1 public-inbox design notes
2 -------------------------
4 Challenges to running normal mailing lists
5 ------------------------------------------
8 2) bounce processing of invalid/bad email addresses
9 3) processing subscribe/unsubscribe requests
11 Issues 2) and 3) are side-stepped entirely by moving reader
12 subscriptions to git repository synchronization and Atom feeds. There's
13 no chance of faked subscription requests and no need to deal with
14 confused users who cannot unsubscribe.
16 Use existing infrastructure
17 ---------------------------
19 * public-inbox can coexist with existing mailing lists, any subscriber
20 to the existing mailing list can begin delivering messages to
23 * public-inbox uses SMTP for posting. Posting a message to a public-inbox
24 instance is no different than sending a message to any _open_ mailing
27 * Existing spam filtering on an SMTP server is also effective on
30 * Readers may continue using use their choice of NNTP and mail clients.
32 * Atom is a reasonable feed format for casual readers and is supported
33 by a variety of feed readers.
38 * Freedom from proprietary services, tools and APIs. Communicating with
39 developers and users of Free Software should not rely on proprietary
42 * Existing infrastructure, tools, and user familiarity.
43 There is already a large variety of tools, clients, and email providers
44 available. There are also many resources for users to run their own
45 SMTP server on a domain they control.
47 * All public discussion mediums are affected by spam and advertising.
48 There exist several good Free Software anti-spam tools for email.
50 * Privacy is not an issue for public discussion. Public mailing list
51 archives are common and accepted by Free Software communities.
52 There is no need to ask the NSA for backups of your mail archives :)
54 * git, one of the most widely-used version control systems, includes many
55 tools for for email, including: git-format-patch(1), git-send-email(1),
56 git-am(1), git-imap-send(1). Furthermore, the development of git itself
57 is based on the git mailing list: https://public-inbox.org/git/
58 (or http://hjrcffqmbrq6wope.onion/git/ for Tor users)
60 * Email is already the de-facto form of communication in many Free Software
63 * Fallback/transition to private email and other lists, in case the
64 public-inbox host becomes unavailable, users may still directly email
65 each other (or Cc: lists for related/dependent projects).
70 * git is distributed and robust while being both fast and
71 space-efficient with text data. NNTP was considered, but does not
72 support delta-compression and places no guarantees on data/transport
73 integrity. However, a read-only NNTP gateway is implemented.
75 * As of 2016, git is widely used and known to nearly all Free Software
76 developers. For non-developers it is packaged for all major GNU/Linux
77 and *BSD distributions. NNTP is not as widely-used nowadays.
82 * Perl 5 is widely available on modern *nix systems with good a history
83 of backwards and forward compatibility.
85 * git and SpamAssassin both use it, so it should be one less thing for
86 admins to install and waste disk space with.
88 * Distributing compiled binaries has higher costs for storage/cache
89 space is required for each architecture. Having a runnable,
90 source-only distribution means any user already has access to all
96 * Stick to dependencies available in Debian main, this should make it
97 easier for potential users to install, and easier for distro
98 maintainers to pick up.
100 * A list server being turned into an SMTP spam relay and being
101 blacklisted while an admin is asleep is scary.
102 Sidestep that entirely by having clients pull.
104 * Eric has a great Maildir+inotify-based Bayes training setup
105 going back many years. Document, integrate and publicize it for
106 public-inbox usage, encouraging other admins to use it (it works
107 as long as admins read their public-inbox).
109 * Custom, difficult-for-Bayes requires custom anti-spam rules.
110 We may steal rules from the Debian listmasters:
111 svn://anonscm.debian.org/pkg-listmaster
113 * Full archives are easily distributable with git, so somebody else
114 can take over the list if we give up. Anybody may also run an SMTP
115 notifier/delivery service based on the archives.
117 * Avoids bikeshedding about web UI decisions, GUI-lovers can write their
118 own GUI-friendly interfaces (HTML or native) based on public archives.
123 * Getting users to install/run any new tool is difficult.
124 The web views must be easily read/cache/mirror-able.
126 * There may also be a significant number of webmail users without
127 an MUA or feed reader; so a web view is necessary.
129 * Expose Message-ID in web views to encourage replies from drive-by
132 * Raw text endpoint allows users to write client-side endpoints
133 without hosting the data themselves (or on a different server).
135 What sucks about public-inbox
136 -----------------------------
138 * Lack of push notification. On the other hand, feeds seem popular.
140 * some (mostly GUI) mail clients cannot set In-Reply-To headers
141 properly without the original message.
146 See the public-inbox-v2-format(5) manpage for all the scalability
152 Copyright 2013-2019 all contributors <meta@public-inbox.org>
153 License: AGPL-3.0+ <http://www.gnu.org/licenses/agpl-3.0.txt>