A rolling epoch, as suggested in the paper, not only helps keep the blockchain length tractable it also prevents "submarining" from assets which haven't moved since the early days of a blockchain (e.g., Satochi/Finney and new blockchain pre-mining).
On Mon, 2 Jul 2018 06:49:04 -0700
Steven Schear <schear.steve@gmail.com> wrote:
> In 2013, a paper I contributed to offered a solution to the ever growing
> blockchain delema: a finite epoch. The solution is similar the one Chaum
> used on Digicash. It would fix, temporally, the blockchain to include only
> transactions for the past 2 years,
and what happens to funds older than 2 years?
"Any money still held from transactions in these blocks would be freed up, and released back to the network in the form of a lottery."
lolwut - that's trolling, right? =)
> for example, thus creating a blockchain
> of tractable and predictable size for affordable full nodes.
>
> https://www.coindesk.com/bitcoin-activists-suggest-hard-fork-to-bitcoin-to-keep-it-anonymous-and-regulation-free/
>
> More recently, I've co-written a paper proposing a distributed service
> solution that could solve thin wallet privacy and security issues without
> needing to trust individual full nodes under the control of others.
>