Community Call #16 w/ Team Teku

Transcript

[Music] [Music] [Applause] [Music] [Music] [Applause] [Music]

[Music] so [Music] hey everybody uh my name is superfiz i am a dude from the eatstaker community uh we are a kind of an open group who works to support anyone who is interested in uh

in staking on the eth2 network uh whether that's solo staking or pool staking um whatever your interest the east acre crew is here to help you out uh today this is community called 16 i believe and we are here with uh i say this frequently i say some of our favorite people but i think that's because i like so many people in this space but we're here with ben edgington and adrian sutton from tekku before i introduce them i would like to introduce our east staker group and then we will hop right in uh today let's see uh we have buddha buddha is

uh buddha works with bitfly so he is uh associated with um beacon c-h-a-i-n um you wanna say hi buddha yeah hi guys and yeah you probably forgot the terrible minus [Laughter] i'm encouraging him to be in recovery and ben is sipping his wine as buddha brings up uh we have nolan who you might know is lamboshi uh he's the brains of the crew while

some of the brains we have a lot of brains here yeah um michael or unvetica eth who does our devops stuff also i have regarded michael as my emotional support human which is a fun way to regard him uh and uh i i do want to send a shout out to colfax um colfax i i is not necessarily part of the um key staker core team but he does a lot with us uh so he's very much an eath staker and we're glad to have you with us and finally patricio worth halter is in charge of the poet program and i saved him for last because he will talk just a moment about

um how you can get a poep for this call hey everyone yes i'm patricio and because we have big plans for these pow-ups um like giving you access to great things because you are a loyal each staker communicatingly um instead of giving the po-ops to everyone in the call like you used to use like we used to do i'm going to deliver them to people who has video on or to people i know so if i don't know you or you're not active show yourself on discord prove that you're a human we want you to be a happy member we want you to collect powers we just want to discourage people farming them and

getting multiples multiple tokens if they are only one person with that enjoy the great uh thanks and we we will talk a couple more times throughout the call about claiming poeps um some calls that's the only reason we show up but i know this one will be well they're all pretty interesting uh some other people i want to send a quick shout out to uh terence from prismatic is here um he uh and it's okay you don't have to unmic i just want to say hello that we're glad to have you um and someone who's been really helpful to eat staker steaker with distributing our genesis co-apps is ktlxv and we are glad to have you and we

appreciate your support of e-sticker um so let's rock on uh ben and adrian uh develop the teku uh beacon chain validator client for eth2 um would you all like to begin by giving us a brief introduction uh ben will you go first for us yeah sure so um i am product owner for teku for what it's worth that's my my uh day job my title as it were um you might also know me from what's new in eth2 which uh i write every couple of weeks you can find it at eth2.news

and i've been hanging around ethereum for about five years now so come across many of you at one time or another awesome thank you uh and yeah i the what's new in e2 might be it's it's one of those things like you know you get a lot of newsletters and a lot of them you just kind of like ah you know kind of wave off or mark as read but yours is one of those that i drop everything every time i see it and uh look through it so it is i think i've probably called it the most valuable piece of of regular content in um ethereum 2 news so hopefully we'll get to talking more about that today

adrian can you tell us a little bit about uh yourself yeah i'm adrian i'm one of the developers on teku i've been working on techyou for i don't know must be coming up towards you a year now eight months or so um before that i was a developer on baysu still had a little bit of attention there but don't have a lot of time with each two uh rocking along um yeah pretty much uh in ethereum because i like this low-level technology more than anything so the higher up the levels you go the less i know but i like it down here in the in the dungeons making the protocol work cool you know adrian's having fun because it's been a year and a half since he

started work on tekku wow i had not realized you know time dilation in the ethereum space is like one of the most real concepts ever like it's some some things seem like they've been weeks that have been years and other things seem like they've been years that have been weeks it is uh the most interesting thing uh before i shut up for a minute and let the team get into some of these questions been something i i didn't know that i didn't understand until just now you described yourself as as product owner but i know that there's a deep affiliation with consensus what is the uh what is the the link there

yeah so uh i joined consensus like three and a half years ago originally to work on enterprise stuff i very quickly deemed that quite boring and my passion was for public chain things and at that time consensus wasn't really doing any protocol level work in ethereum for various historical reasons some bad blood with the ethereum foundation and things like that but things were warming up and we were getting into doing protocol level stuff and uh i um wanted to make that my mission so i set out kind of on a solo journey to work on whatever was next the next generation of ethereum it wasn't eth2 at that point it was like staking um

you know running contracts on eth1 to manage validators and stuff like that but um it soon developed into what we know as eth2 so i think i first met terence uh back in taipei i think in early 2018 uh and it kind of uh grew from there i'm gonna have to close the chat box i'll concentrate on anything yeah that's like speaker of you only like i don't stop letting people in and um and so over time that was uh we built an r d function around ethereum two working on all sorts of interesting topics and then we decided we wanted to build a client so

uh i moved over into the product team well we built that in r d so we had uh people like johnny ray and joe delong and stephen schroeder were working on that uh and aquila raju as well um and then about a year and a half ago we transitioned it over to the product team adrian came on board and our sort of you know um it's a kind of 80 20 rule right it's easy to make something which is functional and a prototype but make it product quality is actually 80 of the work and that's what we've been doing for the for the last little while and um and that's it so i don't write a lot of code they won't let me write too

much code but i um i try and set the direction and oversee things and gather requirements from customers and you know just keep in touch with the wider uh business and and community okay so i am i it's funny well how little little of this like in my mind i think i know a lot about you and then we'd start talking and i'm like i don't know any of this um are would you consider yourself the bridge between uh research and implementation is that what role you serve yeah so product owner product manager um whatever whatever you call it you're kind of at the intersection of uh three three things you got the the people doing the um development work so the techies and you need to speak their language and kind of understand what

