I like walkscore, though I'd be the first to admit it has many shortcomings. We found it very useful when looking for a place for my mother to live once she gave up driving. She couldn't make the decision based on walkscore alone, but it was a useful additional data point. It also creates awareness that is very lacking in American decision making about where to live. So, it's value is as much that it attempts to quantify something as the actual outputs of that quantification.

If I recall correctly, the one-mile radius was set because that was a reasonable distance for a person to carry (or pull/push with a cart, etc.) groceries for a household. Random, yes, but still useful. I walk the two-miles to our local grocery store or pharmacy, but not if I have to carry two gallons of milk.

We had a local "pedestrian friendly" movement start a few years ago by politicians prior to election time (I live in Virginia inside the beltway). This translated into funding overpriced brick sidewalks and crosswalks and a few WALK/DON'T WALK signs where none had been before. Helpful, but hardly a serious commitment, and the money would have been better spent on building new sidewalks where there are none rather than tearing out perfectly good concrete sidewalks and putting the brick ones in. What we really needed was an aggressive eminent domain statute to connect the culs-de-sac (cul-de-sacs?) with pedestrian easements that run between houses. Of course, nobody wants it running by their own houses . . . .

Google is really good at stuff like this, and I'd like to see them more aggressively pursue walking maps. One thing I'd like to see the Google Maps walking option include is cutting through parks. I can see the parks on the map, but Google will not let me walk through them. The same goes for pedestrian easements. Walkscore is road based, but imagine what it could do if, for example, it could include all the stairways in Pittsburgh. . . .