2013-08-22 Came across the Fake MX trick. [0] That looks like a worthwhile idea. The amount of spam in the email system is really annoying. The spam filtering on my server uses more memory than the rest of the system together. This is not how it should be. Having a better approach, i.e. less resource-wasting especially through superior concepts, would be a delight. The related Tarbaby project [1] is a great idea, but there is a privacy problem included. I'd rather not participate. Also found the Brief History of Debian [2] today. It is really brief but it is good to have such a thing. When I've been work- ing on old software, I've often wished for such an explanation. Drawing the information out of (hopefully available) mailing list archives and the changelog is difficult, even more if you start with no idea what the project was like back then, even more if it was dead already before you started working on it. It is impor- tant to leave descriptions on the structure of the team and how the work was done behind. Besides that, describe the rationale of all (technical) decisions. As I believe I'd know what is important, do I already act accord- ingly? Concerning rationale for decisions, I think I do, but in describing my view on the history and structure of the projects, I should improve. [0] http://wiki.apache.org/spamassassin/OtherTricks [1] http://wiki.junkemailfilter.com/index.php/Project_tarbaby [2] http://www.debian.org/doc/manuals/project-history/index.html http://marmaro.de/lue/ markus schnalke