Digital photos are an important aspect of the modern multimedia experience. The contributions of online collaborative communities can now be leveraged for the task of automated image annotation. We present a web tool, SpiritTagger, which suggests tags for a user's photo provided the location where it was taken. We accomplish this on a world scale by mining data from nearly 1.5 million geo-referenced images from the website Flickr. By applying statistics on the geographic distribution of tags, we can bias towards suggesting ones that occur more often locally, hence capturing the spirit or genius loci, of the place.