Download websites that use Mailman

About Mailman

Mailing list software is mail infrastructure, and mail infrastructure normally leaves nothing behind that a web crawler can read. Mailman is an exception, because it ships a web side: subscription pages, list information pages, administrative screens and public archives, all served over the same port as the rest of the site.

Since the web interface is the part we can see, the count measures something narrower than deployment. A list server whose web pages are closed to the public, restricted to an internal network or placed behind an authentication wall does not appear at all. As of October 2026 we detect Mailman on 25 sites, and the honest reading is hosts that publish a list interface openly, which is a subset of installations rather than a random sample of them.

Our record files the software under Email, and the label is a poor fit worth flagging. What it actually runs is discussion lists, announcement lists and working group correspondence for universities, standards bodies, community projects and public institutions. Campaign sending, audience segmentation and open tracking, which is what the category normally implies, are not the point of it. Anyone filtering our data by category alone will pull it in alongside commercial newsletter tools and end up comparing unlike things.

We hold no vendor for the software, which is what a community maintained project looks like in a field built for companies. Written in Python and distributed under GPL v3, with our earliest record at 1998, the software predates almost every hosted newsletter service a site could rent today. An organisation still running it has made a choice: keep the lists, the archives and the subscriber data on hardware it controls, and pay for the administration in staff time rather than in a subscription.

The signal is therefore about the operator more than about the mail. A domain carrying it usually has technical staff in house, self hosts other services too, and holds correspondence going back far enough that the archive is an asset in its own right. Hosted list providers looking for migration candidates, security researchers auditing an exposed administrative surface, and archivists mining public discussion history all want the same thing from us.

What they want is the enumeration. A per domain lookup confirms one host you already knew about. The question that pays is which hosts across the whole web still publish a list interface, and answering it needs the domains up front, in a form you can filter, sample and revisit later to see what has gone quiet.