they do um you have the business um which is represented in the kind of companies priorities and what what they want to do and what's the kind of commercial model if any and so on so on and you've got the customers of course the users and sometimes you get to interact with them directly which is great i can do that now in my old job i was always interacting with customers via sales people which was just horrible so this is definitely an improvement but yeah so product manager sits at the center of a web uh in in a way and just uh tries to keep things aligned on on several axes let's i think you're perfect for the role because i love to see you uh

popping in the e-sticker reddit and helping solve things that um you just add that that unique perspective that you you don't come across as as a product you come across as a very engaged and willing helper and i i love that attitude so uh thanks for that i am going to shut my mouth for a minute and let my friends um kind of work down these questions do you want to go buddha i'll go um can you guys tell us about um you know earlier on in your careers

what's your background maybe some education if you have it what you used to do for work i'm sure they have some education too um yeah i'm in education i did a bachelor of information technology with griffith uni here in brisbane uh majored in computer science and artificial intelligence um but with quite a lot of software engineering in there um i've actually had if you ignore you know a few little jobs straight after uni i've actually had three jobs in the last 20 years including this one so i started out actually as a visual basic developer building an activex control

for in-browser html editing before there was dhtml so back in the netscape 4 kind of days they very quickly discovered my visual basic was awful and i started working on the java applet i spent nearly 10 years doing that worked my way up to cto there and they've then moved into javascript so if you use wordpress that that editor in there is now the same company i used to work for it started in a different place um so a lot of web content management that kind of thing working in the enterprise space we sold a lot through ibm and places like that which is really quite interesting to see big uh big companies trying to manage creating that much content and websites for thousands of pages and so on and

getting the average user using it um the whole point was to kind of decentralize content creation i guess i've never put it that way before but it's it's it's been interesting to engage with lots of users there and then i spent about eight years uh with a company called lmax doing finite uh foreign currency exchange so they built and ran a low level a low latency exchange in java they basically realized that they spun out from betfair actually and realized that during something like the world cup betfair was doing more trades in the london stock exchange by an order of magnitude and maybe foreign currency wasn't that difficult to get into maybe maybe instead of just

betting they could have this spin-off that went into finance uh so i've seen a lot of fun stuff there and uh that got me into kind of this world so when we start start doing cryptocurrency they've started moving into there and some trading as well um and just really interesting the low-level stuff lots of java the amount of stuff you can do with with java and the speed at which you can do it really surprises a lot of people um i just don't want to get that in a few times this call um java's not slow uh yeah and then um consensus actually hired a couple of people that i worked with previously um back in that first job and say what you know if they're going to hire good quality people like that it

seemed like a company i should check out um and it's kind of been a wild ride ever since uh bouncing around a few teams here at consensus and then getting into the e2 stuff i just love the technology and that's what keeps me coming back every day you're in australia right yeah i'm curious because i australia seems to me to be one of the on one side like one of the hotbeds of each development um but on the other side you know people like when i hear about stakers from australia they're like oh well our connections are terrible you know we have latency so how would you characterize the australian east community

this is such an american question what do you want from me patricia i can't be my flag is right there i think it varies a lot if you come visit a single country in the world let it be australia well i mean i don't think it's too american i i like i can characterize the american ethereum scene and you could probably characterize the argentinian ethereum scene why i mean i'm just curious to know how he sees it there it has sounded us as you do guys have blockchain developers how is that but yeah

that's not how i intended it yeah yeah i mean we ship packets in the pocket of a kangaroo but um [Laughter] i think it varies a lot you know australia's a big country and so there's kind of different worlds in between sydney and melbourne and brisbane um sydney is a lot bigger in technology to brisbane but funnily enough consensus has more people here in brisbane than anywhere else i think maybe sydney but certainly in the protocol level we've got quite a big team here of researchers and developers and you get those little hubs to kind of build up it's it's pretty exciting in terms of the stuff that's going on um there's been a few projects uh i think the australian stock exchange and and some places like that where it is

starting to attraction with government and with with some of the bigger companies so it's it's not a complete backwater whether your internet is good enough for staking is very hit and mess here um there's lots of people with really good internet here i'm very fortunate that that i got a bit of the nbn that works well um and there's lots of people that just complain a lot and do have a lot of dropouts and a lot of problems and the further you get out from the cities the worse it gets obviously in my defense patricio i i did not imagine i didn't imagine the the backwoods aspect i imagined it as a thriving community and i just i wonder how how it looks from there like yeah i mean i think the the big thing

that you see and i think this probably applies right across kind of the apac time zones is that everything else happens overnight um so if you ever watch the these two call calls or develop calls i'm never there because it's midnight um and you know this is 6am this is really quite civilized for me it's one of one of the better meeting times i get in my week the cigna crime team always makes it [Laughter] yeah yeah i mean that's a big effort i think they're also more more night people than morning people than for me but you do see that there is there are long distances it's hard to

get to a lot of the conferences and things like that but there's still a lot happening here and particularly through asia um in apac there's there's a lot of blockchain going on there and it's really is picking up in australia as well how do you guys manage the communication so if you are awake ben goes to bed so how do you do internal communication yeah lots of async stuff between ben and i a lot of the time but we get little bits of overlap so it's one of those things in the modern world you're very flexible about the hours you work so even if i knock off relatively early in the afternoon because i've been up early i'll still be on you know on messages

late at night you know so ben can i mean ben's ben's waking up something like 6 p.m my time it's not too bad so after dinner you know once a week we have a call in the evening it's about 8 p.m my time um and those kind of things mostly though it's just yeah a bit of back and forth on on slack and it kind of all works out you can you can do a lot of async stuff um fairly well it's basically mandatory to be async i mean the full team currently active on tech you uh adrian and paul in in brisbane we've got meredith and gem um in east coast time us and we've got anton in uh some petersburg russia and myself and you know there is no time which is good for

