science mag and a lot of others thought he was worth listening to...
the argument is that it's not something simple like calorie in, calorie out (seen this argument on hn a few times), but rather how certain calories trigger certain behavior with insulin in our bodies.
After all, if he's right - it quite changes how we approach the problem, and considering the backing this research has from other reputable sources - it's quite foolish to just ignore such a paradigm shifting idea like that. perhaps it was worth ignoring before the guy wrote a giant book looking at one study after the next....before he presented such a good case for the argument...but at this point, obviously wrong is not true.
if you've got a good reference, a link to an article perhaps, to why Gary Taubes is wrong -- I'd love to see it.