Ethereum Foundation 2020 and beyond

Transcript

good to go ready week great hi everybody my name is Albert and today I'll be talking about the if they're in foundations multi-layer strategy for supporting aetherium so first quick reminder of what we're doing here we're working towards a world where aetherium is widely used hopefully for good and we're aetherium values are widely found so why is this hard well there are a lot of reasons well one of the biggest is that none of us see the full picture of aetherium today it's awesome to see that aetherium already means many different

things to many different people I'd say that was one of the goals so this is good but it certainly doesn't make navigating toward the future any easier of course the picture today is already so incomplete it only gets blurry or five years from now another complicating factor but also good is how much the etherium ecosystem has left to grow even today etherium already has this vast vibrant ecosystem and it's already something that makes aetherium special i actually think it does more than make aetherium special i think it's one of the theorems super powers so what's the etherium foundations part in all of this

and what's the etherium foundations plan for all of this one of my favorite quotes that I cite often when talking about the EF strategy is from a book titled good strategy bad strategy you can read the quote but basically it says that real life doesn't work like the movies you know in the movies you have a hero they have a crazy elaborate plan they manage to dodge bullets by three inches and everything works out in the end this never actually happens in real life this quote is also very useful because as anyone who has spent even three months in this space knows the situation is nothing if not dynamic now the future may be uncertain but we

already know some of the things we need to I'm going to go through this list very quickly but I'm happy to talk more about it if you grab me afterward so just going quickly here we want to grow the usage we want to grow the community we want to grow the developer base we want to increase the developer output which I would describe as making existing developers more effective we want to grow the researcher base and we want to make sure that we are supporting the research that can make the future of aetherium a reality similarly there's a lot of development that goes into making the future of aetherium work and along the way we can't forget about security maintenance upkeep all of these incredibly important things that if we don't keep doing them also make it so

that theory ins future can't be a reality related to that there's just general downside protection of all kinds and whether you want to call this upside out-of-the-box ideas or explorations there are also many things of that nature that we want to do and by the way if you weren't if this didn't seem complicated enough we want to make sure that we do all of this in ways that are consistent with the theorems values whatever those turn out to be so this is quite a challenge but we at the etherium foundation accept this challenge because we have no choice so what is our approach to these problems first let's remind ourselves of what we're working

with we know that aetherium is growing regardless of whether the etherium foundation is adding subtracting multiplying or dividing a theory omits going to grow a lot faster and a lot more than whatever form the etherium foundation takes we also know about the power of the etherium ecosystem even the one that already exists today so again let's come back to this quote and the critical concepts in it how do we create options it involves a multi-pronged approach but I've realized that it might be easier to think of it as a multi-layered approach since in aetherium we're already used to this

multiple layers metaphor in other words just as aetherium as a system has multiple layers the etherium foundation and it's portfolio of allocations does as well and that's what I want to talk a little bit more about here today so let's walk through these allocation layers first off is what I would call allocation layer 0 these are the traditionally supported projects and functions when you think about a team inside an organization that's the model that fits here so examples would include you know go aetherium AKA Geth solidity parts of the etherium to R&D team a lot of eath 1x awasum remix the administrative side you

know the people behind the grants and ecosystem support program legal finance accounting general operations and and several other development teams that I've left off all here now how do allocation decisions get made at layer 0 for the EF well it's actually largely team lead driven it does have some similarities to how grant applications get evaluated but there isn't a notion of here's the direction from above that everyone follows rather teams and their leads are asked to submit budgets and roadmap proposals and others work together to evaluate these proposals review them and help improve them in practice it's been rare that massive adjustments get made so you can think of

this more as like a checks and balances system than a centrally planned one the decisions for how the resources distributed an allocation layer 0 are pretty distributed actually and in practice they turn out to be surprisingly equally distributed not really by design but you end up with I think 15 or 20 people across which the majority of the money and the decisions for where it goes get made most of the people are focused on their slice their team but there's about three to five people who are more focused on reviewing across slices mostly to check for obvious outliers as I noted in practice very few significant and adjustments get made in this process but a lot of minor ones do moving up to allocation layer one we have grants namely direct grants

