The Deepwater Horizon accident is a terrible lesson in not testing backup plans.  Yes, it is probably impossible to test the various way of trying to stop a well under exact conditions (i.e. with a gushing well), but the constant refrain of “this has not been tried at this depth” is deeply disheartening.  That’s like hearing from your operations team that they have only ever practiced restoring a single development machine and never anything in the production environment.  At this point I am with former Labor secretary Reich – the government should put the US operations of BP into temporary receivership and take complete control of the cleanup effort.  I have lost all confidence in communications emanating from BP and would rather see the US Navy in charge.

I certainly hope that we draw the right lessons from this disaster, but I am fairly skeptical about that.  A similar event that comes to mind is Three Mile Island.  Since then the US has not built any new nuclear power plants but is still relying on nuclear power to supply a meaningful chunk of electricity.  That has left us with the worst of both worlds.  We are now operating the existing nuclear plants which have relatively unsafe designs (compared to what researchers have come up with since) way past their original due dates.  Much like driving cars to very high mileage that means stuff breaks (I should know, having two cars with way over 100,000 miles each).  Only here it is radiation not radiators we are talking about.

The only really consistent lesson to draw from the Gulf accident is that we need to invest heavily and rapidly into alternatives to oil.  Personally, I am open to those alternatives including nuclear for base load in the grid, although it would be great if eventually we could do without that too.