Stateless Ethereum must abandon the eth subprotocol (eth/65)

[pipermerriam]

Simply cross linking this post with my latest thoughts on the data retrieval problem as a whole:

The Data Availability Problem under Stateless Ethereum

Here is a write up on a more “middle-of-the-road” approach to mitigating this issue which I’m now referring to as the “Data Retrieval Problem”. Here I pose an approach which: Attempts to limit the ability to create stateless clients via “abuse/misuse” of the on-demand state retrieval primitives which are aimed at facilitating full node syncing. Greatly improve the overall efficiency of the network with respect to syncing the state (unlocking the data prison) Facilitate the creation of statele…

Note that the proposed solution doesn’t solve the problem that a large number of stateless clients on the network will cause, but it appears we may be able to mitigate the ability for stateless clients to abuse the ETH protocol.

I’m still in general agreement with this post: that the ETH protocol likely ends up needing to be abandoned, or changed drastically in order to support stateless clients.