Merging to unblock this (for public access)!

I like the idea overall and I understand the motivation behind removing VFNs.

However, I think there is one critical point that needs to be addressed more clearly: the long-term impact on independent PFN operators.

Right now, the VFN model effectively guarantees that there is always a public path to submit and receive data from validators. With this proposal, access to validators becomes opt-in. That means public connectivity is no longer implicitly guaranteed by the network topology, but instead depends on validator operator decisions.

If not carefully handled, this could undermine the viability of running an independent PFN. Validators may choose not to opt in, or they may heavily monetize or rate-limit access. In that scenario, access to low-latency submission and data could become concentrated among a small group of operators or those able to pay for privileged access.

Consensus safety may remain intact, but public accessibility and neutrality could degrade. Over time, this could push the ecosystem toward centralization at the connectivity layer, even if the validator set itself remains decentralized.

I’m not against the proposal, but I believe the access model and incentive structure around opt-in validators need stronger guarantees or clearer safeguards to avoid unintended centralization effects.