everybody so yeah we we use slack a lot are you happy with the size of your team or would you hire someone else if you found the right person well funny you should ask i um so uh we wouldn't we don't tend to hire directly into the team so pegs we're part of a bigger organization which is pegasus pegasus is i don't know 60 people or so protocol engineering group in consensus so we tend to hire protocol engineers uh into pegasus and then people find roles within within the teams we may be looking to add one to the the tecu team so if anyone has uh strong

java chops and is interested in working on it then let us know and we will um see how things go speaking of java that is probably the perfect time to ask or to have you all tell us about the teku client uh what sets it apart what is unique about it um and i that's not my question let me start then adrian can put me right um so you know the java thing is comes out of the the pegasus and consensus the background so the the initial ethos was around enterprise um ethereum and uh

felt you know we didn't want to scare customers so we didn't want to program in some obscure language they'd never heard of there are more java devs in the world than any other um language and so it is just easy conversation licensing as well was was an issue so we chose a very liberal license apache2 and so on so you know those are the people we had available when we started writing teku and it was just a natural choice for us and as adrian says it doesn't hamper us in any way i mean java is quick uh it's efficient there are vast libraries available for everything under the sun um so you know i i kind of feel for the status team that um the nimbus team uh because they've basically had to write

everything from scratch i mean all the core libraries and everything you know it's a lot of work um whereas you know our biggest problem is you know which one do we choose so that is um as for language choice we we took a decision early on i i felt it was strategic for the clients to sort of differentiate themselves in terms of target market and you know one level all the clients are the same right they all do the same job they run on similar hardware they have similar kind of spec requirements um and they all have to talk to each other and they're basically by definition interchangeable but at another level i felt it was good to try to differentiate our target market

um to give us a sort of a focus and you know we're not all competing for the same users all the time which is you know just a bun fight so given consensus and pegasus and where we were and the kind of background of our devs and our sort of dna uh we decided to go after the institutional staking market um and you know we can come to my kind of personal preferences later that which was slightly at odds with that direction but um you know that made sense as a product emphasis for for teku so that is when we imagine our target clients when we think of features when we're talking to users and prioritizing user stories it is the institutional stakeholders in mind so

we're developing features which make their lives easier we're not so much focusing on people running one or two validators at home though i am quite happily i'll take you and it's terrific um but you know we're looking at people running thousands of validators and and trying to do things which make their operations better so that's that's the kind of um high level description of where we're going can i strike while the iron is hot real quick you said that you take odds with um being identified as an institutional client just because i will forget to follow up on that i'm curious about it yeah all right i mean i i have a personal a passion and i have a kind of

professional kind of role as well my personal passion is you know why i've kind of thrown in my lot with this community i won't be able to be staking at home i love that decentralization mantra i'm fully bought into it and you know in a sense having my client be the client of choice for institutional stakers kind of goes against that a little bit and i always have to think twice when i say it but you know the these people will be involved they need to run something i'd rather it or our client than something else so um you know we'll do our level best to make the the best client for institutional stake as we can although you know outside office hours

my passion is for for you know doing what you guys are doing that is a uh that's a really great uh line and i i hope that maybe later in the call if we have time left um or maybe it's one of the questions i would love to talk more about um how we see decentralization of the network and how it's how it's evolving in the first few months um but i don't want to hijack so let's continue to move on and i would love to hopefully get back to i'm going to add it to the notes to get back to decentralization later on no wonder michael do you want to uh yeah actually uh real quick so uh ben in your editorial

and your latest blog post you mentioned many changes to the ethereum roadmap so can you kind of explain and give a brief overview of the original roadmap and how that contrasts with today's roadmap yeah yeah this kind of favorite topic of mine uh i just love the way that ethereum is adaptable that the things evolve we're not you know i used to work on like five-year gantt charts and everything was nailed down from the beginning and we'd have this massive project plan and now we we do a lot more exploration and um i i love that um so we a couple of years ago we laid out a very neat uh phase zero phase one phase two linear road map where we deliver the beacon

chain we deliver data charts we deliver execution environments on top of that and then we have this kind of merge of eth1 into eth2 as an execution environment but as we've gone along new things have come up and new priorities have emerged roll-ups have come on the scene since then and look like they can solve some of our problems uh without necessarily having to do kind of hard stuff in protocol we can sort of move that into layer two so things have evolved and the current roadmap is roll-up centric so there are three things we need to parallelize to scale there's uh data input which is

blocks uh there is state which we need to hold and there is processing of the data so if we if we want to scale we need to parallelize all three things the original plan was was to do that um you know phase one was to parallelize the data input the blocks and phase two was to paralyze the processing and and the state we've moved away from that we've said we're only going to parallelize in protocol the data processing the shards and we're going to make the handling the state and the execution uh the problem of rollups so they're going to handle the uh parallelization of uh of estate and execution so that

makes our job a lot simpler as as protocol developers and we've taken advantage of that to bring forward the merge of eth1 and eth2 so we can do that very quickly if we need to um and so we've got this executable beacon chain concept that was first published by our colleague mikhail in november last year so quite recent and that's independent now sharding we don't have to do sharding first but we don't have to do the merge first we can we can pick and choose and uh yeah i mean i i was seeing a lot of comments on uh twitter and elsewhere just people saying you know these gas fees are killing us uh when heath two uh hurry up guys get a move on and actually eth2 is currently envisaged does not

solve the gas problem roll up solve the gas problem and that's not what we're working on so if we aren't scaling the layer one is there any concerns that entering and exiting roll-ups will still be massively expensive for people uh that's outside my area of expertise to be honest i i think mostly people won't be entering and execute exiting very often and there will be protocols to do cross roll-up transfers without hitting layer one

