Eth 1.x Explained by Hudson Jameson (Devcon5)

all right a theory m1x let's talk about it for you those of you who don't know just a quick intro at the etherium foundation I worked at a bank for a few years I have a wife and three cats and I mainly work on community security and DevOps at the etherium foundation so if you see me around please say hi I'm friendly so the thing I want you to take away from this is that eath 1x represents sustainability that is a really cheesy hashtag that we came up with at one of the 1x meetings and I really like it because it describes how we're trying to keep the chain together and how we're trying to make aetherium 1.0 last for as long as necessary and keep the network running and efficient for everybody using it Oh before I start this there's gonna be a time for Q&A at the end so please get your questions ready we'll have mics at the audience probably just in the aisles I believe so when that happens I will let you all know oh it's actually gonna be at the front so line up at the front when you have questions once my presentations over and Wow that caught that code here's the beginning we had ad hoc meetings in Prague during DEFCON 4 there was about 25 people there you could see some of them in that image and we talked about how to achieve sustainability we were saying you know there's a lot of state bloat there's things going on that are gonna like fundamentally change aetherium when 2.0 comes around how can we best prepare for those in 1.0 while keeping the network very very active and are not active but very efficient very optimizable for people to use so from that meeting in Prague we came with four working groups the simulation working group state rent working group the chain pruning working group and the ewaz and working group so simulation working group was put together to understand how can we look at how much state bloat is affecting aetherium what can we do about like coming up with graphs figures numbers to make it help us understand more the state rent working group which has now turned into the state be working group which has now turned into the like another whole working group kind of a thing is about addressing the growing state maybe advance snapshot sinking and stateless clients are also explored later on there's just a ton of stuff like eath 2.0 Danny's going to talk about this after me but if 2.

0 is gonna have a lot of changes and figuring out how to store information in a way where it's not going to blow up people's hard drives so can we address some of those problems now that I mean the fact is people are just storing this data for free and it's you know a difficult thing to incentivize and figure out what to do with you can't even run in aetherium client on a regular hard disk drive anymore it has to be on a solid-state drive because of the disk IO reads or the disk IO so that's that working group we have chain pruning can we archive blocks can we archive logs how many of you here adapt developers keep your hand up if you use logs in your DAP keep your hand up if you're a cheater and you're using logs because there are cheaper storage than on chain storage okay yes that's exactly why we might need to archive logs it was never really like explicitly said if we're going to have logs last forever they're supposed to be for error messages and for just you know tracking certain things but now that they're a cheaper way of storage that's unbalancing things a lot and it's causing some state bloat so can we archive those can we maybe make it so that you have to fetch them from an external source what we're looking into things like that and then there was the awasum Working Group that determines what for me was going into 1x and 2.0 a lot of he was and workshops and talks are around Def Con so feel free to jump into those for more information so at the time of Prague but basically between late last year early this year meant a renewed effort to invest in the existing eath 1.0 network to ensure it can survive and thrive indefinitely so that's what I alerted to earlier that's the first version of what sustainability is I'm gonna get to the second one later so if 2.0 is sexy right now it's like the new hot thing everyone wants to work on it but we still have this 1.0 chain and up until you know last year in Prague there was some kind of discomfort and some like things that we really need to address about the fact that developers were dropping off doing stuff with 1.

0 to do stuff on 2.0 and we need to make it so that 1.0 is still sustainable and 1.0 is still fun to work with and has goals and like a know its own driven architecture just like 2.0 has so not that we don't like 2.

0 but it is let you know the sexier younger brother of eath 1.0 and we need to make sure that 1.0 and one point X stays in the forefront until we can get to 2.0 ok so this was kind of I don't know if how many of y'all remember this controversy from last year but like the Prague meetings that I talked about the ad-hoc ones with 20 to 25 people a lot of people said those were secret meetings and the reason for that was we did have this note sheet passed around during it that I think someone from consensus took and they basically took notes down and they were released later or they were going to be released later but they got like leaked early because some one of the core developers thought that people should know about the notes before we were ready so we were talking about one X in Prague it was like 20 people in a room there were other ad hoc meetings all over the place and so people interpreted this as oh these people are coming together to figure out what's happening to aetherium without anyone else's input this isn't like the core developer meetings where it's all open and out there and streaming so we did an experiment because of the 1x initiative starting we wanted developers to be able to speak hand without fear of media misinterpretation sorry Li sorry Kim with Coyne desk I know you're out there but yeah without fear of media interpret interpretations so we had an eighth one x-coordinate meeting on the 30th of the November of last year to solidify plans for workshops and things like that but we use Chatham House Rules and what that means is everybody attends the meeting and you take notes but you don't attribute those notes to anyone in particular what you do is you just write the notes down and no one everyone can talk about what happened at the meeting but they can't mention names so we decided to just try that we didn't record or stream the meeting and the community pushed back on that so we decided to not do that again so that's kind of the end of that but it was really interesting to me and I put it with us in the presentation because this kind of jump-started a lot of enthusiasm for you know transparency and really showed that like showed us what the community values within aetherium and I thought that was really important we had a 1 X meeting at Stanford that was streamed and recorded in January and that one had a lot to do with Alexi kind of running the meetings and doing a really good job of getting us all organized and talking about the things we needed to do in 1.0 to achieve sustainability and on chain pruning and stuff like that Berlin I didn't get to go to that one it sucked because I didn't get to go but in April we had another meeting that was also streaming reported you can find all this on youtube it's hours of really good content on one point oh and one point X and a lot of different stuff came away from that including more working groups we had EVM evolution which is improvements to the current EVM around speed safety and and operability that ones currently not doing anything right now that one got some feedback from different developers that they didn't really want it to be put in so that working group is no longer in existence as far as I know there's the testing and infrastructure team they're kind of have a working group where there you build these tools these cross client testing tools and other things for compatibility there's the finality gadget which is using the 2.

