I want to first comment on the negative space discussion, which is very interesting.
The question of copyright's negatives is closely related to the question of meta-IP, which also no one understands. Why are some things protected by copyright? Others by patent? Others by nothing at all?
One theory is maybe something like a historical public choice theory. In other words, behind every grant of an IP right lies (1) a group who wants protection against misappopriation, (2) organized enough to press its claims, (3) without powerful opposition. Where that pattern has existed over time, you find a trail of intellectual property rights.
This leads to a highly path dependent, or founder's effect driven shape to the various IP domains. Copyright begins by dealing with one problem (books), makes up rules for books, and then moves on to everything that seems analogous. Patent begins with things like wrenches, makes up rules, and then goes onward from there. Under this theory there's no use looking at the nature of things, as opposed to the interest groups and litigants who were in a position to ask and get protection.
Maybe by analogy, if we were studying a tree that grows in Africa, if someone asked, why don't we find this tree in South America, the answer is that the organism didn't quite get there, and instead another tree is growing where it might have.
That explanation may be one of the more accurate, but perhaps less fun. I think there's alot to think about in Clarisa Long and Henry Smith's approach to these problems, which is to look at the information costs created by the items in question and the IP regimes in question. So, for example, (as Mike pointed out), copyright has a tendency to cling to physical things that, well, can be easily copied. Patent, at least traditionally, was stuck to the physical ebodiment of an invention. So some of the negative and positive spaces in IP might be helpfully understood this way.