During a panel discussion on Skepticism and Magic at TAM7, the often-asked question came up yet again: should skeptics learn something about magic to help them in investigating paranormal claims? The response was the one I've heard many times before about how dangerous that can be: someone learns one way to do a particular trick, goes charging off to a psychic (or whatever) armed with this limited knowledge, then gets caught off-guard because there are other ways to do it that they don't know about. The term magicians use for this is "half-smart."
While I agree with the concern, I don't think it's a valid reason for stopping skeptics from learning about magic. Sure, professional magicians have to develop skills and knowledge that most of us don't, and never will, have. That deep understanding of their field allows them to see through the charlatans that purvey paranormal poppycock to the public.
But the same concern applies to just about any non-trivial endeavor — different jobs require different levels of skill and knowledge. The problem arises when we don't know what we don't know about the job we're undertaking. The "half-smart" charge ahead, not considering their limitations. That can lead to failure that is irritating, expensive, or even dangerous. In my own field, software engineering, I've seen plenty of examples of such failures, and I've had to clean up the messes left behind by half-smart software developers. However, I wouldn't use that as a reason for not teaching people about software. Why should magic be any different?
So by all means, I think skeptics should learn about how magic tricks are done, so they can understand how those tricks can be used to fool people. They should also learn about chemistry, physics, and geology so they can understand how to debunk homeopathy, perpetual motion machines, and creationism. Magic shouldn't be off-limits.
At the same time, skeptics shouldn't expect that learning a few tricks makes them expert magicians, any more than learning a few formulas makes them expert physicists. We should be skeptical of ourselves, too, and avoid the mistake of being "half-smart."