While I am generally in sympathy with your (David Merrill, et al.) statement and your criticism of constructo-babble, I can't agree with everything.
Although ID can draw on science, it is not and cannot be a science itself.
I will argue by analogy. The design of cars is based partly on science. Factors such as weight strength, power, etc., are a part of the design (same would apply to other design fields such as architecture), but science cannot tell you to design a Cadillac. Nor can it tell you whether a Cadillac is better than a Lincoln (or a Jeep). What a car should be is determined by "local" factors, such as the state of the economy, availability of materials, availability of skilled workers, fashion, intellectual trends, new ideas, and other stuff.
Instructional Design is, if anything, more problematical than car design. People's expectations change and they get bored quickly. A lesson that looked very effective last year will fail this year because students will regard it as "old-fashioned" or something. That lesson might still work if students would just be more charitable, but unfortunately, as the Shah of Iran once said, the world is not a place of charity.
I would like to see more incorporation of science in ID, but that science which does exist is rarely taught. The log-log power law of practice predicts how long it will take to achieve a certain level of skill. You'd think this would be useful.
Instructional science, which hardly exists, cannot be a subset of ID. It might be a cousin or an uncle. The closest I have seen to a statement of instructional science is Eckel's Instruction Language, but probably no one has read this from cover to cover, except me (and you, of course). Since instruction is a form of dialogue, instructional science should really be based on linguistic discourse theory. Problem is, discourse theory isn't very robust at the moment. I would say that instructional science has about as much chance of succeeding as political science.
I agree with your important distinction between learners and students, but even you fall into the constructo-babble fallacy, of claiming that learners construct their meanings. I have tentatively agreed with Lloyd to produce a paper demonstrating the error of this line of thinking so I will not elaborate here. You should retreat from this position before it is too late.
You should give ample consideration to the principle of analysis by synthesis. First build something that works well. Then explain it scientifically. The Wright Brothers didn't understand all that the were doing. But they did know how to fly, because they had built and flown many gliders. They were probably the best pilots in the world at that time. That knowledge did not come from science, although it came from careful empirical experimentation.
Warning to constructionists: since it is a principle of yours that people "construct their own meanings" and thus have internal systems of truth, you cannot claim that I am wrong without contradicting yourself. To say that I am wrong is to judge me by some external (i.e., non-constructed) standard of truth.
Looking forward to your indignant reply.