9 ms·
Personally feel that a better solution long term is Ethereum L2 payouts.
by zeroclip 4y ago
Personally feel that a better solution long term is Ethereum L2 payouts.
- Karrot_Kream 4y agoL2s are still pretty young. Polygon is popular but is ultimately a sidechain. Optimism hasn't rolled out fully. I understand why some would not want to trust L2s yet.
- zeroclip 4y agoAgreed, although short term solutions like having freelancers paid in ADA just to avoid ETH fees might end up hurting users and eroding their trust in blockchains in the long run.
- jsemrau 4y agoThe future of payouts will be omni-chain. Now that ETH is PoS, it is actually one of the benefits of the move, that there is a treasury with non-random payout cycles.
- imtringued 4y agoL2 is a fragmented mess. There can only be a handful of solutions that must all be interoperable with each other. Honestly the number of L2 for Ethereum convinced me it is a deadend. Even bridging is a deadend. I have seen 6 wrapped versions of DAI on Solana. The crypto skeptics were right, you can fork cryptocurrency endlessly until it becomes difficult to agree on which chain is the right one. Honestly either you do everything on L1 Ethereum or you do everything on L1 Solana(or Avalanche or any other scaling blockchain), there is no room for solutions in between.
- zeroclip 4y ago> I have seen 6 wrapped versions of DAI on Solana. What does Solana have to do with Ethereum L2s? And what are you referring to about bridges, forks, chains? L2 = rollup. Users have several to choose from, and are not forced into a single processor like Stripe or PayPal. Withdrawal back to L1 is trustless. Interop might eventually be solved with recursive zk proofs.[1] Trying to build a decentralized payment infrastructure on a centralized chain like Solana sounds like a disaster. [1] https://blog.succinct.xyz/post/2022/09/20/proof-of-consensus https://blog.succinct.xyz/post/2022/09/20/proof-of-consensus