Hacker News Reader

Comment by te_platt | Hacker News Reader

I'm curious how this would look if it were somehow "divided by" population density. It wouldn't make sense for somewhere in the middle of the ocean but it might highlight more remote, scenic places. edit: For example I noticed that southern Utah is significantly brighter than you would expect from population density.

View on HN

That was my first reaction too.

But zooming all the way in (to Manhattan, for example), it's clear that the heat map corresponds to actual scenic tourist spots, it goes all the way down to street level, pretty amazing.

And since these are "panoramio" photos, not selfies snapped for Facebook, it would seem to be a pretty accurate reflection of tourist destinations, as opposed to population.

When you zoom in, you do in fact control for differential population density to a lot of extent.

Not for the markers though, they are mostly for cities with high densities of people.

The "all sizes" selection has a similar effect: take "small", "tiny" or "remote" to put markers on sparsely populated locations only.

I agree, I think the anomalies are more interesting than the expected.

Here's an example:

http://www.sightsmap.com/index.php?lt=45.104179&lg=-94.37551...

Also I'd love to do the reverse. Show me public places that have very infrequent photos. Especially campgrounds. I'd pay to know the least used campsites in my area.

Southern Utah is Arches and Moab etc.

Also Zion, Bryce. Spectacular sights.

I wonder if you could also normalize against weather - to differentiate between "Steady state" photos that happen regardless of weather vs. photos that are taken only when the view is clearly good

Reply on news.ycombinator.com