Possibly stupid question, but why not stop e-mail spam in the same way we do IMs?
I don’t see how I could ever get spam-messages from, e.g. an xmpp account. Worst-case scenario is that I get a bunch of ‘subscription’ requests, and I can only add friends when I trawl through the requests, or if I know they’re adding a request at the time, then look out for that request.
Emails seem to let everything in, with a reliance on the admin to sort this out. Why not do the same thing?
Specifically, I’m thinking of writing a script:
- If this person’s in my contact-list, they’re cool.
- If they’re on the shit-list, they’re deleted.
- If not, they get into the ‘waiting room’.
… then set up a shortcut to put someone on the shit-list. So there’s no more ‘you’ve got mail’ notifications from random spammers, and I can review it once a week or so to pull the good-guys out.
Seems like a good idea, but then I wondered, why hasn’t this been done before? If the script works, it seems like someone could do the same thing with a GUI.
Spam detection is HARD to get right. How do you ensure your spam filter never has false positives? How do you know #2 on your list won’t cause problems later? And most people don’t have time for item #3, sifting through everything in the ‘waiting room’ which will never be empty.
Your system seems to implement a whitelist of people who would be even allowed to contact you. That goes against the fundamental “push” nature of email, if you see what I mean. Remember that just because an email is unsolicited, doesn’t mean it’s spam.
Same attitude as a blocklist, broadly. Once someone’s on the shit-list, you just need to be okay with those messages lying in the bin, unseen.
Yea, that’s what I’m going for. It seems to work for IMs. And if someone emails me from nowhere, offering wonderful things, I’ll get back to them late, but don’t have to get a ‘ding’ from all the random crap.