hmmm. i don't know about everyone else, but 100% of my users use SMTP
AUTH if they want to send email out.
How do you verify that?
How can you know they do not input their email address else where?
One thing all ISPs could do to help is do SPF before sending as Meng suggested
today. Then you change a potential silent delete into a send fail. I found
that to be a good idea, if every SMTP would implement, then that could be a
good way to stop the unpermitted sending, but will customers like it if they
can not send email the way that is easiest for them? Too many send failures
might lead to a lost customer.
Perhaps you can hope to know by complaints of lost sent email if you enabled
"-all"?, but when did senders complain to their own ISP, they complain to the
receiver....and if you only have 50 users and "for every 1 complaint there are
100 more who did not complain", then how would you know statistically?
And what sort of ISP is this? What demographics?
When you set up an account, you
just tell them what to do.
Tell me what to do with my Eudora 4?
Upgrade first correct?
About the greeting card. I agree, I haven't had any luck deciding what
looks like spam or not. People complain when they get spam, and they
complain when they don't get spam, IMHO.
AccuSpam knows a little bit more (than what you could gleem for content
analysis), because it correlates not only votes on domains, but correlates how
much users agree on the recipient user's votes.
APOP was back in the gay nineties.
I have not even graduated to POP over SSL yet.
Maybe this forum needs to collect some stats on the internet before it blindly
advises "solutions"