Neighborhood Boundaries by Tom Taylor uses Flickr Shapefiles and Yahoo! Geoplanet “to show you where the world thinks its neighbors are.” Yahoo! provides access to the Where on Earth (WOE) database, which attempts to describe locations as a hierarchy. For example – a town belongs to a city, a city to a county, a county to a state. The Flickr API stores shape files identified by the WOE ID. Here’s the punchline. The shapefiles are built using only the latitude and longitude from geotagged photos on Flickr. There’s no GIS involved here.
Why this matters, I can’t really say. I think it’s mostly to show how much data is stored in geotagged Flickr photos. I’m no GIS expert though. Anyone care to comment on the significance?
Well, by definition, GIS (Geographic Information Services) is a system for “capturing, storing, analyzing and managing data and associated attributes which are spatially referenced to the earth”.
Having that said, I believe Flicker is doing half the job here (capturing & storing the geographic information via GeoTags), Tom is doing the calculation and Yahoo is doing the management and association of that data to a map… Doesn’t that make the visualization a GIS? Or at least something very similar to it?
When I first saw the blog post on Flickr’s development site, I was pretty impressed/amazed that they had already accumulated enough geodata to be able to define geographic areas. What intrigued me though when I started playing around with Tom’s implementation of the data however, was the notion of psyhogeography (http://en.wikipedia.org/wiki/Psychogeography) that appears to be at play here.
Right now, some boundaries are pretty loosely defined, but as the data matures, there will be some very well-defined boundaries. Already some metropolitan areas have their neighborhoods quite well outlined. What would be very intriguing is to apply a time filter/slider to this data, say over several decades, to observe how the notions of neighborhoods and people’s overall sense of place in general changes with its inhabitants (or visitors) and then even over generations. It’s pretty far-flung and would require this data to be archived over time, instead of constantly amalgamated to a well-defined data set, but either way, it’s kind of fun to think about.
I checked it out for my neighborhood. Not exactly perfect. My city was showing way smaller than it actually is. That could be because the way people may loosely define the area they live in. I live in Oakland Park, but many people will call the area Ft. Lauderdale since it’s part of the Ft. Lauderdale metro area. And that is reflected on this map.
I don’t know whether to be amazed by the new tech here or be possibly scared by it. This has so much potential for good as well as evil in terms of what can be done with this level of information. One can only hope that before this becomes widespread as it has the potential to be that safeguards are put in place as to how this data can be used.
Pingback: Border Patrol, or: Where’s Dallas Again? | Clubs in Dallas
The hierarchical relationship is valuable. I’ve personally had to build an app that lets you drilldown through the relationship hierarchy. This was a pain because the boundary data we had did not come with these relationships. A simple thing like showing all the counties in Texas becomes harder without this information.
From what I understand about the technology in use here, this is to a GIS what wikipedia is to an encyclopedia. In other words, instead of being developed by ‘experts’ as a formal planned project, it depends on the input of many people to become complete and accurate. So it’ll be interesting to see how long it takes for the map to develop into something as accurate as a USGS map or similar. The early maps more or less represent what’s been contributed so far (as ellen points out). Also keep in mind that the overlays depend on an accurate base map to be geographically meaningful – not to say that they aren’t meaningful in other ways that are independent of the projection.