[bitcoin-dev] Proposal to replace full blockchain with recent history plus UTXO Set

[Dave Scotese <dscotese@litmocracy.com>]

[-- Attachment #1: Type: text/plain, Size: 1799 bytes --]

I've been working on an idea that relieves full nodes of storing the entire
blockchain. Open source software generally relies on the fact that "enough"
people agree that it's secure. Bitcoin software works that way too. So if
you understand enough to see that a UTXO set is valid at a certain block
height, and there are enough other people who agree and that set is
recognizable by humans, then we can use that UTXO set and ditch the
blockchain that existed up to that point. It would save a lot of storage
and make it a lot easier to run a full node.

Have you reviewed the source code from which your wallets were compiled?
At some point, we all trust third parties, but generally (at least among
people who understand Bitcoin) they are large composite groups so that no
small group or individual can profit from cheating.

I look forward to answering any concerns and also to any offers of help.   I
used block 542324 of the Bitcoin blockchain to make a memorable experience
using the game of life. I wrote a script for the open-source Game-of-Life
software Golly and shared it in the paste at https://pastebin.com/k5Ssc0qk.
It produces the image at https://imgur.com/a/rwIQuVz. If someone can tell
me how to get a UTXO Set from the bitcoin client, I'll send them $50 of
bitcoin. Then I could get the SHA256 hash of that set and try to make a
recognizable checkpoint for the Bitcoin blockchain. If someone runs Golly
and shares a video of the game playing out (into the apron-shaped image),
I'll send them $50 of bitcoin too.

In a few decades when the blockchain has grown to a few terabytes and the
UTXO Set is still just a few gigabytes, I'd like to see more people start
running full nodes without the hassle of a long wait and loads of storage
space. That's what stops me from running one.

[-- Attachment #2: Type: text/html, Size: 4593 bytes --]