so it's only when you need to have a dispute or sort out an issue that you have to appeal to uh layer one and hopefully that will be rare but that's my current take on things so i wanna unpack that a little i just want to unpack that just a little bit in the sense that so the takeaway here i'm getting is that we're moving from more of a roll-up centric architecture over sort of prioritizing sharding as a scaling solution to ethereum two can we really kind of like eli five this and uh for the newcomers to the scene and why that is different and why sort of the

pivot to roll up centric versus the sharding priority let me try out an analogy that i've been working on it needs a bit of polishing but so evm is an engine uh and currently is underpowered um so our initial solution to that was to have um many many evms so we originally had in mind 1000 1024 engines and that was how we were going to get power um but that came with all sorts of problems about how these engines work together so what we're doing now with roll-ups is roll-ups are a way to turbo charge your engine so they take the input data they compress it so like the fuel air mix and they ram it into the engine and give it

more power and they also kind of deal with the state problem but that doesn't really fit in the analogy so we'll ignore that um and the great thing about so you can do that on ethereum one and you can get some you know 100x or whatever depending on your application and your roll-up type speed up more power out of your engine your single engine by sticking a roll up in front of it ramming the data into it the fuel air mix and you get more power out then with eth2 with the sharded data you compound that gain because you've now got 64 fuel lines now so the shards are like extra fuel lines coming in and so your your roll-up turbo charger has you know 64 times as much or actually

quite a lot more than 64 times as much um fuel that it can now fall into the engine and then you get this enormously powerful uh engine but there's only one of them we're not like doing a thousand engines anymore this is helpful for you as an analogy on the early days of the internet um countries connected between themselves with uh dedicated uh optic fiber lines and and when they needed more capacity they they dug more trenches and they put more pipes and optic fibers and and during the dot-com bubble they were laying pipes with hundreds of different fibers spending huge amount of capital and and a couple years later and and it was

quite groundbreaking pretty much like roll ups they discovered that they could do wavelength division multiplexing so so they kept using the same pipes and and having uh hundreds of magnitudes so i mean many others have funded more capacity and it's up to these days that we are barely using the the lines that there were dug 20 years ago so so maybe that's an easier analogy to put together yeah yeah i like it i'll think about that so i i'm still trying to wrap my head around the current purpose of the shards does it mean something like matic which has become polygon or optimism that

each of those might have its own shard that it checks data into or does it matter does it work that way i think they can watch multiple charts and take data input from multiple shards and it will depend on the application and honestly speaking these are unsolved questions so as i noted in my editorial um this is a rapidly evolving space it's going to be chaotic for a while uh as these layer two solutions are built out and it's kind of one of the disadvantages of the new direction is it's no longer neat and ordered and phase zero phase one phase two is now this kind of big mess of layer two stuff that kind of needs to shake out but it but it will it

will quickly um and um and then we'll know a lot more about what the future looks like it's funny because i i tend to regard myself as the leader of the east staker community and i feel like i should understand a lot of these things and then they're like these really basic questions and i'm like well i've really fizz i got to know that and then i you know i i build up the courage to ask and you know you're like well that really hasn't been solved yet my god well maybe that's why i don't know so it is it's good to know that uh to recognize that it's so complex that um while it's still being worked out there aren't a lot of answers but it's also the kind of thing in two years when it's done it will be

completely obvious like aha you know it was right there all the time we just didn't see it then so thanks for that perspective i think the other thing worth noting in in that is that while roll-ups are kind of undefined they're actually working at least in part the the phase two that we originally envisioned for ethereum um there was some really great research into it and it's not completely undefined but there's a lot of problems still to be solved in it um and a lot of work to be done i think it's a lot less defined and a lot less known than where rollups are which is what makes them a a good choice to go there now and if we needed to then we can still scale layer one with with that phase two plan in the future potentially once we work out how

is it wise for us to repeat the line that uh eth2 isn't the scaling solution that it's roll-ups is that something because i it is a understanding i'm beginning to solidify is it the right message to to push out to everyone yeah i so it's it's not quite that simple so eth2 does um scale roll up so roll ups give n scalability and and e2 gives you know end time 64 or whatever so they do compound um but in the short term uh then uh roll ups are the way forward for scalability i think i think of eth2 a lot like the e1 point x moniker um which started i

think it was prague um but there were some talk of where we're not getting to eat too fast enough the iphone chain's not going to survive and they came up with three or four task force to look at this technology and i can't remember what all they are what all of them were but i don't think any of them shipped um some might have gotten through but eth12x is just kind of rolled on anyway and every time we change anything on what's currently called the e1 chain we call it d1x it's just improving ethereum and i think e2 is like that um people keep asking why we aren't rolling this out as part of e2 and why are we doing it now or why are we you know e2 is just the future of ethereum at this point um the road map has changed so many times that you can't point to it

and be one specific thing and it will keep changing and that's that's really cool so i basically say yeah roll ups are a part of each two now i think the one thing we can probably agree on is that we we will transition from each one to each to at the point when the merge like the block when the merge activates would be the moment that we say we are now in eth2 is that is that a fair assumption we kind of keep talking about these two rollout after that anyway this will stop talking about equine yeah um the people are quite down on it

vitalik says he tries never to use um the uh expression dany is moving away from it um as the purveyor of what's unique too i'm i'm a huge advocate of the term but yeah i mean yeah that's the point where we turn off proof of stake which is a proof of work rather which is a huge milestone it's been part of the ethereum story since almost day one um and i think is worthy to say that point at which we do the merge we're now running on proof of stake not proof of work you know that's a massive massive milestone it is worth a name and serenity is the old name um but uh e3 will do i'm still comfortable using eth2 in

regard to the chain i think that makes perfect sense in regard to the the coin layer as in the value layer it is all ethereum um but when you're discussing chains you know what what chain are you on i feel like it does make some sense then to describe e1 or e2 because they are such different entities yeah yeah true um yeah beacon chain we could call it whatever but i'm i'm content with e2 i don't feel like we need to be revisionist about this it's fine if you guys were starting teku from

