Based off posts the past three years on this topic the behavior FoE is trying to mitigate via multi-accounting rules is pushing. Specifically pushing Great Build Forge point contributions. But notably, not goods trading from the same "internet connection". FP's is bad, goods are ok. OK whatever.
So it is working? Are there better ways to accomplish this that use data and mechanisms you actually control? No and yes, from what I can tell. You don't need to discuss the methods you are using because they are essentially public domain already. Inno is not doing anything countless other Internet Gaming and Marketing sites are doing. It's not a particularly large problem domain, and it is VERY well known, aka, public domain. IP and device ID based auth systems simply do not work. It is why those in my profession never use those techniques.
My point is, stop spending money using marginal and weak methods to mitigate behaviors you feel are "cheating", and instead spend them on techniques that much more effective. According to one of the answers here, Inno has already developed sophisticated usage pattern detection systems. So USE THEM, not this silliness that does nothing but drive legitimately honest players from the game and locks out an entire class of user.....thus costing Inno money! Wasted O/E dollars supporting weak authorization and lost revenue from an entire class of potential player.
Again, it simply makes no business sense.
- - - Updated - - -
For obvious reasons, we're not going to publicly discuss how we do cheater hunting.
But I don't really understand your point. You think that we should do away with the rule about letting people play multiple accounts, and focus only on pushing?
Basically, yes. Because the former is unenforceable and ineffective and does nothing more than to drive away a whole class of player. The latter is something that you can probably develop very effective means to mitigate.