I forked the Polkadot SDK.
For those reading posts for the first time, i built identity infrastructure for polkadot and got ghosted by the technical fellowship, so shipped to Paseo anyway. XCM didnt let me teleport, and ondemand coretime would happily take my tokens but my parachain wasnt assigned a core so it never made blocks. Instead, i noted the overall pattern, and forked the network.
From there I culled heavily: the relay chain, XCM, the identity pallet, and a lot else that wasn’t needed. No more parachains. Removing the tech debt was its own project; the pallet dependencies were a spaghetti mess and untangling them was necessary before anything else could be done cleanly. Found a bug that was an escape hatch for validators to avoid getting slashed.
Governance has been replaced. One person, one vote. Built to be capture-resistant from the ground up, not retrofitted onto an existing system. Parachains are gone, the architecture is like this: anyone can run a node, the infrastructure binaries are hashed and replaced in case of drift for any reason. Repeated binary file modification gets you dropped from the network. We use my patented x.509 standardized fido2 complaintmethod to makes bots extremely expensive, as well as solving the proof of personhood issue with requiring silly tattoos. Again: 1 person, 1 vote.
The tokenomics are not extractive.
The treasury only can pay bounties on shipped code that is driving real network transactions. Contractors paid up front only never to be seen again is the oldest contractor grift in the book.
Its an identity platform first, but has a near frictionless path for developers to launch their own products.
The VM has been tuned. As of today it is outperforming Wei Tang’s microkernel (and polkavm, interpreter for interpreter on some benchmarks, by a factor of 2,000%+).
More to come.