To summarize, I think the line of evidence you mentioned is NOT enough to establish an intelligent designer. Either additional evidence needs to be found (and Brad seems to have some ideas what that might be), or they need to be able to establish some properties of the designer. Otherwise their hypothesis is not unique, and the same result can be explained both by an "unknown intelligent designer" and an "unknown CI algorithmic generation procedure."
Right, and this is what makes it difficult to take ID seriously as science. ID seems to rely on assertions that certain problems are "impossible to solve" without direct manipulation by an intelligent agent. For the IDist, it's all or nothing.
IDists routinely present (pseudo)statistical calculations of the impossibility of this or that biological event. Leaving aside the biological and mathematical errors of such specific calculations
for the moment, the premise behind them reveals a structural flaw in the ID line of argument.
For example, in computer science, there are known to be problems that are not optimally solvable in finite computational time (NP-complete), but nevertheless algorithmic optimization procedures do exist to find suboptimal but "good enough" solutions in finite time.
e.g. the "traveling salesman" problem
http://www.tsp.gatech.edu/problem/index.html
IDists seem to completely miss the point that it is quite plausible that this is the mechanism by which evolution works, i.e. as some ensemble of optimization algorithms that produces "good enough" solutions in the form of genomes and biological structures.
So for the ID line of argument to "evolve" beyond an argument from incredulity, it must make positive, testable predictions about not only the nature of the designer, but about the precise mechanisms by which that designer intervenes, and at what times, to assemble genomes and biostructures.