That is exactly what I'm thinking. Just like when the press reports the cause of a fire to be "electrical" when in fact it was due to an overloaded 18/2 extension cord. That's not an electrical fire. It is a fire caused by carelessness.
Well, actually, that 18/2 extension cord is an electrical fire. That is the way that the statistics are gathered. An electrical appliance that starts a fire is also an electrical fire.
A long time ago, EC&M, I think, ran an article that was titled" All Those Electrical Fires, Hogwash." In the article, the author referred to a fire report that stated the cause of the fire could not be determined because the building wasn't wired.
Few things are harder to put up with than the annoyance of a good example.
Alan, there is absolutely no requirement to make any manner of ground or bond between the electrical system and the metal siding.
As our code reads, only metal 'likely to become energized' must be bonded, and most often this is done within an appliance.
Now, our meter cans are metal, and household siding, metal roofs, and all manner of things are nearly impossible to instal without there being some manner of path to ground.
I've seen some pretty impressive scorch marks result when a service mast breaks, and the results of some pretty severe arcing when a meter base fails. I don't doubt that a fire COULD be caused this way.
Yet, I suspect that the news story has left out some information, and we're not getting the full story.