I agree that this is exactly what he does. The question is whether this approach might lead to an explanation or whether it just begs the question. I'd say the latter.
The boundaries of science itself are indeed metaphysical, you're right. However, the theory would not have to deal with such boundaries, because anyone questioning the validity of the theory based on a problem with such boundaries is questioning the validity of Science, as a whole, not of any particular theory.
Not sure I quite understand you here but I think I agree. To explain consciousness requires crossing the epistemilogical and metaphysical boudaries of science, not simply producing a new scientific theory. (If that's what you meant). I suspect you see this as a argument aagainst any such 'non-scientific' theory, and from a scientific perspective it is. However it may yet be the only way to an understanding. Until science can define consciousness there is no secure footing from which anybody can argue that it has a scientific explanation.
Maybe we should worry about what he says about his own theory, rather than what other people have said.
Ok. As far as I can see he leaves out consciousness from his explanations. I find his writing convoluted in the extrame so I may have misinterpreted him. However I don't seem to be alone in concluding this.
Well, I didn't read the whole thing, but I did read his explanation of the hard and easy problems, and I'm moved to repeat what Edelman and Calvin and even Dennett, have been saying for so long: there really is no hard problem.
Ok, but you've got some healthy opposition.
If one can explain these things (and Chalmers seems to think we can):
the ability to discriminate, categorize, and react to environmental stimuli;
the integration of information by a cognitive system;
the reportability of mental states;
the ability of a system to access its own internal states;
I would say that, more simply put, Chalmer's position is that we need to expain the existence of 'what it is like to be'. As this is not a meaningful scientific description or definition he argues that science needs to consider redefing itself in order to include consciousness, and thus have a chance of explaining it. In a sense the 'hard problem' is finding a scientific definition for consciousness, for only if we can find one can we say that it is a scientific entity. Calling it physical is a sort of bulldozer of a strategy, and is completely ad hoc and counterintuitive. Many people, me incuded, cannot make sense of this hypothesis.