Rufus Pollock, a researcher at Cambridge University, has produced a rather brilliant but difficult, because formal and mathematical proof (whilst yet embodying many empirical propositions), that the term of copyright should be much lower than it is now, and that it should in general become shorter as technology advances. He suggests that the optimum term of copyright should be about 15 years from creation. The argument is complex, but he provides some neat informal guidance vis:

……..consider the situation with respect to books, music, or film. Today, a man could spend a lifetime simply reading the greats of the nineteenth century, watching the classic movies of Hollywood’s (and Europe’s) golden age or listening to music recorded before 1965. This does not mean new work isn’t valuable but it surely means it is less valuable from a welfare point of view than it was when these media had first sprung into existence. Furthermore, if we increase protection we not only restrict access to works of the future but also to those of the past.
As a result the optimal level of protection must be lower than it was initially in fact it must fall gradually over time as our store of the creative work of past generations gradually accumulates to its long-term level. Forever Minus a Day?…

Pollock does not consider the related question: how will the efficient and optimal pricing of information services respond to changes in technology which reduce barriers to access? As more information becomes available how should a commercial information supplier i.e. a publisher, price his subscription services? How much should be given away and what to charge for that which is sold?