3 ms·

Yeah, I think people want a better l1 from a variety of perspectives (some may be in conflict with other perspectives people think should exist), not necessaril

Yeah, I think people want a better l1 from a variety of perspectives (some may be in conflict with other perspectives people think should exist), not necessarily a l2, which really only even exist because of eth limitations, which has a whole bunch of other tradeoffs people hardly mention (like cross chain settlement risk within the same[bridged] asset being introduced with the roll ups). I know only one evm compatible l1 (avalanche c-chain) that's better from a consensus perspective (not nakamoto), decentralization (#of validator nodes that actually participate relative to stake) and time to finality. wrt decentralization, lots people ignore all the choke points in the eth/polygon model (infura, nakamoto consensus, local vs cloud run nodes [people like to handwave this in the avalanche discord as well and focus purely on stake distribution…]) which must not be ignored, esp when considering PoS.

sanderjd 5y ago
Could you say more about "local vs. cloud run nodes"? I get the sense you think local nodes are better, but I'm curious what you think the trade offs are.
- cinquemb 5y ago
  I don't necessarily think cloud run nodes are worse per se, I just think the crowding that happens when people run on a few big (aws, google, ovh, hostgator, etc) cloud providers is bad for decentralization and makes it easier for well resourced actors to easily compromise the network if they can compromise the big providers (legally à la 5eyes, and/or technically with server/node vulns) and modify the chain state of nodes with enough stake to make the modifications pass consensus (and potentially make non compromised nodes byzantine, unless they update to compromised nodes states). I think part of this tradeoff has been made because for a couple years with no real competitors to the evm (probably because from early on, most people seemed not to care for smart contracts, so the jury was very much out on it for some, compared to just forking btc) combined with peoples acceptance of pooling with PoW systems, most have become resigned and decided that the capabilities needed to run a node for a evm compatible chain will need to be delegated to the cloud providers (and even though newer consensus algos are out, people might still make this tradeoff to deal with the not quite answered state space issues being worked out/researched now).
  - sanderjd 5y ago
    Thanks for the insight! So I think the "local" here is more about spreading it out across more data center providers than doing it at home. I've been looking into running validators a bit and haven't quite wrapped my head around all the trade offs. The problem with ethereum is that it requires a huge amount of money to stake, which I can't really justify. It seems possible to run a to run a Cardano stake pool at home, but I'm not sure the incentives work, because stakers seem much more incentivized to choose professionally run pools. It seems somewhat reasonable to run a Polkadot validator, but only really in a data center, either in the cloud or with dedicated hardware. So in short I haven't really figured out a good way to get involved in validating as an amateur. And maybe that's ok, but the comment about "local" validating made me curious.
    - cinquemb 5y ago
      > So I think the "local" here is more about spreading it out across more data center providers than doing it at home. As well as geographically, more nodes spread throughout more existing national governmental distributions is better than less. > huge amount of money to stake > So in short I haven't really figured out a good way to get involved in validating as an amateur. And maybe that's ok, but the comment about "local" validating made me curious. Yeah, same with avalanche as it will require 2000 avax (maybe it will be lowered in the future when the dao goes online and existing stakers ok that). But at least at this point, I know people who are validating for x/p/c-chain on an RPI with 8GB of ram and external HDD or SSDs, which is way more accessible than pretty much everyone else (and has about ~973 validating nodes now).

sanderjd 5y ago

Could you say more about "local vs. cloud run nodes"? I get the sense you think local nodes are better, but I'm curious what you think the trade offs are.

cinquemb 5y ago
I don't necessarily think cloud run nodes are worse per se, I just think the crowding that happens when people run on a few big (aws, google, ovh, hostgator, etc) cloud providers is bad for decentralization and makes it easier for well resourced actors to easily compromise the network if they can compromise the big providers (legally à la 5eyes, and/or technically with server/node vulns) and modify the chain state of nodes with enough stake to make the modifications pass consensus (and potentially make non compromised nodes byzantine, unless they update to compromised nodes states). I think part of this tradeoff has been made because for a couple years with no real competitors to the evm (probably because from early on, most people seemed not to care for smart contracts, so the jury was very much out on it for some, compared to just forking btc) combined with peoples acceptance of pooling with PoW systems, most have become resigned and decided that the capabilities needed to run a node for a evm compatible chain will need to be delegated to the cloud providers (and even though newer consensus algos are out, people might still make this tradeoff to deal with the not quite answered state space issues being worked out/researched now).
- sanderjd 5y ago
  Thanks for the insight! So I think the "local" here is more about spreading it out across more data center providers than doing it at home. I've been looking into running validators a bit and haven't quite wrapped my head around all the trade offs. The problem with ethereum is that it requires a huge amount of money to stake, which I can't really justify. It seems possible to run a to run a Cardano stake pool at home, but I'm not sure the incentives work, because stakers seem much more incentivized to choose professionally run pools. It seems somewhat reasonable to run a Polkadot validator, but only really in a data center, either in the cloud or with dedicated hardware. So in short I haven't really figured out a good way to get involved in validating as an amateur. And maybe that's ok, but the comment about "local" validating made me curious.
  - cinquemb 5y ago
    > So I think the "local" here is more about spreading it out across more data center providers than doing it at home. As well as geographically, more nodes spread throughout more existing national governmental distributions is better than less. > huge amount of money to stake > So in short I haven't really figured out a good way to get involved in validating as an amateur. And maybe that's ok, but the comment about "local" validating made me curious. Yeah, same with avalanche as it will require 2000 avax (maybe it will be lowered in the future when the dao goes online and existing stakers ok that). But at least at this point, I know people who are validating for x/p/c-chain on an RPI with 8GB of ram and external HDD or SSDs, which is way more accessible than pretty much everyone else (and has about ~973 validating nodes now).