注意TrueCrypt/VeraCrypt的隐藏卷不靠谱(来自cryptsetup的FAQ)
https://gitlab.com/cryptsetup/cryptsetup/wikis/FrequentlyAskedQuestions#5-security-aspects 太长不读版本:https://security.stackexchange.com/questions/135846/is-plausible-deniability-actually-feasible-for-encrypted-volumes-disks
“5.2”和”5.18”详细描述了”隐藏卷”策略如何不靠谱, 大致意思是说: 隐藏卷在现实世界中面对不人道的怀疑(有罪推定)的刑讯逼供时根本没用, 因为你无法证明"在随机数据中不存在隐藏卷, 只要怀疑你, 就会不断对你刑讯逼供, 不论是否真的存在隐藏卷. 而且隐藏卷本身技术上也没法做到"隐藏", 本质上还是"加密成伪随机". 也就是说隐藏卷能糊弄得了外行, 但糊弄不了内行, 尤其是内行+刑讯逼供+有罪推定, 更是没用. 本人翻译和理解水平均有限, 我甚至自己都无法保证是否理解正确, 准确详情需自己查看, 同时也欢迎各位指点.
truecrypt, veracrypt, cryptsetup, faq
The natural processes of the Windows operating system, as well as applications like Microsoft Word or Google Desktop, can leak significant information outside of the deniable volume. For example, lists of recently changed documents, audit logs of recent file actions, and data saved by application programs can all serve to in- hibit deniability. As our results suggest, any DFS will not only have to encrypt and hide data — as file systems like TrueCrypt do — but must also erase any traces of that data left by the operating system through normal operation.
不要老是想要搞個大新聞，把 hidden volume 批判一番！說什麼已經欽定了，VeryCrypt 不靠譜。 豬隊友這種事兒，能怪 VeryCrypt 🐴？