Geofence monitors have buffer zones because a small GPS on someone’s ankle inside a house, potentially a high rise, isn’t going to be that precise.

33 feet is well within the allowances IIRC. Varies by state.

▲

chaps 2 hours ago | parent [-]

That might be true for a place in the middle of nowhere with no features, the same device model managed by the same company and the same call center person and same sheriff and judges and prosecutor. But with GPS in an urban environment where many homes are over 100 years built from many different materials, 33 feet is a lot.

▲

Aurornis an hour ago | parent [-]

Urban environments have larger buffer areas due to multipath interference.

The story you linked has examples about false alerts triggering in a different state.

Do you have any actual examples of 33 feet getting someone arrested? I’ve worked enough with GPS that I could tell you that any system with thresholds that low would result in everyone getting false triggers

	▲	chaps an hour ago \| parent [-]
		Yes, everyone on the devices received false alarms. It's why there was a call center that managed the false alarm flows. I highly suggest you read the articles I've shared. I fought tooth and nail to get access to the data showing how often devices reported "invalid" status (where the device's specs said it was something like, within 150 feet of inaccuracy it reported "invalid"). They initially told me they couldn't get me that data because it'd crash their database running that query. Basically threatened to sue them and over the next 6mo or so I got two 100GB MSSQL dumps. Hundreds of millions of rows of devices reporting back that their GPS signal was invalid. It would take a lot of time to make it sharable because they included the actual GPS location infos.