That works well for images because we have a tool for it (thanks Google!) and for blocks of text (I've often found whole paragraphs taken from other sources). What about entire works?

One thing I'm trying to work through is figuring out what the default assumption is. If someone posts on Steemit, should we assume it's original content? My thinking is, yes, we should. If they link to a source, we'll know it's not. Should we have to reverse search every image and every bit of text before we know we're not upvoting plagiarized material? That's really time consuming. It seems the whales are getting into a lot of trouble from the community for not doing this.