really, shadowbanning as an antispam measure is indefensible; spamassassin is almost entirely transparent and it works excellent
Conversation
it really depends on the adversary - when spammers evolve instead of moving on, you'll want to keep your methods secret
2
now I can't cite any case studies but from my own antispam experience and from watching gmail and other large providers
1
I run nothing except spamassassin in a default config and I leave my email everywhere. I get lots of spam... all filtered
1
Now you may claim this is atypical but I would like to hear why, for one
2
Replying to
nobody is trying to poke holes in your system specifically, actively extracting its filter rules
1
also note that my experience is from web antispam, not mail antispam
1
"filter rules" in a loose sense - this is when I wish this was IRC and I wouldn't feel like I needed to save chars
1
also you can blackhole email effectively and not give spammers an information backchannel, but anything gmail-scale has leaks
Replying to
and this is only email; meanwhile website users expect low response times and websites leak information like buckets

