The author mentions that anyone who can update the software can easily create a fake node that appears to have completely legitimate channels to one or more real nodes, thereby making SBN worthless. They suggest that to make proof-of-uptime expensive, SBN should charge higher than current feerates for premium reliable service. The author also suggests that adding more OTS-like servers may still allow a minority OTS-server to effectively censor users. Therefore, the author proposes groups of independent citizens put funds in a k-of-n address and continuously spend that k-of-n address to itself on every block. This would require a complicated setup but would ensure the generated R indeed commits to the Merkle tree root of the group uptime proofs. The author concludes by mentioning that although this approach would use roughly the same technologies as SBN, it would be distributed, and follow the principle of risk-sharing at the cost of increased blockchain usage.