0 beacon chain to finalize eath 1.0 blocks which is a really really cool idea that one's actually about to spin up a lot more that's gonna be a lot more news from that coming soon from what I've heard eeew 1559 the fee market change that one if you've seen Eric Connor Eco gnar orbit Alec deuteron talking about that that's making transaction fees more stable and predictable by splitting the transaction fee into two different types of fees one of them is burned on chain and the other one is the predictable fee so after all these meetings it's been about a year since Prague where are we at with sustainability what it what is it now and I'd argue that it's about improving processes and making protocol changes to ensure the sustainability of the etherium 1.0 blockchain and what I mean by that is unlike the version 1 sustainability where it was just about you know let's keep the chain alive let's you know do some stuff with the reducing state bloat and rent and all this other stuff now we are looking at improving processes making protocol changes making things more energetic and exciting in the 1.0 chain and I've been seeing a lot more participation because of that I think that people are getting really excited about what we can do today on aetherium and without having to exactly know what the future brings this is an example of the current AIP process though so this is like my favorite or one of my favorite improvements right now which one yeah a theory in the IP process so I'm an e IP editor I'm involved in that and right now it's kind of hard to jump into the EIP one is our technical standard for how to submit a change to the etherion protocol or a community standard and it's really hard to Digest right now it's kind of long there's all these different IPS you have to format them a certain way and we want to make that a little bit easier so right now we have work-in-progress draftee IP accepted e IP main mint client implementation network update and then the miners mind the new blocks and then we're at consensus and what we do is we take these e IPS and we stuff them into these hard Forks and they're on a certain date and time and we do that my block number and that's not really efficient because it makes us rush we don't have test cases up yet we have to pressure everybody to work really hard and really fast at this you know normal this original model of how we do things today instead we want an e IP VIP focused network upgrade model where we have our draft and then we have the all core devs green light these Cori IPS from there we get into the implementations from there we get into testing and then it's only accepted after the test cases and the implementations are finished or near finished and what that's gonna help is instead of us deciding early on like okay we're gonna put the IP 1 2 & 3 into this hard fork on this date so we're gonna have to rush to the finish line now we just do a hard fork when the EIP is done and if there happens to be more than one AIP done at the same time or there's like really ones that are really close together that's great some of the challenges behind this are gonna be letting clients exchanges people like infrastructure projects like in fira and cloud pod front know when these hard Forks are happening so we're gonna have to bump up our communications we're gonna have to ramp up and assign people to know when these things are happening and that's actually a job that the etherium cat herders which I'm also a part of is going to hopefully have a really big job in accomplishing here's another really cool change this one is a graph from Alexia Kuna and this is how whoops this is how things want to work in the future metallic or people on the internet or other core developers have an idea and they're smiling and they like I really like this idea so they're going to put an improvement proposal out there maybe on a theory of magicians form but it's not formal yet then there's a working group formed around each proposal and research project and then others can optionally participate there can be client implementations other improvements but it's not like we're having every core developer look at every EIP we have these working groups of people who are interested and they present their case and they do the research and the simulations and the testing cases and the implementations and work with the team and this is a lot better than how we've had it before where the core devs have to look at every IP I want to make a note here that these improvements that I'm talking about aren't done yet it's a work in progress and we aren't even at consensus as this that this is how it's going to be we kind of have to come to rough consensus about this but this is definitely what I'll be pushing for as we go forward into this new era of eath 1x so how is this all gonna get funded we all know that there's all these grant programs out there for e 2.0 and different tooling projects the etherium foundation earmarked 8 million dollars back in May for the next year of eath 1.

0 and one point X research that was announced at ether real and we're working to bring full-time eath 1x teams and coordinators together that's also a work in progress that you'll be hearing more about in 2020 so yay we got funding we have excitement we have everything we need to really bring the eath 1.0 chain into its next evolution before we hit 2.0 [Music] so if you want to know more google Alexia two knobs medium his blog has a ton of good post on that there's an aetherium wiki page that has the eath 1x history on it it talks about the Berlin meeting and the Stanford meeting and it says links to the YouTube pages also check out the PM repo on aetherium github that's the one where we have the bi-weekly core developer meetings that myself and a few others help coordinate and run for every core developer meeting that happens every two weeks we have a full transcript of notes that's in the PM repo and you can go back three years and see notes for every meeting also YouTube streams and live recordings of almost every meeting contact me if you want to help if you're like a low-level developer this stuff is more like not for people who build apps necessarily except to get their input this is more for like low-level client developers and people looking at working on the protocol or the networking layer but if you're interested in that and don't know where to start please reach out to me I can connect you with the right people and I'm not like an angry frustrated person I'm happy happy to help alright so everyone thank you so much start to come up to the front for questions the microphones over there and yeah anybody got questions if not I could talk about my table top RPG podcast but that's like very unrelated so I don't know how many people are gonna enjoy that Oh some keep some fans hey what's up introduce yourself and then pose your question hi I'm sorry I'm wondering what's the soonest that yeah if you wanted five five nine might be implemented awesome the IP five five nine one five five nine that was on this one the fee market change that's a good question for vitalik and Eric I'm thinking 2020 they can definitely go in we still have to come to consensus around it but I have not seen much opposition personally for it I think it's a really important change that a lot of people are very you know like they really like the idea of that it will change the user experience a little bit and the experience for Daffy Debs so that is something to take into consideration but otherwise I think that that's a really good idea you