I'm not really too worried about the demographic skew of the people posting the pictures. Even if people who just bought iPhones are excited to take pictures with their new toys right outside of the store, after that first few minutes they're still probably going to go to more or less the same places and find more or less the same things interesting as anyone else would, and do us all the favor of taking pictures of them and posting them online. The most-geotagged places are, in general, the most popular tourist sites, and that suggests honest mainstream tourism to me, not hipster irony.
But even if there are biases, it's still an amazingly phenomenal data set and I'm grateful to have it. Never before in human history has there been compiled such a vast list of places that somebody, regardless of who it was, found interesting (ten million of them in North America alone!) with precise locations, ready for analysis, and without the self-consciousness that would be involved if these were people trying to write guidebooks (or even just write down a list of their favorite places) instead of just taking pictures. My own plotting of the data puts the concentrations right where I would expect them, plus a few surprises, and now I look forward to checking out the surprises to see what other people found interesting that I never knew about before. And it seems like (unless people start intentionally rigging the data, which I hope will be detectable) it can only get more inclusive and more detailed from here as more and more people contribute. Eric On Mon, Apr 27, 2009 at 4:00 PM, Paul Harwood <[email protected]> wrote: > I was thinking about Hans Rolsings lecture, but instead of using government > data, using accidental data, the forensic, macro, data that the internet is > good at producing, to compute things of interest to all of us. I know GIS > has a much more scientific bent, but such anecdotal data can lead to new > ways of thinking about 'us'. After all, science was largely discovered by > accident. > > > On 27 Apr 2009, at 23:39, Paul Ramsey wrote: > > The original article glances off a defect in the data, with the >> mention that, curiously, the Apple Store is one of the most >> photographed locales in New York. The full sentence should be "most >> photographed by hipster doofuses who upload their photos to Flickr". >> There's probably a pretty serious urban/rural bias in this data as >> well as a large generational one. >> >> P. >> >> On Mon, Apr 27, 2009 at 3:36 PM, Paul Harwood <[email protected]> wrote: >> >> This flickr data is valuable though? not just eye candy, especially if >>> shared and open. Feeding it into something like the new wolfram engine >>> along >>> with say coordinates of power stations, you could compute VERY rough >>> temporal energy consumption maps pretty quickly I guess amongst a >>> multitude >>> of other much better probable uses. (don't shoot me down, probably the >>> worst >>> example I could have thought of). Interesting ? :-) Accidental mapping >>> could reveal a lot about 'us' when coupled with other bits of pretty >>> mundane >>> static data. >>> >> >> _______________________________________________ >> Geowanking mailing list >> [email protected] >> http://geowanking.org/mailman/listinfo/geowanking_geowanking.org >> > > > _______________________________________________ > Geowanking mailing list > [email protected] > http://geowanking.org/mailman/listinfo/geowanking_geowanking.org >
_______________________________________________ Geowanking mailing list [email protected] http://geowanking.org/mailman/listinfo/geowanking_geowanking.org
