Dembski's problem is this, which he acknowledges in his third paragraph:
For ID to be scientific, it needs to be able to distinguish actual design from the illusion of design. Dembski claims that this can be done, but examination reveals only buzzwords and vague promises. He's constantly running up blind alleys:
What has kept design outside the scientific mainstream since Darwin proposed his theory of evolution is that it lacked precise methods for distinguishing intelligently caused objects from unintelligently caused ones. For design to be a fruitful scientific concept, scientists need to be sure they can reliably determine whether something is designed. (p 104-5.)
The problem, however, is that these well-defined methods are ... never defined. Dembski repeats that there are methods to detect design, but again and again these methods are revealed to be nothing more than, "If it looks designed, it must have had a designer."
As a theory of biological origins and development, ID's central claim is that only intelligent causes can adequately explain the complex, information-rich structures of biology and that these causes are empirically detectable. To say intelligent causes are empirically detectable is to say there exist well-defined methods that, based on observable features of the world, can reliably distinguish intelligent causes from undirected material causes. (p 105.)
That Dembski includes SETI, the Search for Extra-Terrestrial Intelligence, as an example of design detection, is surprising, because SETI researchers are emphatically not looking for patterns in radio signals from space. What they're looking for is a narrow-band signal — any modulation of a potential carrier wave is expected to have been smeared out with time and distance, so there's likely to be no information present.
Dembski elaborates on his favourite buzz-phrase, specified complexity:
Unfortunately he never elaborates on how to identify specified complexity, other than variations of "if it looks designed, then it must have had a designer."
Within the theory of intelligent design, specified complexity is the characteristic trademark or signature of intelligence. It is a reliable empirical marker of intelligence in the same way that fingerprints are a reliable empirical marker of an individual's presence at the scene of a crime. Design theorists contend that undirected material causes, like natural selection acting on random genetic change, cannot generate specified complexity. (p 106.)
At the top of this post I wondered if we were on to the big guns. Apparently not, for this pea-shooter consistently fails to fire:
That's what they claim, but they don't do it — not rigorously, or at all.
ID's chief claim is this: the world contains events, objects, and structures that exhaust the explanatory resources of undirected material causes and can be adequately explained by recourse to intelligent causes. Design theorists claim to demonstrate this rigorously. (p 107.)