the above is a nice article describing versions of “predictive policing.” however, the underlying problem is not the details of software but the tension between policing by the numbers and policing by relationships.
traditional policing has valued numbers, even though we haven’t necessarily paid a lot of attention to them as we work day-to-day. some of us have paid a lot of attention to relationships while others not so much, focusing more on “catching bad guys.”
to what extent is that tension driving ambivalence among officers — “productivity” versus relationship-building. is there yet a department that validly quantifies relationship-building? or does it remain as ill-defined as pornography, “we know it when we see it?”
moving forward, does what we measure comport well with what we are trying to do? if not, does measurement matter? how do we move forward?