Ethereum 2.0 Beacon Chain Decentralization & Transparency - Daniel Hwang

okay moving right on to daniel from steak fish we're uh talking about ethereum 2.0 beacon chain decentralization and transparency so do i got daniel here there he is hello good to see you are you are you ready to you ready to go yep okay floor is yours all right so hi i'm daniel i'm a protocol specialist at stake fish so i'll be talking about decentralization transparency on the eth2 beacon chain so yeah so i i guess like relative to other networks it would be nice to have some transparency and insight into the state of what our decentralization is on ethereum 2 beacon chain so stakefish is a proof-of-stake validator and we've been supporting ethereum-2 deposits so yeah i'll be talking about a few things so first talking about some how we can get from raw data to reveal data uh going over some terminology um talking a little bit about transparency and understanding what that actually means um some concepts of i guess security through obscurity and if that is actually something that's necessary uh for the robustness and security of our networks the tragedy of the open source commons and the bystander effect and what that has to do with our decentralization transparency and whose responsibility ultimately it comes down to and then i'll end with some sort of actionable insights so yeah i mean we're very familiar with like the moniker don't trust verify but i think what actually happens is we oftentimes have to trust uh in either infrastructure providers applications or even just popular figures that kind of say hey this is the way that things are this is the state and then we are no wiser if we don't have the technical acumen to verify first for ourselves so i think an important part is right right anyone can definitely run a node um but do many people actually run those nodes like i would say very few of the um of the people who are interacting with ethereum today do um and so i think like that type of transparency in like these public networks should be publicly available and accessible to to anyone um and and that means that right there should be ways of being able to verify uh easily right and and so like i say here like vanity metrics aside beyond that of like leaderboards for how much stake that you do control over the system um right objective transparency into that is important uh for for robust security and also to kind of maintain like the open source principles of our of our network so i'll go over some terminology because i think there definitely is a bit of ambiguity between other proof-of-stake networks and that of ethereum too uh specifically with regards to what a validator is and what a validator is not and then what service providers are and independent setups so um this comes down to like naming conventions i think like with other proof of stake networks if you're not familiar you can there are validators who are basically like the miners of proof of stake uh so to speak and they you know accept delegations on your behalf and then um you know you kind of get like a little reward from that from helping participate in the security of the system um but i think when and and those delegates delegation amounts um on those different networks are uncapped right you can basically provide send as many as many tokens on those networks to a validate or service provider but on ethereum it's you technically are able to send like an infinite amount of tokens but the way that uh validators are described is basically broken down into effectively 32 increments where a validator is determined by uh 32e right and so it's somewhat capped also there's kind of this concept of like what is what and who is who who are custodians not non-custodial providers custodial providers centralized exchanges independent setups and then also right like certain i guess like white label agreements that happen between service providers and other service providers and service providers and even non-custodial exchanges and this is just like an example from i think this guy's alex was from like the tezos ecosystem that does have like a right as i mentioned the uncapped delegation amounts uh for which is attributed to a particular validator in this case validator service providers um and he's basically saying here right you know ethereum is showing there's like at this point in time in december 2nd um like 21 000 independent validators but right it's not actually 21 000 independent validators it's 21 000 32 e denominated validators that may represent one 21 000 people or just a few of those and you kind of get into these umbrellas that you kind of portion them off into where how many of those 21 000 are actually independent setups how many of those are kind of denominated into validator service providers like like us stake fish how many does stake fish actually control right like the the validation keys the sanding keys obviously we we're in non-custodial and we don't hold your withdrawal keys and then how many of those are custodial exchanges um or just custodians like like bitcoin series for example so i think that like uh that that that terminology disambiguation is important to kind of understand right just like the pure raw numbers we see is is not necessarily uh what is actually um presented right and so for example i just broke it down here right this is one way to kind of look at it right now we have individuals at independent setups we have non-custodial service providers like stake fish we have decentralized pools like lido and rocket pool and we have custodial exchanges like centralized exchanges and then custodians like bitcoin suites and so within these buckets i think we can kind of kind of come to a a a start of where this uh what decentralization actually actually looks like sorry about that um and if we can bucket these instead of the 21 000 amount i think right now we actually have 70 000 but um right how much of those are kind of bracketed into existing uh service provider umbrellas how many of those are individuals and obviously from that we can kind of see kind of unique amounts of where the decentralized action decentralization actually is and what state we're at um and that kind of come leads into the concept of like trans transparency right obviously like the scalability trilemma is decentralization security insurance uh scalability right right um and so right these are all important concepts uh they're inherent to it and then i think perhaps with the way that some people are familiar in terms of exposing their validators let's say independent versus large service providers it's sometimes misunderstood as