These are topics that should go in the paper somewhere. This can be thought of as a checklist to ensure we don't miss anything. * DEFINITION OF REPLY BLOCK!!!! * why we use IDA instead of just duplicating everything. * Principle of survivability/robustness over accountability - if it keeps working, we don't need to know who's attacking * Rubber Hose Cryptanalysis and Eternity's answer to it - do we lose on rubber hoses compared to Eternity because our info expires? - no, because we expect others to republish it even if the original author dies - and this doesn't make it content-based in the same way Freenet is * "Ideal model" for anonymous publication * lots of others? roger's notes for last minute: make sure receive is spelled correctly (not recieve) make sure all references have url's (ps or whatever) if we can manage it. the 'trust' section needs a string of references similar to the commchannel section, so we look like we've done our due diligence i want the phrase 'trust algebra' in the paper somewhere towards the end make sure we mention freenet in the intro, in big letters