scratch is there anything you'd change [Music] yes i think any time you you working in a code base there are things that you don't like um and you've got to do that again um but ultimately in terms of what a an e3 client needs to look like i think we need to go through three or four more hard forks um and really get an idea and probably the merge and you know it's in two years time that you can start to look back and go oh that's that's what a client needs to look like and in fact these one clients are just doing that now they're all refactoring madly the

low level data storage because the way they've stored it up till now isn't efficient enough and there's actually this this whole new way that took alexis three or four years to try and prove out and convince people and i think that kind of approach is going to apply for these two clients as well so at the moment we're all talking you know attestation inclusion rates uh by the time you merge in each one are they going to matter as much probably because you do get paid for them that probably still be a fair bit of money tied to that but your your block import rate and the way you work with that e1 client how you handle latency of having them on different machines because the scale is going to be quite different in terms of the resource load can you outsource that to someone

like infuria and and have a lot of latency between you want to know the ethon node all of those kind of considerations are going to be so new and so much more of a challenge that we just don't know what it's going to look like in terms of how that client is structured so you know then it kind of comes back to the relatively boring answer of well i'd like a client that does what it needs to do now that's well tested and is easy to change because it's going to have to change and i think that's pretty much what what every client team is working towards now is just making it easy to change because i know the tsunami of changes is coming okay i think i can continue now i actually wanted to say like we were

talking about proof of work so i i thought i'd jump in um adrian was writing a post about the half truths of eavesdaker about stakers and that with proof of stake that you pay the minimum to secure the chain whereas with proof of work you pay slightly more and i wanted to know what your thoughts about proof of work are i think the key thing is that the policy ethereum has had since day one is to pay the minimum possible for the required security uh so i think that's true both of proof of work and proof of stake and that's why proof of safe has this built-in reduction in rewards as the number of

elevators go up because the more validators we have the more secure our chain is the less we need you so if you drop off it's not as big a deal whereas if we have few validators we want to pay more to try and accentuate and that's exactly the same mechanism you see with the hash rate and difficulty kind of balance the miners unless there's more motors come on to me what do i think of proof of work uh it's a pretty expensive kind of kind of endeavor to burn a lot of processing time for not a lot of good really when we can do better without it i think i think people a lot of people actually get caught up in in hating on minors in

that and particularly as stakers i think we've got to really watch that because and this is why i kind of write that wrote that post was stakers are the new miners everything that applies to miners today in terms of you're a service provider and you're paid for your job and the rules might change at any time that all applies to stakers as well um and that doesn't mean that the world is out to get you or that you know people often phrase it as miners aren't part of the community and i hate that phrase i think that's so wrong i think mine is a part of the community as our takers and i i think one of the big advantages of proof of stake versus proof of work is that when you exit proof of stake you get your money back

so you make an investment and it sits there and as long as you don't get yourself slashed or perform really badly your money is still there at the end so if you find that the rules changed or that the rate of return is dropped because there's enough elevators it's really easy to exit and it's not it's not a personal flight it's not a huge loss on your investment you're not kind of you know i spent a million bucks on mining rigs and now i just burnt them you know they're worthless it's it's a really much better alignment of incentives of i'm happy to take that rate of pay no i'm not tomorrow i am again you can kind of jump in and out so much more easily and without the fast

you know a lot of this started because i've been paying attention to 1559 mining channel which is uh yeah i think um until we like until eve 2 doesn't prove itself proof work will be the choice but um until like once we move over and if two proves itself like i believe that other chains will follow to proof of proof of stake but until then we have to wait i guess yeah seeing a lot of change starting with proof of state um for a variety of reasons uh doing it at scale of ethereum is really really hard that's the unique thing about these two is the scale of decentralization that you can do with

proof of stake um bitcoin will never change from proof of work uh it just is um what other new chains do i think will be a variety of things they're like kind of trading off the decentralization with you know throughput on different things like that and and it's good to have those choices yeah uh just for my hey pennath um there are two things i like about proofwork it's super simple uh you know it's got properties that are easy to analyze and you know having been involved in implementing proof of stake over the last couple of years i think i'd tell you it's not simple um and it's much easier to bootstrap uh which is quite a subtle point but you can

participate in the protocol without having any of the coin of the protocol you can make coins without having coins uh originally and so for the original blockchains that made a lot of sense right bootstrapping it prove a stake not quite so easy to bootstrap but uh we're not there now we're in a different place now and we're you know we've all got magic internet money so it's uh we can um don't have to worry about that aspect of it um yeah i mean one of the very early things i read so i said i got into kind of ethereum world about five years ago uh i read very early on a draft of a book i can't remember the name of it now but it's written by some consultants and it it was about

uh blockchain technology and one of the very late chapters in the book just mentioned in passing this idea of proof of stake where instead of spending your money on a bunch of graphics cards you you put it into the protocol and i just thought this was brilliant and it just grabbed me it's one of those oh wow this is incredible moments a flash of insight and i thought yeah um we should do that so when i learned later that ethereum was planning to do that it it really drove my my interest in ethereum and and has kind of like been a uh a goal ever since probably was purecoin the first something like purecoin was one of the

first uh proof of stake systems that just seemed really at the time it seemed really hokey um yeah because it was the first but i'm sure you know we look at the transition now it made so much sense to begin with proof of work where like you said anyone can can run hardware to get into the ecosystem and it made sense where you know four or five years after that when the people who were invested in the technology uh that's kind of like a uh like an equal opportunity airdrop uh in a way in a way of a manner of speaking and that those people could then progress to the next level um so i i think that that uh progressive

model is really a cool way to look at the growth of ethereum over time which is my two cents this is probably a good segue since we're sort of dancing around this topic anyway um but one of the platforms that i manage is the ethereum study master site and the focus of that platform is to help educate newcomers to the ethereum ecosystem so i'm wondering if we can in the spirit of that educate newcomers to the ethereum scene and sort of dissect and break down one of the hot topics within the industry and that at least in the moment one of them is eip 1559 so i'm wondering if either ben or adrian want to sort of break down what that is in a very

