Why? Because it'll cause a great deal of FUD around cryptography
I assume that you're making this statement because of prime factorization, right? This is actually one of the problems that people aren't really too sure about which complexity class to place it. Proving P = NP might not matter all that much in this case. Of course, it mattering would spur on alternative encryption strategies and/or increased awareness of existing algorithms not based on prime factorization.
solvable (in human-time) solutions to so many interesting questions' rather than just answers.
Well, it might be a bit less useful in the real-world than many of us have come to believe. First, there is the possibility that the solution is something like n^100, which would probably not prove to be very useful. Second, while many NP problems might be interesting to computer scientists and mathematicians, I'm not too sure that it will be viewed as warmly by the general media unless computers could suddenly perform tasks like language translation, speech recognition and object recognition perfectly. I don't believe that these tasks are actually in NP -- ditto for Chess, Go, and many other games. Of course, many of the computationally expensive tasks that people already perform, eg HDTV encoding, are presumably already in P and won't benefit from a stunning P = NP finding.
Anyways, I hope to get to the paper early next week. It should be interesting.