This is a very interesting compilation of various approaches to the cryptographic time-lock problem. The Bitcoin idea is particularly enticing, even if impractical, since the idea of using the Bitcoin blockchain as a relatively tamper proof "timer" feels like something that could be made to work in practice.<p>One gap though is that the opening example is the idea of creating an insurance file that is publicly decryptable at a certain date if a particular real-world condition exists, like the file's author being kidnapped or whatever. But to make that work in practice you need not just time-lock encryption, but <i>resetable</i> time-lock encryption where it's possible to delay the release of the time-lock if that condition has not been met. I can see why the examples don't tackle that since even the basic problem seems unsolved, much less a tunable version of it, but it's a very interesting thought exercise.