general sense and why that's such a polarizing topic within the ecosystem i need more wine before i'm going to tackle that i hope you may 59 is a change to the way that uh gas fees work essentially uh it's a change to the auction mechanism uh the way we currently price gas fees it's actually incredibly hard to know how much you should pay it's kind of like bidding on a house it's the analogy that i think was from the tim rough garden pipe paper um and you kind of put in your bid and you want to make sure it's high enough that you actually get in because if not you miss out and it's a real problem

so you and if you get it that's the price you pay whatever you bid is what you pay so you really incentivize to put it down low but you kind of can't because you've got to make sure it's high enough all the time that doesn't work right so 1559 changes that so that the protocol sets a base fee and that's the amount that it will cost you to get in it changes slower it's about 12 per block at most so you can kind of predict it but more importantly you can set a cap on the amount you want to pay so you can actually say this transaction is worth it to me even if i have to pay 100 to getting but you don't have to pay 100 unless

that's where the price actually is up to so if the real price is one nano week you pay one nano and the rest of money you get much easier pricing much easier for users to actually understand and it's been hard for people to wrestle with this because there's a lot of economic analysis in it but that's what it boils down to is you can say maximum with confidence that you'll only pay what's actually required why is it controversial well it turns out that the way to make things work in the incentives line up the base fee that you pay gets burnt instead of being paid to miners now when 1559 started work transaction fees were a very very small component of minor income they're really

dominated by the block rule which is the security reasons what you want from a blockchain you don't want your transaction to be dominated in the meantime over the last couple of years transaction fees have blown up and it's now a huge part of miners income and you know we're giving these people a pay cut basically um people don't like taking a paper you know it's a hard time to sound the outrage um at the same time it's there's a lot of confusion in that because of the difference between what we pay miners in total and what any individual miner will wind up making so at the moment it's good because hashtag is trying to catch up with the price of ethan with the price of transactions so individual modes are actually making a

decent amount at the moment i think um but if you take a sudden drop in that you get the effect a sudden drop in the total amount we pay to minus which 1559 you get the opposite effect um in the cash rate will be really high so it's hard to find a block and you're earning a small percent of the total rewards and it'll take time to come back down so individual miners will actually wind up a lot worse until that hash rate comes back down um and then then you just get all the politics and drama of we hate miners we love miners all that stuff mixed in and you get this huge blow up ultimately though i think i think it's a lot less controversial

than it seems like everyone seems on board with the idea of it it is just a matter of um you know working through with miners of how to how to adjust that that hash rate so that we still wind up with a good distribution of miners and that they're not too badly called out some of which will just be the fact that it's still at least six months away from shipping so well five months now probably but yeah the the notice of when people notice things when it actually ships is always a lot longer than anything and i know your uh wife is staking can you do a psa about what redundancies and failovers

she has set up yeah yeah she's more into the kind of um uh you know uh d5 dj stuff herself so she delegates a staking uh to me um so i i i manage her stay quiet she's using [Laughter] i manage her stake on her behalf and as um yes no i do recommend teku so we're running teku here uh i've got a little dedicated box um it's just an i5 6 core processor 16 gig um ram and it's running super smoothly running teku i just on a home adsl connection my bandwidth isn't super but yeah it's it's terrifically reliable um i checked today it's been up a month i

haven't upgraded recently and it's just been running a month no issues um i've got a ups which helps occasional yeah we get five or ten minute power outages every now and again i had a couple of internet outages which yeah you you lose some reward but you quickly make it up again it's been great yeah really good really good experience one of the big uh concerns that we have from people who are interested in um mining and this is one of the things i see every day on eatstaker on reddit eaststaker is you know i have 4000 ether and i want to stake but i can't or maybe it's 32. i have 32 ether and i want to stake but i can't or i won't because i

don't want to get slashed because it seems like everybody gets slashed um can you can you help put that in perspective for people who may not really get the context uh yeah i can say a word adrian i love you as well um and the number of individuals who've had their state slash is tiny um i i forget how many slashings we've had altogether but the the vast majority have been a couple of staking services um who screwed up and there have been a handful maybe six individuals who who've had their stake slashed um and slashing isn't as bad as it sounds i mean they've lost about 0.6 of an eth um their stake is locked up until we do the merge which is unfortunate at this

current time but you know they haven't lost their whole stake or anywhere near it just as a small fraction so yeah it's not a disaster um and you had just have to take some very basic precautions and the clients have all got built-in slashing protection now um and it's a there's an interchange format which is an eip and i think we're all supporting now i think prism's just put that in uh so if you change client as long as you port that data over you can um be sure that your your the client you're bringing up is um not going to do anything slashable uh so just taking some fairly basic precautions about making sure you never run

two validators managing the same key at the same time and it ain't rocket science so if you are if you follow that basic maxim you will not get slashed uh it's it's it's just not going to happen um while we're talking about slashing uh what happens if people don't or forget to update their if onenotes at the berlin heart fork can they get slashed for not updating no no and actually there's an analysis i i did on um you can read on the e2 specs repo that currently only three percent of rewards come from proposing blocks so the only if you've not connected to an eth1 node the only thing you can't do is propose

blocks uh and only three percent of the rewards come from proposing blocks and it turns out that actually most of the time even if you you're not connected to an e1 node you can actually propose blocks so i forget the exact number but basically if you never have an eth one node connected you might expect to lose about point four percent of your total rewards maybe ten dollars um a year uh which makes it kind of questionable about why anyone would bother running an eat one note and i don't i i connect to infuria um so uh which is bad you know i'm i'm a bad person but you know it's just not worth it you're a home staker that makes you a contributor to the network that's you

