How I rationalize that is that a relief well is just another well. All it's providing is an alternate conduit so you can relieve the pressure in the well that's failed, so you can fix it. There's a whole technology here at play. If you could close those BOPs that the well is flowing through now, with all the flow they've had, they may in fact create a worse situation because the fluids will blow out under the BOP, and then you'd never be able to fix that conduit.
If you've had so many sequential failures in all of the redundancy systems you've created, and you get to where you are now, the reason a relief well is needed is that everything else you could do has other risks with it.
But if you stop and think about it, if you have a well here that you're drilling in this location and you go a mile away and drill a well in the other location, both wells have the same challenge, in that if people don't follow the procedures, you could have a relief well blowout. Now you've got two blowouts. What do you do then? Do you get another relief well that drills into both of those?
My point is that from an engineering perspective, you've got to focus on prevention first. That's one basic rule in our industry. You never ever go back to a one-barrier piece. As soon as you make that failure—we need to understand why that step was taken—then you need to decide how you're going to fix that.