Right. I was conscious of that when we started this program. One of the things we're sensitive to is the workload. I didn't want to have my team responding to lots and lots of false alarms.
What needs to happen is the conditions have to complement the technology and the technology has to complement the conditions. For example, I believe the committee's heard of something called drift. There's a satellite drift. If you think of it as a map and the dots are populated on an offender's house, the dots will sway back and forth. You need to make sure that when you're building, let's say, exclusion zones, you're doing it in such a way that you're capturing drift. You are anticipating drift when you set those up to reduce the number of false alarms.
Third-party monitoring is another way we layer this. For example, another alert we get is called a strap tamper. If an offender knocks the device really hard, maybe he's tripped and fallen down the stairs, the device will momentarily believe he's trying to take the bracelet off. Protocols need to be established for this. So now you have to build in a layer of monitoring. Do we respond right away? Or do we allow the device to reconnect, re-establish, say the device is fine, and then move along? That's an important question.