That statement attributed to Dr. Taylor in the PCWorld article regarding 200nm feature size has been troubling my subconscious ever since I first read it. I originally chocked it up to either a misquote by the author of the article or maybe the good doctor just misspoke and didn't catch his mistake. It just doesn't make sense with everything we've learned about the technology so far. There are multiple sources for references to feature sizes of poet and speed that I recall. Here's one from the Toronto presentation.
http://m.youtube.com/watch?feature=relmfu&v=NH9yj3u5Nig
Just go to 5:56 in the clip to listen to Lee Shepherd talk about the speed of Dr Taylor's first chips he produced from about 3 years ago.
Green