4 * Bremner and dkg are co-hosting a BoF at [debconf](https://summit.debconf.org/debconf15/meeting/217/improving-privacy-and-security-for-notmuch-mail/).
6 * The meeting is Monday 2015-08-17, 1700-1800
8 * Video streaming should be [available](https://wiki.debconf.org/wiki/DebConf15/Videostream/Amsterdam)
14 Moving parts for secure e-mail
28 * wrong key selection during composition
29 * reply (message mode defaults)
31 * webmail authentication/authorization (muliple users?)
32 * webmail message escaping (XSS, etc)
34 * terminal escape sequences
37 ### usability as security?
39 * indexing encrypted mail
40 * Memory Hole protected headers
41 * key selection indicators during compositoin
47 * based on moving part
54 -------------------------
58 One of (at least my) primary motivations for working on Notmuch is reducing my dependence on cloud services, and supporting the secure sending and receiving of signed and encrypted mail. Like any realworld piece of software, notmuch is far from perfect, and several areas related to privacy and security could clearly be improved. During this BoF we'd like to plan out some topics to work on in followup hacking sessions. Anyone is welcome, even if they don't feel like hacking on notmuch. Potential topics of discussion andhacking include:
59 * S/MIME signatures and encryption
60 * Improving the security of the Emacs MML mime composer
61 * Searching of GPG encrypted mail
62 * Auditing and fixing "webbug" style problems in front ends
63 * Making notmuch build reproducibly
69 * S/MIME signatures and encryption
71 * integration with other keyrings
72 * signature only (easyish) versus encryption (more work)
73 * Improving the security of the Emacs MML mime composer
74 * automated "encrypt-when-i-have-keys-available" mode or other convenience functions?
75 * can an adversary force signatures based on quoted text?
76 * generate memory-hole-style messages
77 * Searching of GPG encrypted mail
78 * possible implementation mechanism: "notmuch reindex --with-filter=decrypt"
79 * Auditing and fixing "webbug" style problems in front ends
80 * can we instruct emacs to restrict all network access from notmuch?
81 * what other frontends might call out to the network?
82 * Making notmuch build reproducibly
83 * https://reproducible.debian.net/rb-pkg/unstable/amd64/notmuch.html
84 * Protect against spoofed signature verification?
85 * how do we deal with multipart messages where only a subtree is signed?
86 * are other sorts of spoofing possible?
87 * read and display memory-hole-style messages
88 * "safe" ways to display html parts (e.g. without text/plain alternatives)