I’m currently running postfix + dovecot + rspamd, and use a smarthost for outgoing mail (because some of the entities I need to communicate with would need me to be on an allowlist that costs thousands of dollars a year, and eff that). However, all three of those have various levels of slop infection, so I’m currently migrating away from them.
I will be using Aerogramme for imap & lmtp, opensmtpd (or Exim) for smtp, and my own milter instead of rspamd.
Postfix has slop? I have seen W Venema post about fixes for bugs that were found by AI… Also, I think only W Venema and V Dukhovni write Postfix, and they know the codebase inside-out, so I reckon that any assistance from AI would be well-vetted and well-understood by them.
It’s arguably not much of slop, indeed. I prefer projects that ban all kinds of AI contributions, bugreports included. This is my stance, and it will not be changing.
Exim does that, and would’ve been my first choice, but its configuration is way too complicated for my tastes. OpenSMTPD does not have an anti-AI policy, but I don’t see anything suspicious there, and it’s config & filter system is nicer than either postfixs or Exims.
Luckily for me, the smtpd is the easiest part of the system to replace, so if opensmtpd turns to slop, I can trivially migrate to Exim, or whatever else is available at the time with an anti-AI policy.
The comments here are somewhat pessimistic-looking for a complete anti-AI stance.
I suppose that since the baddies will use AI to find ways to abuse your code, you probably would want to do the same to be ahead of them? AI is here now (as are guns, for example), and unless the State declares a monopoly on the use of AI (like it does, in many countries, on the use of firearms), with an obligation to protect citizens from AI helped attacks, there’s going to be a difficulty here, I think.
Not trying to convince you, not at all. Just stating some thoughts of mine.
I’m currently running postfix + dovecot + rspamd, and use a smarthost for outgoing mail (because some of the entities I need to communicate with would need me to be on an allowlist that costs thousands of dollars a year, and eff that). However, all three of those have various levels of slop infection, so I’m currently migrating away from them.
I will be using Aerogramme for imap & lmtp, opensmtpd (or Exim) for smtp, and my own milter instead of rspamd.
score 1 · canonical
Postfix has slop? I have seen W Venema post about fixes for bugs that were found by AI… Also, I think only W Venema and V Dukhovni write Postfix, and they know the codebase inside-out, so I reckon that any assistance from AI would be well-vetted and well-understood by them.
score 1 · canonical
It’s arguably not much of slop, indeed. I prefer projects that ban all kinds of AI contributions, bugreports included. This is my stance, and it will not be changing.
Exim does that, and would’ve been my first choice, but its configuration is way too complicated for my tastes. OpenSMTPD does not have an anti-AI policy, but I don’t see anything suspicious there, and it’s config & filter system is nicer than either postfixs or Exims.
Luckily for me, the smtpd is the easiest part of the system to replace, so if opensmtpd turns to slop, I can trivially migrate to Exim, or whatever else is available at the time with an anti-AI policy.
score 1 · canonical
The comments here are somewhat pessimistic-looking for a complete anti-AI stance.
I suppose that since the baddies will use AI to find ways to abuse your code, you probably would want to do the same to be ahead of them? AI is here now (as are guns, for example), and unless the State declares a monopoly on the use of AI (like it does, in many countries, on the use of firearms), with an obligation to protect citizens from AI helped attacks, there’s going to be a difficulty here, I think.
Not trying to convince you, not at all. Just stating some thoughts of mine.
score 1 · canonical