The World At One BBC interview was fun - lots of dodging around the 'why didn't you have a backup?' with a 'the real failure only lasted a short time' answer. Funny, I could have sworn the IT was really failed for a lot longer than that, even if you don't count the knock-on scheduling nightmare. On outsourcing/sacking the experts - "This had nothing to do with it. They were local systems manned by local staff" - I'm afraid at this point I may have yelled something at the wireless impugning yer man's truthfulness and parentage.
So far, it's been 'a power fluctuation' that 'disrupted the network in the data centre' for a short while, and when the network came back up 'the messaging' wasn't working. Which, assuming some correlation with what actually happened, are three things that should have redundancy or resilience in themselves, and furthermore shouldn't then trigger the next one in the chain. I'm going with poor design, poor implementation, poor testing, poor ops and nobody left who understands the whole system - the latter especially so, because the prior four factors won't exist if there's someone in charge who knows how things work (and has the authority to make them so, but that's another story...)
As for who's ultimate responsible for this, it's the board. Half the stuff I've written for CIO consumption has been about how to make a case to the board for the resources and buy-in to do things properly, because y'know, IT is just a cost centre and why the hell are we spending all this money on it when Mrs Chairman can use Google for free? (I wish that wasn't an accurate parody, but...)