Right but couldn't that type of system be roughly accomplished by looking up the canonical URL and/or title from the meta tag (from a scrape) and search against

Right but couldn't that type of system be roughly accomplished by looking up the canonical URL and/or title from the meta tag (from a scrape) and search against the last posts? Having a stronger policy on repeated duplicate offenders should be a thing. It just farms points that can be later used in a bot farm.