer is caught because of it. And then the same police department fucked up and ruined a legitimate and justified but innocent suspect's life.
> But what does that have to do with a legitimate and effective use of his geolocation data? Again, a murderer is caught because of it. And then the same police department fucked up and ruined a legitimate and justified but innocent suspect's life.
There is something fundamentally wrong with believing that the ends justify the means in these cases. As has been said, "[i]t is better that ten guilty persons escape than that one innocent suffer." [1]
Your argument can be applied to absolutely every investigative technique. Neighbours were interviewed, a suspect was identified, and then something went wrong for unrelated reasons. Clearly the problem is the very idea of interviewing neighbours, so we must eradicate the technique for privacy reasons, because Blackstone ratio and justice for all. The argument that this case is about police geocasting holds about as much water.
The question then becomes: How much harm will society suffer if that shoplifter got away or the murderer was not caught? In the former case, the store's insurance policy should cover the loss and prices are increased to offset the loss, and the shoplifter may do it again. In the latter case, it's possible that the murderer may kill someone else.
But the overall rate of shoplifting and murder isn't going to change substantially even if the police ruin a residence and refuse to pay for damages or arrest and jail the wrong person for a substantial period of time.
So it appears that there a negligible increase in overall risk to society and property if criminals sometimes get away, but there doesn't appear to be any substantial decrease in overall risk to them if they're caught at great cost to certain innocent individuals.
In other words, there's really no difference on a societal level whether or not these individual cases result in the arrest and conviction of these criminals, but there's a substantial difference in personal cost for the innocent individuals who were in the wrong place at the wrong time.
Where it went wrong is a completely orthogonal matter. He was kept in jail for days after it was crystal clear he's not the murderer, and his name was dragged through the mud also after that point. He's completely legitimately suing for that, and hopefully wins. But what does that have to do with a legitimate and effective use of his geolocation data? Again, a murderer is caught because of it. And then the same police department fucked up and ruined a legitimate and justified but innocent suspect's life.