Donald Sensing thinks it possible that the North Korean bomb test wasn’t a fizzle (the prevalent theory), but that it was instead a successful proof-of-concept test for a bomb design. The North Koreans are probably short on fissionable materials and don’t have any to waste on showy tests. It’s a credible idea.
Another possibility is that the test was a complete success – of an attempt to fool Kim Jong Il into believing his country has a WMD program. North Korea is desperately poor. It is on the borderline of credibility that they could mount a successful nuclear research effort. But it is believable that Kim Jong Il has ordered such a program nonetheless – in fact, any other proposition here IS unbelievable – and that some poor bastard in North Korea is trying to make it come true. And probably failing – but where it is difficult to build a nuclear bomb, it is relatively easy to make a stack of conventional explosives – and to pass it off to your nut of a boss as a successful small-scale test.
There’s reason to believe it’s happened before; it’s almost certain that Saddam Hussein thought his regime had certain capabilities which in fact only existed in Potemkin fashion. This is a standing problem with certain forms of authoritarian governance. But sometimes the royal visitor says “that’s a lovely shop, let’s dash in for a moment”, and Potemkin is forced to go to more extreme measures. It’s possible that’s what happened here.