4 ms·
Distributed Proof-of-Work is not vulnerable to this: https://org.saito.tech/eliminating-51-attacks-in-proof-of-work-blockchains/ https://org.saito.tech/elimina
by trevelyan 6y ago
Distributed Proof-of-Work is not vulnerable to this:
https://org.saito.tech/eliminating-51-attacks-in-proof-of-work-blockchains/ https://org.saito.tech/eliminating-51-attacks-in-proof-of-wo...
The community of users simply attaches harder cryptographic proofs to their transactions, forcing up the cost of producing empty blocks. The approach is capable of bankrupting nation-states as a side-effect of eliminating 51% attacks.
- nine_k 6y agoWhy won't it make transactions prohibitively expensive? Each transaction needs a trustless agreement of many, many nodes.
- Vecr 6y agoNot unless the block sizes down, or the block times/ block/transaction relaying latency go way up. The great firewall of China might cause some issues, but in theory it should not get worse.
- trevelyan 6y agoYou don't need a transaction to be routed by many nodes: how quickly it propagates (and thus how quickly it gets included/confirmed) just depends on how much hashing you do before sending it. The halving penalty on work just forces attackers (who have +1 routing hop compared to honest nodes) to double the work of those honest nodes if they wish to outcompete them in producing a longer chain (i.e. no 51% attack). It also allows users to respond to attacks in real-time by affixing harder proofs to their transactions, increasing the attack-cost on the network in minutes should anyone try to pull off the sort of gambit described in the linked page.