Microsoft says the app will use crime statistics to determine what parts of town are to be avoided. But it’s unclear where the data will come from and how it will be interpreted.
Microsoft has filed a patent for the app, but the actual product is unnamed and not available yet.
Opponents like Wallace fear it could hurt minority communities.
“It’s almost like gerrymandering,” she said. “It’s stereotyping for sure and without a doubt; I can’t emphasize enough, it’s discriminatory.”
A couple of things jump out at me here:
- As the story suggests, interpretation is critical. Consider Madison.
- Most areas are reasonably safe during the daytime, but some get dicey in the evening and night. Does time of day figure in? How do you flag an area that is dangerous day or night?
- Some areas have a high number of crimes because the population density is high; the rate is not out of line with other places. Do they use rates or raw numbers?
- One area with high crime stats is State Street, which is also one of the attractions of the town. Drunks are easier targets, and some are aggressive; so you can avoid a lot of it if you choose the right time of day. Frat row (a block away from State Street) has high stats too; battery, sexual assault, and lesser offenses; though I can't imagine that area as a city attraction.
- Most areas are reasonably safe during the daytime, but some get dicey in the evening and night. Does time of day figure in? How do you flag an area that is dangerous day or night?
- This is exactly the opposite of stereotyping. Areas where strangers are apt to be attacked are also areas where locals die regularly. All this does is make local knowledge available to out-of-towners.
Perhaps Wallace fears that entrepreneurs visiting town would pass by the violent neighborhoods and miss the opportunity to look at a graffiti-ed building and have an epiphany: "That must be a crack-house; what a wonderful spot to base my order fulfillment team!" Or if there is a restaurant that manages to attract people into a dangerous zone anyway, it isn't likely to lose much business from passers-through who probably haven't heard of it in the first place. The only (honest) people likely to be harmed are those in intermediate crime areas, if prospective shoppers look them up and say "Well, this other place is only half a mile farther, and the crime rate is lower." In a medium sized city I'd not expect that to be an issue, and in a large city most people seem to deal with a medium-sized subset and can probably be trusted to interpret the numbers.
It seems unlikely that Wallace somehow profits from the fate of unsuspecting travelers, and she can't be so foolish as to imagine that the people living in the city don't know the shape of it, so I'm led to conclude that what she really objects to is calling things by their names.
2 comments:
I'd call this the opposite of stereotyping, not because we have any idea whether the violence is happening equally to locals and strangers, but because it's based on public crime statistics rather than, say, local gossip about which neighborhood is presumptively more dangerous, perhaps because of assumptions about the meaning of its ethnic mix.
It's not stereotyping to believe a neighborhood is dangerous on the ground that it produces a lot of violent crime.
Perhaps Wallace believes that NAM neighborhoods will be unduly prominent, and is trying to shove away any racism in advance. If so, then it would be Wallace who is doing the stereotyping.
Post a Comment