[Talk-us] Neighborhoods / Zillow

[Serge Wroclawski]

On Tue, Jun 11, 2013 at 6:06 PM, Clifford Snow <clifford at snowandsnow.us> wrote:
>
> On Tue, Jun 11, 2013 at 5:47 PM, william skora <skorasaurus at gmail.com>
> wrote:
>>
>> Given the subjective, fluid nature of neighborhoods - especially
>> boundaries - where one neighborhood ends and one begins - may change from
>> person to person, they are best represented as a single node in the area
>> where there is greatest consensus that the neighborhood is located. This can
>> be very roughly estimated by OSM mappers who locally live in or near the
>> area.
>
>
> One reason for including boundaries is querying to determine what exists in
> a neighborhood. Another is to see the result from a search using nominatim.

Clifford,

Your reply really doesn't address what William is saying, which is
that neighbourhood boundaries are subjective. I think we all agree
that neighbourhoods are useful, but they're worse than political
boundaries in terms of being unsurveyable.

> And while the boundaries may not be exact, people can always change them!

OSM's model is about improving surveyed information, but does not
handle subjective data well.

If you think that a boundary is on one place, and I think it is
another, the fact is that we both may be right. OSM doesn't handle
this concept.

- Serge