While I'm mulling over Sonia's many thought-provoking points, let me return to one of Lee's. As she points out, “hardwired servitudes” (technological protection measures that prevent forbidden uses) are an interesting test case that I don’t address in The New Servitudes. Lee suggests that “[t]hese built-in constraints might be viewed as problematic, but not because of the time and remoteness concerns we would typically associate with running servitudes.” I’m not sure about that.
Technological protection measures can in fact impose restrictions on everyone who encounters them--including people who are remote in time and space from the entity that imposed the restriction. The problems of obsolescence and non-negotiability that I associate with "running" and remoteness in my previous post therefore seem pretty serious here. And to the extent these restrictions impose external harms on third parties, the fact that the technology makes the restrictions self-enforcing (and therefore presumably more enforced—that’s the idea anyway) could magnify those harms. (This is different from the mechanism by which "running" magnifies third-party harms, but the consequences could be similar.)
On the other hand, there is one way in which TPMs seem in practice to solve the type of information cost problem I associate with servitudes. I haven’t yet studied this systematically, but recent controversies over TPMs (copy-proof CDs, non-interoperable digital music files, etc.) suggest that the (relatively) airtight enforcement effected by TPMs makes them quite noticeable. Consumers might go for years without noticing that the terms of a written EULA forbidding behavior they would like to undertake. But when a company distributes copy-proof CDs, at least some consumers notice and quickly spread the word (which has led in some cases to the commercial failure and abandonment of unpopular TPM schemes).