I remember when I was a student a friend of mine telling me this guy was giving a seminar and he'd just completed day one and everyone was really excited he was going to prove FLT. Of course, the guy in question was Andrew Wiles. He then spent months patching up problems they found prior to publication and finally the whole thing got published. It was a hugely exciting thing when you were studying mathematics.
All of which is a long way of saying the line "the old-fashioned 1990s proof" makes me feel _really_ old.