31 August, 2015

The Spam Solution

What's the solution to spammers, telemarketers, conmen and trolls?

What's the best way to deal with telemarketers? It is NOT by having a blacklist, screening calls or a do-not-call registry. This is the most basic solution, one that should be available everywhere in 2015 but sadly isn't. But what's wrong with this? It just allows them to concentrate their efforts on those who aren't blocking them. Oh that's not your problem you say? It still congests networks, and you have to pay (e.g. some of your credit card fees go to combating fraud) indirectly. Also, once enough people have blocked them, they'll concentrate their efforts on defeating the filters or flat out ignoring the registry. It's like trying to fix a dripping ceiling by painting it - you're just diverting the problem to a less obvious symptom.

So what's the solution? The solution is to occupy them. Fruitlessly. Some people pretend to be interested and lead them on for hours. You don't have to keep talking to them, you can simply ask them to wait while you look for a pen, or make them repeat everything. We have to take this one step further and automate it. A simple AI with prerecorded messages:
Hello
(wait for pause)
Oh that sounds great.
(wait for pause)
Can you clarify that for me?
(wait for pause)
Hang on, I need to write that down. Let me find a pen.
(30 seconds of rustling)
Ok I-
(pen dropping, 1 minute of rustling and footsteps)

This will tie them up and prevent them from hassling others. What does this have to do with trolls and online spam? The point is, just blocking and banning them is a cat-and-mouse game. We need to keep them busy.

At this point, I need to introduce the concept of phasing, as used in World of Warcraft. Phasing means something appears different to different people. We need to apply this to trolls. Do you see where this is going? Let them post. Let them do whatever they want. But only they can see it. To them, their junk looks like normal, but it doesn't appear to anyone else. A global ignore list, if you will.

This will work for all kinds of things. Spam, trolls, people trying to start arguments, even people trying to game voting systems. Oh but they can log out and see, as the public, that their works are gone. This needs to be coupled with IP monitoring to work. Hey, ad companies already monitor IPs to see if it's an unsuspecting customer before serving malware. If it's a server checking, they serve a clean ad. Even Wikipedia blocks anonymous users who misbehave, so it can be done.

Update 5 Jan 17
(wow, I thought I wrote this in 2016)
I found out there's already a term for this - shadow ban, or banning, or shadowban. Wikipedia even has a page: https://en.wikipedia.org/wiki/Stealth_banning

No comments:

Post a Comment