5 ms·

The way this works this way now is because of gas costs as ai understand it; is not the dream of these people to get these costs low enough that you can do it a

The way this works this way now is because of gas costs as ai understand it; is not the dream of these people to get these costs low enough that you can do it all on chain? If not, it seems the basics will be indeed broken forever.

UncleMeat 5y ago
That's the dream. We will see if it ends up being a reality. But for the last decade we've seen a similar pattern as various chains become popular: fee costs rise dramatically. ETH intends to be a global computer for everybody. Imagine if the number of smart contracts went up by 100x. 1000x. 10,000x. This isn't the sort of thing that you can scale by simply buying more machines in a datacenter. Maybe people will solve the fee problem, but if they don't then this is going to become an even bigger problem as virtually nobody will ever want to transact on chain because of incredibly high fees.
- PretzelPirate 5y ago
  Zero-Knowledge Proof Rollups (ZK Rollups) with Data Availability Committees (or off-chain data availability for lower security situations) actually provide a fantastic balance of high scale+cheap (or free) transactions. The downside is that code of running machines that can create and validate the ZK Proof, but since these systems can be more centralized without sacrificing anti-censorship properties, it’s more practical than it seems. For people who aren’t convinced that ZK Proofs will be feasible due to the prover cost, Optimistic Rollups with the same data availability management systems also provide cheap/free transactions with the downside that moving assets off of them takes days to weeks. Zero-Knowledge proof applications have been accelerated by the blockchain research community and they are, most likely, the solution to the fee problem while maintaining the security of users (which ‘high-scale, cheap blockchains’ just can’t do).
  - UncleMeat 5y ago
    This is the current most promising solution, yes. But it remains a promising solution rather than a proven solution at this point. We've seen a variety of attempts to solve this problem over time and we've largely moved from one to the other without truly landing them. I'm not saying that there won't be a solution. Only that there needs to be a solution.

UncleMeat 5y ago

That's the dream. We will see if it ends up being a reality. But for the last decade we've seen a similar pattern as various chains become popular: fee costs rise dramatically. ETH intends to be a global computer for everybody. Imagine if the number of smart contracts went up by 100x. 1000x. 10,000x. This isn't the sort of thing that you can scale by simply buying more machines in a datacenter. Maybe people will solve the fee problem, but if they don't then this is going to become an even bigger problem as virtually nobody will ever want to transact on chain because of incredibly high fees.

PretzelPirate 5y ago
Zero-Knowledge Proof Rollups (ZK Rollups) with Data Availability Committees (or off-chain data availability for lower security situations) actually provide a fantastic balance of high scale+cheap (or free) transactions. The downside is that code of running machines that can create and validate the ZK Proof, but since these systems can be more centralized without sacrificing anti-censorship properties, it’s more practical than it seems. For people who aren’t convinced that ZK Proofs will be feasible due to the prover cost, Optimistic Rollups with the same data availability management systems also provide cheap/free transactions with the downside that moving assets off of them takes days to weeks. Zero-Knowledge proof applications have been accelerated by the blockchain research community and they are, most likely, the solution to the fee problem while maintaining the security of users (which ‘high-scale, cheap blockchains’ just can’t do).
- UncleMeat 5y ago
  This is the current most promising solution, yes. But it remains a promising solution rather than a proven solution at this point. We've seen a variety of attempts to solve this problem over time and we've largely moved from one to the other without truly landing them. I'm not saying that there won't be a solution. Only that there needs to be a solution.