[snip] Bernard Vatant kindly pointed me at http://www.geonames.org/statistics/total.html which gives some raw material info on which features are most common.
A sample from spot features section, where the number is the count in geonames 132 S.CMPLA labor camp a camp used by migrant or temporary laborers 131 S.RSGNL railroad signal a signal at the entrance of a particular section of track governing the movement of trains 128 S.SNTR sanatorium a facility where victims of physical or mental disorders are treated 128 S.ZOO zoo a zoological garden or park where wild animals are kept for exhibition 125 S.PKLT parking lot an area used for parking vehicles [...] I don't think this ranking correlates in any simple-minded way with the kinds of features people care most about for mainstream search apps. But then, if there are more labor camps than parking lots in the world, it seems important that this be preserved in structured data sharing. Most likely more many many people search for parking lots than for labor camps; but many people also use search products to decide travel plans, a scenario in which both kinds of feature might be equally relevant. My gut feeling is to patch up the high level stuff we have in schema.org to fix any obvious omissions, and then find a mechanisms based on http://blog.schema.org/2012/05/schemaorg-markup-for-external-lists.html for including the full detail by reference. cheers, Dan _______________________________________________ Geowanking mailing list Geowanking@geowanking.org http://geowanking.org/mailman/listinfo/geowanking_geowanking.org