know one of the things i'm i spend all my time like you don't have to be perfect you don't have to be ideal but be the best you can and can you help so i'm glad that we talked about what it takes to to get slashed can you from your perspective talk about and this is great because you run or you develop the client that is used by industry for staking most commonly can you talk about the difference in slashing um for someone who might be staking with an exchange or a pool versus the slashing that happens if i'm a home staker and i have you know one node running why where's the counter intuition in there and why

why would you suggest that it's better to stake with a large pool or with a single validator at home depends in conflict with yourself so i'll take that so the the advantage of staking from home is that slashing penalties scale up as more people get slashed at the same time so if you have just you get slashed you're one validator the penalty is really small if you have a thousand validators all being slashed at the same time maybe more than that it starts to really ramp up because that the network is basically trying to

determine whether you've accidentally done something stupid whether you're actually attacking the network and it's trying to penalize you accordingly um they slash the bigger the penalty and even with a staking setup and they stuff up there's a good chance as we saw with with one that you get a whole bunch of slashings at once and then the penalty is increased so actually staking from home um much as you know we love tecu and staking providers are awesome and yeah support them and thank you uh you know second time is actually a really really good option if you keep your setup simple and don't worry about the downtime right so as ben said if you have some downtime it's smaller you can

make it up um a really simple setup that just sits there and runs it's not going to correlate with other values all the time your chances or your risk even if you do get smashed is quite low um and the same thing for if finalization stops you're not going to be contributing to that you'll be one of the good people helping you just keep going and things like that plus the the general thing of of aiding decentralization where i think staking pools are really useful uh is especially for people who don't have 32e it's a great way to onboard and for people who running even a simple node is too much for them and i think that's they're both good markets to be chasing

um running a note is not as hard as you think so it is worth you know if you've managed to find yourself into this community there's a very good chance you're technical enough to run a node um so it is worth worth a look but i don't think there's any shame in using a staking service it's like you said you'd be as good as you can be um and you're still contributing to the network because you can always take your stake and move it particularly once withdrawals are there it's a bit of a funny time at the moment but but longer term it's like mining calls you pick a staking pool that is operating in a way that you you are happy with and that you believe is contributing to the network so i keep hearing developers say

if we need to do the merge we can do the merge which is sounds exciting to me do you do either of you want to uh justin drake up a prediction for me a prediction on when the merger i don't make predictions about the future anymore yeah i'd love to talk about the merge though because um teku is is very much involved in this so um part of pegasus is our r d uh division which is uh thriving and mikhail formerly of the the harmony ether2 team is leading a leading r d team

within within pegasus and he's used teku and he's working with the ethereum foundation with guillaume over there who's got a fork of geth and they have made a prototype um executable beacon chain so this is eth1 running on eth2 it works so it is teku and geth basically uh versions of um talking to each other and running um proof of stake eth1 and this is very cool and actually there is a blockchain out there which uh fabian uh fergus stella um who invented erc20 defined uh that standard he has put together his own blockchain which is

called luxo has just finished his reversible ico um they are running this they are planning to run this in production this which is yeah interesting uh this this teku geth combo so they'll be running a version of the of the executable beacon chain uh for their their blockchain so yeah yeah good things are happening and um the slow part about this is the governance right the technical side is is actually pretty straightforward michael's demonstrated um it works the slow part is the governance and we need to that that is you know softly softly um

a lot of value depends on this going correctly uh and people need to be completely convinced that it's safe and it's gonna work and it's not gonna be disruptive so the next n months uh are going to be about that uh and and a big step forward was taken yesterday when there was a good long discussion on the eth one all cordettes call about what the merge involved uh and that's you know the the the teams the sides eats to eat one are getting much closer together now do you think that there is some amount of if required to be staked so we can consider it a safe so because right now there is like three

million eve stakes that's like if you were to merge right now it would secure a two 200 million network um how much if do we need to have staked in order to consider this safe uh yeah it's a good question and i think nobody really knows the answer it's not half of all eve um and plus half of all d fight uh because that just doesn't make any sense um but in the ballpark figure which makes the protocol work um fully is is quarter um it's 250 000 validators whatever that is in yeah i can't i can't without four million each no

yes something like that uh eight million yeah that's right um and vitalik's actually proposed putting a cap on the number of validators that are active at any one time of of half a million um so that's somewhere between seven and fourteen percent of of all ether state and currently we're at what three percent or something so we don't need a lot more i think for for to be in the right ballpark i i'd like to begin wrapping up the call um and as we do there are four things that i'd like us to talk about and i will i'm gonna assign those really quickly um patricia if i can get you to talk about the co-app one more time michael if i can get you to talk about

the installation workshop that we had had talked about um nolan if you will talk about the uh upcoming call for the teku eth1 merge client uh and then after we wrap up um if if anyone who's on the zoom wants to remain on the zoom after we cut off youtube we will do a bright id um group scan so uh patricia do you want to go ahead yep um i think anyone who wanted their pow up already got it so so in case you don't know what it is a pop is a proof of attendance protocol every person that is part of an

event and an event can be many things um it's entitled to get an unfashionable token that works as a proof that you've been part of that event we've been running this thing for two years now it's been going well there are many pow-ups the each taker community has been a testing ground for new po functionality and and we plan on on running experiments here so if you didn't get your po up chime out somehow either here or discord or somewhere else if you did get your pow up stay tuned because we are working hard for making them more valuable for making them useful for different things

so so don't miss them if we are building the consensus layer of the feature the consensus network the distribute system our grandchildren will use it would be nice to have a proof that you've been part of the early days of it so so i see that only one person didn't get it tom reagan bogan i'm going to reach out to him and and if someone has any question feel free to reach out and and please michael tell us more about the workshop that we are doing in march all right thanks i really see poep as one of those public goods kind of things um and i was

