Okay, message received. Bad idea, but you want to hear an ever worse one that could possibly work and not require any changes to the current MagiCoin infrastructure?
Hand-approved pool registrations in conjunction with pool-side hashrate limits, where users would need to disclose their estimated hashrate upon registration. That way, if a user goes over their registered hashrate (or if their registered hashrate is exactly the pool limit, then the pool limit), then they will be kicked from the pool. Sure, it adds more work to the pools, but it ensures the fairness of the rest of the miners who are actually following the rules. You could just solo-mine with a botnet, but at least that's less unfair to pool miners because you're not invalidating the blocks of miners who did nothing wrong. You're just lowering the block reward for everyone on the network, which is still unfair.
So I've been a programmer for several years now, I am quite fluent in C, but not so much with C++ for the last couple of weeks, I have been going over the Magi code and the m7m algo in the hope I could suggest some ideas and fixes to the dev, problem is without knowing the exact method the attacker is using, its hard to suggest a fix. How the attacks are happening, I would expect is being kept very secretive as if this information was to make it into the wild, then many crooks would be jumping onboard and hitting the network hard.
Aside from building wallets and custom miner code, I have been looking at starting a pool.
I'm not sure if I will yet, but if I did I would be enforcing strict hash rate limits, a place to answer queries and complaints from pool members. Email and or support tickets. One of my biggest problems with mining pools, has been little to no communication or reply from pool owners.
Another feature that I have thought about, could be a weekly random draw with prizes in XMG. The prize fund would come from 100% of donations. All who donate would automatically be in the draw.
I have a wealth of experience with running servers, web applications, including CMS portals, databases, forums, live chat, support ticket systems.
If I do go ahead with this, then I will obtain a paid SSL cert from a trusted CA, a reliable dedicated server, rather than a shared VPS. Low server latency would be key.
Maybe if I got a lot of interest from sub 150Khps miners who are sick of the whales killing the pool rewards, I could get something started. If mining could be more rewarding for people with lower hashrates, maybe It could give Magi a small boost in popularity and value.
Just thought I would put this idea out there for opinion.