The web has always been semantic to some extent, just not in the rigid academic sense. It will be scraped, chewed, regurgitated and recycled. That's better than catalogued.