is essentially the list of limitations of XP as given in Beck's books.

You can't apply it when all 3 variables are fixed, youcan't apply it to large or distributed teams. It's hardly applicable when major interfaces are published.

While the failure of C3 project was a big bad news for me, I am not sure it's fair to assume the worst before we hear the other side.

His opinion on pair programming is interesting. But this happens to be the hardest part of XP. I know that I could not master it, and kept slipping into mentoring mode. This is a hard thing to do. If some study subjects fail to deliver, it's not a big deal - try looking at good people, not random selection.

Some things he says are just insane, or demostarte that he does not know what he's talking about. To say that "refactoring" is doubious practice is indeed insane. To say that time is wasted adapting test cases means he does not know what he's talking about.

XP is not for every project, not for every programmer and not for every toolset. But the elements of it that I managed to implement are precious to me, and no one will convince me otherwise.