actually pinging patricio privately about this the other day like when airdrop but uh and he said no airdrop anytime soon don't worry but it is one of those public good kind of things like pull together like uniswap like tornado cash like they are not in it to um to profit or win they're in it to provide a public good um and so there's a lot of potential in just getting interested and involved in poep and if you don't know about it yet find us anywhere on eath staker because we love to talk about it so thank you patricio and then upcoming within the east acre community for those of you already familiar with the community we've hosted one of these last year which is the validator

workshop course we hold a live session and we start from bare metal and spin up a working functional mainnet validator so that's what we did last time uh with sort of the next iteration of this what we're going to do is uh we followed some mayor esats guides and tutorials last time what we're going to do this time is create sort of an mvp just the bare minimum what you need to set up and spin something up and then those are going to be our two tracks so we're going to follow samara's guides and then the mvp guide and we're going to go through different scenarios getting used to some

of the more advanced feature sets of the deposit cli we're running through different scenarios what happens if you want to recreate keys running multiple validators or keys on validator working through some of the minutia and different scenarios so if that's something you're interested in uh if you're just interested in diving in and learning more about validators if you're new to the scene i highly recommend you check us out um that will be saturday march 6th so that's two weeks from today exactly uh next week wednesday which is february 24th we will make a formal post on the east staker website at easttaker.cc outlining

the specific date and times the link and how to access and any questions that you may have to set up we'll make sure that you're taken care of so that's my my spiel on the validator workshop i hope to see everyone there and our next event we're going to have the previously mentioned mikhail and gil on to talk about the tecu plus geth client and all about the merch i don't think we've nailed down an exact date for it yet but uh probably early-ish next month i thought we had a date do we not uh maybe we now i i don't think we do um

um we think that that call is going to be quite significant because the merch hasn't been yet explained to the people on the ground so so so stay tuned to it it might be it's likely going to be the first opportunity to really understand what's going on from the people building it and so we have confirmed it we just haven't established the time is that right yeah yeah okay uh and one more thing uh buddha if you want to take this um the uh do you want to tell about the hard fork naming uh are we is that something we can talk

about publicly patricio uh yes we can uh so yeah buddha yeah yeah sure on github there is an issue where you can post names for the upcoming hard fork you can be creative and we will choose i think we'll choose the top five and then post him on vote and people who own a poep or have been in the community and have a poet can vote on the on his on on their favorite one i'm just not sure if we are going to pick the top five or uh or not i think it's going to be more of the researchers and clients and it's not about picking the name of the next hardware but about picking the

naming convention for all the protocol upgrades that are coming to each tool so so there are some guidelines and and they are all on the if your github repo it's an issue by um icepair yeah one of the two researchers it has become a bit of a joke and i have contributed to the joke and just having some fun but if you scale back or zoom out you'll realize that this naming convention uh will probably last for many years so if you can make a great proposal and if yours is selected then you will have that claim to fame for many years

to come so aside from the haha joke it may be worth submitting some quality responses there and we'll put that link in the video description to the recording of this as well for those of you not familiar with where that github location is yeah and if you don't have any idea um almost all of us the easter egger people have made our own proposals so maybe you want to support any of them or or mine [Laughter] so uh ben and adrian we are delighted to have you guys today do you have any closing thoughts you'd like to share with us no i don't know that i do but uh it's

been you know it's been wonderful to be here i think it is a real joy to see the sacred community and the amount of help that you're giving people it's such a wonderful community to have um blockchain communities aren't always known for being so friendly and supportive so do keep that up it is it is a wonderful thing thank you thank you yeah totally echo that and um love the uh education side i'm an alumnus of the uh the first round of the um i forget what it's called now but the studymaster you're not an alumnus what are you say say what you are ben uh yeah the valedictorian is that what you thought my kind of final thought is yeah

notwithstanding the tecu branding is you know primarily aimed at institutional stakers it is i speak from personal experience super client to run at home i would love more people in this community to be running tech running a minority client is good for the network it's good for you um and uh you can find us on our discord um thankfully it's usually busy during my night so adrian carries a load of answering questions and stuff but you know adrian and i hang out there and some of the other devs as well so um that's on the consensus discord um and we've got a little community there is it actually a minority client uh it's hard to tell exactly what the proportions are but um i have

uh looks like about seven percent of the nodes are teku from a crawl that was done recently but i know that at least 14 of graffiti is teku so um and so we're a minimum of 14 percent of validators and it looks like most uh tecu nodes are running twice as many validators as average so that kind of indicates it tends to be larger operations using it but yeah so if 14 is a minority it's a minority when you say there was a crawl it's like someone had a bot poking and fingerprinting every note in the next yeah yeah so leo who's been around for a while from

from barcelona supercomputer center has written a written a bot that crawls the network and it seems to be fairly reliable but that counts modes not validators and that's different thing because you can run many validators per node that's cool to know i would love to get a better link to that because i only see the screenshot i would love to know the home for it i still haven't found it uh yeah it's in what's new in these two dude um a couple of weeks ago yeah no i mean i i saw that i but does that have uh yeah the source code is linked and the uh the repository okay maybe we should we should run the board in the age taker infrastructure for neutrality that's a good idea um and we could update the data regularly from that

rather than just snapshotting um yeah that's that's actually an awesome idea uh as we close i would like to let everyone know that ben and adrian are fantastic contributors to the staking ecosystem they they develop the tekken client is is wonderful it runs very smoothly and there's this idea that because they develop for uh institutional clients that it might be harder to get support but if you've ever reached out to them i can guarantee you that their support is nearly instantaneous and top-notch so um definitely one of the top tier clients

i would encourage anyone to try to stake but if you decided to start staking today tecus is a perfect client to begin with so thanks everyone for your time i'm really glad you all made it thank you adrian for waking up for us um and take care uh if if you're still on the call stick around after we finish recording and we'll do that bright idea link up bye everyone see you everybody