I encountered a New York Times Magazine article about the "Hollywood Model" of employment, in which an ad-hoc team of specialists is assembled for a project and then dispersed on completion; the article suggests that this will become a more dominant mode of employment in the future. It's an interesting article, but I'm not sure I agree with the degree to which they predict the ascendancy of this model.
Meanwhile, at work, my colleagues and I have been struggling to resolve "integration bugs" in new products. These are the problems which arise when individual pieces of a product or design appear to work when tested on their own, but because specifications are pretty much always incomplete and architectures are imperfect, don't quite work perfectly when brought together in a system. Some such problems make themselves felt immediately, but others can be subtle and sneaky, surfacing only under extraordinary circumstances.
Specialists aren't good at figuring out integration bugs. They require a holistic understanding of a system, and a significant strength of my company is that we have a fair number of bright people who have been there long enough to have not only deep understanding of their own specialties and broad understanding of the company, but significant understanding of the systems we work on. The Hollywood Model doesn't seem to allow for that, and so is vulnerable to integration bugs.
I don't think that every kind of work lends itself to managing integration bugs on the fly or to specifying every element exhaustively, so I think there will be a place for the "Corporate Model", in which people form a structure which remains stable across projects, for a long time yet.