kind of anti-pri anti-security um and i'll talk about that concept in a bit but i i would say that basically to sum it up like transparency and favor over privacy is is basically paramount to identifying the state of decentralization the transparency that we can see at least gives us a a a clear picture of state uh of the state and then from an understanding of a clear picture of the state then we can kind of act on that and see what we can do from that right and this is basically an issue of right how many independents we have versus how many kind of pooled uh providers do we have so this kind of so this leads into right the the tragedy i say of like the open source commons right so um and and the bystander effect so um we can talk about and as i mentioned um right security through obscurity where uh there are these security engineering principles that kind of have have i've talked a little bit about this right like cloud shannon the information theorist kind of says like you should have a default state of like the enemy should know enemy knows the system i think another popular one is like kirchhoff's principle where it says like a crypto system should be secure even if everything about the system except for the key is public knowledge and then ross anderson says like the security of a system should depend on its key not on the dis its design remaining obscure and and i would say that these are are are pretty good arguments against having privacy at all costs to obscure your setups or even expose like how much you how much of the system that you own um just for that type of uh of privacy right because because i would say that maybe it makes sense for independents but for large service providers right that's that's something that you get that we all have we as infrastructure providers should have been doing from from the beginning as and as a default right and and right it's also having an understanding of like these open source systems that we should be under the assumption that the enemy al always will gain familia full familiarity with them i think like a good example um that we've seen even like in the weighty test net like one of the consensus features is johnny ria like basically was able to almost trivially be able to to dox certain people right um and and he was able to show that right and so i would say that right this this privacy that you obtain from security obscurity is at the expense of decentralization transparency and at the expense of decentralization transparency you get issues with actually understanding the complete state of the system and if you don't understand the complete state of the system then you can't understand like what you what you may be able to do maybe you may have like an abstract or vague concept of oh right a certain validator service provider has an enormous amount of of control or influence over the current beacon chain network um and then we should do something about it right but but we can't just operate on this under this kind of obs like fog of assuming things are are going as well as their where they're going right um and and so that's the kind of concept of like robustness um yeah sorry i must have redone that slide but yeah so the concept is right at the end of the day who is who is responsible for these things right we're in we're in like an open source environment we're in public networks and public systems but right uh like peter goodman uh had had given a talk on in like a kiwi con in like 2007 where he was talking about like this design mentality of like bugs in the brain right like uh and and how these bugs and maybe apply to the our decentralized uh networks is sort of like centralization is the bug right everyone also assumes that every someone else is looking at it right and and and in the end like the burden of making sure that these things are uh are operating as they should that decentralization is being achieved as as they should it falls on like no one and it also falls on on on everyone at this at the same time and so because everyone is making these assumptions we don't have clarity into the state then then no one's doing it right and and so i'm not trying to like harp on anyone here but like i think there are specific things that right we should be aware of um and then we should be kind of pushing to make sure we have transparency transparency into um right and so i think uh philip monk is a developer i think that with the urbit project has like a fantastic quote that i love it says where sovereignty necessitates understanding if you don't understand the system you're using you don't control it and if nobody understands a system the system is in control and in this case like what does that mean right if you don't understand the state of decentralization transparency or if you don't understand the state of decentralization and you have to don't have transparency into it you're making assumptions then the people who are actually benefiting from it or let's say at risk or um in power of being able to uh benefit from it are perhaps large service providers like like statefish itself right and and right the community and the ecosystem should be aware and maybe should may be more cognizant and demanding of their large service providers to to kind of identify themselves or at least have some pressure there positive i would say right and then again right you have like the don't trust verify moniker um but the thing is that i think we realize is wait what if we can't verify it right most people can't verify it um and this type of decentralization it must be easy to verify and understandable um i know i have like one more minute left but i think i will definitely say like there are initiatives that are in place that need to have these dashboards that are open source independently hosted especially not by validators or validator service providers like stake fish right so there's no conflict of interest they need to be credibly neutral and be treated as public goods and for example these are some some initiatives that are going on right beacon chain is developing some dashboards to support this nansen is as well there's community initiatives that elias from bison trails is doing and then superfiz from the eth2 due diligence committee is also spearheading a lot of this coordination as well so please i mean like go contact them and then that would be some great ways to kind of get started and contribute um but yeah i mean there's public and private interest in here i think we're all the same boat towards kind of developing ethereum two and getting getting to that point yeah if you have any questions please reach out thank you