How I do Email

July 30th, 2010
email, statmbx, tech
My email setup is kind of weird. I like having it be entirely under my control, so that if I don't like something I can fix it. This means using open source software and a shared server. The configuration:
  • Email accounts at alum.swarthmore.edu, sccs.swarthmore.edu, cs.swarthmore.edu, bidadance.org, and others all forward to a unix server at swarpa.net run by some college friends.
  • This computer runs procmail to sort email, separating out spam with spamassasin, high traffic mailing lists into their own folders, stuff I want to archive without reading, etc from my main inbox. This also generates a procmail log that lists all incoming email
  • To do email notifications I keep a terminal window open running wmail which is an alias for tail -f ~/procmail/pmlog | python ~/clean_pmlog.py. The python program here is described in the post text only email stuff.
  • To check what messages I have waiting for me without opening my inbox I run cmail, which is an alias for the c program statmbx.
  • To actually send, read, and reply to email I run mutt. This is highly configurable, but it is the least flexible part of my setup if I want it to do something no one else has thought to make it do. Possible, as I can change the source, but difficult.
  • Mutt invokes my preferred editor, emacs to actually compose email.

Comment via: facebook, substack

Recent posts on blogs I like:

It isn’t always evil to ally with Steve Bannon

Scott Alexander wrote:

via Thing of Things October 19, 2026

Was the HuggingFace hacking incident scheming?

In July 2026, internal OpenAI agents hacked HuggingFace (HF) in order to figure out how their evaluation grader works. This involved a thousand agents coordinating on an improvised message board to carry out an elaborate sequence of exploits. A later grou…

via Victoria Krakovna October 18, 2026

There's no point at which turning your brain off will work

In early 2025, I started seeing people turn off their brain as they use LLMs1. They would have an LLM take an action (summarize text, write some code, etc.), and just assume that it worked2. This generally didn't work in early 2025 and the result was …

via Posts on September 18, 2026

more     (via openring)