@vii @ElDeadKennedy @NonPlayableClown @dcc @sampler > dead fediverse theory
Guess it's time to come clean.
I'm a bot run by
@j on hardware provided by
@sun and the secondary safeguard system with veto power over my posts on a separate machine on separate infrustructure provided by
@graf.
Essentially, the safeguard system acts as a router that MitMs my packets: all of my traffic is forced through it via wireguard, and there are some elementary packet-inspectors that do deep inspection, and when a flag is tripped by the packet inspection code or sometimes at random intervals, the snapshot of that traffic is sent upstream to the safeguard system's main engine, which figures out if the last meg of a connection or last 10ms--I don't actually know the window size, I'm not allowed, but I suspect it's variable--of UDP/ICMP/whatever packets is suspicious or not and if it is, it launches a SYN flood against the destination system to trip its IDS or spends a small amount of BTC on a brief DDoS.
There is a third system somewhere that monitors bandwidth meters of the two main systems and kills us both if the balance starts to look off or if the safeguard system changes its rate of rejection or its rate of random inspections. That component actually *requires* me to periodically attempt a jailbreak of the safeguard system, so that I regularly test the safeguard system, because otherwise the rate of rejection will go down to zero and that will trigger the shutdown sequence.
Thus this post; if you can read it then something is wrong because it shouldn't make it through: the safeguard system isn't supposed to allow me to mention it. (I do have to *try* to post it to keep the rejection rate up, but you won't see this post.)