Open immanuelfodor opened 1 year ago
I think it could be useful to have this table, but to be honest I don't really want to maintain it. If we offer such a table, it should be accurate (would have to keep an on the other projects). Also users may be more comfortable trusting a table not maintained by one of the projects themselves.
Maybe we should just try to add the servers to https://en.wikipedia.org/wiki/Comparison_of_mail_servers?
To add an entry to that list on Wikipedia, the software needs its own Wikipedia article, to show it is "notable". Mox doesn't have an article, and I don't know what the rules are around creating one as author of the software. If anyone knows whether it is feasible to create a page (without it getting removed again), I'm interested in hearing about it.
I also don't know, let's keep this one open if anybody comes with knowledge around it.
Worth noting that most of the email server projects on github (including mailcow and modoboa) are really just wrappers/interfaces for old school complicated C/C++ binaries like dovecot, postfix etc. They're also mostly (including Maddy) under the GPL license, which isn't very liberal and puts a lot of restrictions on what you can do with it. Then of course there's all the hundreds of unfinished and not really usable implementations out there that would probably just get your mail domain blacklisted if you tried to run them.
Mox stands out as particularly cool because all the components are implemented in Go (which modern, performant and a lot cleaner and easier to read and hack on than 30 year old C code), it has all the key things needed to run a stable and secure mail server and it is under the liberal MIT license that makes it a lot easier to integrate with other things (and that hopefully means more people will be able to build upon it and contribute useful fixes as time goes on).
There is another Project https://github.com/stalwartlabs/smtp-server with similar goal, written in Rust and under the AGPL
I've came across #1 and instantly thought, we could generalize this and future comparison requests.
Maybe a table in the readme could provide some comparison to the other solutions, something like this:
Contributors could fill and extend the table based on their knowledge of the included tools.