ColourOfSpring - 10:09 am on Jun 28, 2013 (gmt 0)
People are assuming "clones" literally clone DMOZ - literally clone the content, anchor text etc. Sure, those exist. But it also wouldn't be hard to spider DMOZ and create your own directory based on the listings in DMOZ. You could get a list of URLs from DMOZ, then spider THOSE URLs to create anchors, descriptions etc - you could even take random samples of sentences on the home page of each URL (doesn't need to make sense, after all). You would then omit random URLs so the sub-directories aren't mirrors of DMOZ. In the end you create a "content stew" that is zero quality to a human visitor but unique enough to a Google spider not to know that its origins are from spidering DMOZ.
How would Google properly detect the above scenario? It's not hard to spider (or scrape) a URL and then manipulate what you find to create a unique version of what you found.