Gateway isn't evil, Invitrogen is evil. Gateway is a great technology though as long as you don't mind paying for the clonase. I've found that it works really efficently and consistantly is is particularly useful if I need to put a single construct into several different targets (i.e. lentiviral vector and transient mammalian expression vector). I agree with Aaron with the additional point that if you're PCRing with attB-flanked primers, make sure you're using a proofreading pol.