I think you're right; and I cede that point. I can't claim that most are infringing. I know there are some mp3 sites like beemp3 [https://www.google.com/searc

I think you're right; and I cede that point. I can't claim that most are infringing. I know there are some mp3 sites like beemp3 [https://www.google.com/search?q=related:beemp3.com https://www.google.com/search?q=related:beemp3.com]. I suppose I was thinking about the ease of a human to judge that, yes, beemp3 is nearly entirely made of infringing music. That humans can report these sites and algorithms can find similar sites (and similar files) and develop a pretty decent database of items they could omit from search results. My point is only that it does seem technically feasible.