# HG changeset patch
# User meillo@marmaro.de
# Date 1222975789 -7200
# Node ID 33149fbcac81bbf793c0ee88b25a6201e3cb0cbe
# Parent  563205a1b07e60e7ca7a4daca5b276cfd0c927bd
added mta-comparision and texts about masqmail

diff -r 563205a1b07e -r 33149fbcac81 docs/masqmail-design-goals.txt
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/docs/masqmail-design-goals.txt	Thu Oct 02 21:29:49 2008 +0200
@@ -0,0 +1,17 @@
+Design goals for masqmail
+=========================
+
+- be perfect for non-permanent internet connections
+- small in size
+- have a clear concept
+- follow the Unix Philosophy
+  - be a simple MTA
+
+
+- free software
+- transparent development
+- good documentation online
+
+- preferable portable C code
+- simple configuration
+- conform to RFCs
diff -r 563205a1b07e -r 33149fbcac81 docs/masqmail-security.txt
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/docs/masqmail-security.txt	Thu Oct 02 21:29:49 2008 +0200
@@ -0,0 +1,20 @@
+masqmail security
+=================
+
+masqmail is not intended to listen on a port open to the internet. Its normal
+operation is on workstations and listening only on localhost.
+Generally masqmail should only listen on a port accessable by only trusted
+users. Therefor a firewall should be set up to protect against attacks.
+
+Security is not a primary goal of masqmail, because its jobs is normally not in
+dangerous areas. But secrurity should always be a secondary goal, especially for
+everything that communicates with/via the internet. (And also for programs that
+run suid-root, like all mail transfer agents.)
+
+masqmail should be hardened in future! A common and good way to do that is to
+split it up in several programs, each doing one particular job with only the
+needed rights. This approach is taken by qmail for example.
+postfix took qmail as inspiration and is nearly as secure as it---in contrast to
+sendmail which implements a monolitic architecture.
+But monolitic architectures must not be bad in general. exim for example shows
+that a monolitic MTA can be secure---if it one cared about it.
diff -r 563205a1b07e -r 33149fbcac81 docs/mta-comparision.txt
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/docs/mta-comparision.txt	Thu Oct 02 21:29:49 2008 +0200
@@ -0,0 +1,81 @@
+MTA comparision
+===============
+
+Candidates:
+- sendmail
+- postfix
+- qmail
+- exim
+- masqmail
+
+- (exchange, ...)
+
+
+Author
+------
+sendmail: Eric Allman and others, now Sendmail Inc.
+postfix: Wietse Venema and many others
+qmail: Daniel J. Bernstein
+exim: Philip Hazel
+masqmail: Oliver Kurth
+exchange: Microsoft Corporation
+
+
+License
+-------
+sendmail: Sendmail License
+postfix: IBM Public License
+qmail: public domain
+exim: GPL
+masqmail: GPL 
+exchange: MS-EULA
+
+
+First release
+-------------
+sendmail: 1983
+postfix: 1999
+qmail: 1995
+exim: 1995
+masqmail: 1999
+exchange: 1993
+
+
+Lines of code (with sloccount on debian packages)
+-------------------------------------------------
+sendmail: 93k
+postfix: 92k
+qmail: 18k
+exim: 54k
+masqmail: 14k
+exchange: (no source available)
+
+
+Architecture
+------------
+sendmail: monolitic
+postfix: modular
+qmail: modular
+exim: monolitic
+masqmail: monolitic
+exchange: (unknown)
+
+
+Design goals
+------------
+sendmail: flexibility
+postfix: performance and security
+qmail: security
+exim: general, flexible & extensive facilities for checking
+masqmail: for non-permanent internet connection
+exchange: groupware
+
+
+Market share (by Bernstein in 2001)
+-----------------------------------
+sendmail: 42%
+postfix: 1.6%
+qmail: 17%
+exim: 1.6%
+masqmail: (unknown)
+exchange: 18%
diff -r 563205a1b07e -r 33149fbcac81 docs/rfc-list.txt
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/docs/rfc-list.txt	Thu Oct 02 21:29:49 2008 +0200
@@ -0,0 +1,4 @@
+Useful RFCs
+===========
+
+RFC 974: MX records