because I'll talk about other types of grants later this is the the grants program or the ecosystem support program that some of you already know quite well and how allocation decisions get made at this layer is really driven by this central concept of producing what we call a support menu the support menu is used and evaluated and reviewed by ecosystem support team members as well as people who have expertise that's relevant to a given proposal now I can't really do justice to all the great work that gets done by the ecosystem support team but to give at least some sense of how this works I collected a series of screenshots from the main support menu template that the ecosystem support team

works off of by default so I'm gonna skim through these pretty quickly we'll probably move towards open sourcing this at some point but here are just some screenshots you can see here's the beginning here's a table that talks a lot about the goals we want to understand the tasks in the roadmap for applications we want to evaluate the impact the execution implementation the team we want to figure out what our options are we don't want to be restricted just one way of potentially supporting something we want to consider the alternatives and ultimately we also want to design a structure for the grant people think most grants are just hey put a bunch of money together and give it to somebody up front but a lot of the work goes in deciding how to allocate

the support and across what time frame and under what conditions and so as you can see as I continue to go through here there's all kinds of considerations that go beyond what you might normally consider such as social and second-order effects thinking about this future sustainability of the project because hey they might ask for something but it becomes clear that they're going to need more down the road and this all comes together to form a final recommended structure now we don't necessarily fill out every last part of this template for every single grant but these are all questions and concepts that matter and the folks on the ecosystem support team do an amazing job reviewing and suggesting improvements for an enormous variety of grant applications across a

lot of different domains now let's move up another layer or at least another half layer I think this is where things start to get a little bit interesting in 2019 the EFB to experiment with utilizing delegated domain allocators the general idea is simply that you want the best suited people to be in position to make allocation decisions and the best suited people don't need to be full-time EF people in fact it's unrealistic to expect that the best people would all be full-time EF errs the very process of hiring someone into the EF has its own nuances I'm sure it's got its blind spots and then the goal the EF is to grow its impact but not its organizational girth so I know I keep showing this slide but it's because I

want to establish a clear feedback loop between these concepts and the EFS approach we know we have this vast vibrant ecosystem and we know we want to create options that's just the essential logic behind this layer of allocation how do allocation decisions get made at layer 1.5 well it's a hybrid of what we've been doing at allocation later zero an allocation layer one the idea is that we try to capture some of the best concepts from the previous layers so we want to leverage the coordination and direction giving capabilities of more traditionally structured teams which you generally see with layer zero but we also want to harness the flexibility and

dynamic nature of eco system supports approach they're the ones that have had to figure out how to support a much wider variety of things in a much wider variety of contexts so just as you'd want capable leaders driving projects like e to solidity gap and so on for someone or a group to become a delegated domain allocator we're going to assess their ability to develop and drive drive roadmaps and just like with grants will often start with smaller budgets and smaller scopes and ramp up from there as people show that they can handle it the foray into utilising this layer of allocation has gotten pretty well so far some examples include the local grants programs we've done last year in Korea Japan and Taiwan also utilizing domain

level experts and leaders such as in the zero knowledge proof domain barry white hat and kobe gherkin or alexey hakuna for a lot of work in each one x on the developer experience side we worked a lot with nomic labs on the developer output side we've worked a lot with Sina Hebei bian and then as with aetherium or we've got just Josh Stark Sam Richards and others overall we're pretty excited to expand our utilization of this lair as part of the toolkit and I'm excited to see what that yields in 2020 last but certainly not least we have I guess you could call them third-party allocations or allocation later - as many of you know there are groups that have formed to

collect and redistribute resources in the etherion ecosystem sometimes they use dows to organize and coordinate and while right now it's probably not strictly necessary to use it now since most of them tend to consist of members that all know each other anyway it's still nice to utilize a mechanism that ensures one's ability to withdraw and one that can scale to bigger groups that might not all know each other so the functionality of these dowels I think will become necessary as they get larger there's also grant matching most notably via get coins efforts which have been really cool to observe these have already provided a lot of allocations that are at the minimum complimentary to what the EF does some of which have probably been quite controversial and

many have been the kinds of allocations that the EF couldn't realistically do directly which i think is great the difference between allocation later 2 and 1.5 is that at this layer the EF by design remains largely unappreciated it provides us it provides matching or direct funding but beyond that lets the entities at this layer do their thing so how do allocation decisions get made at this layer well like I said the EF isn't opinionated about the decisions being made at this layer but the EF does need to have an opinion on how resources get to this layer the truth is this is by far the least developed part of the EF

allocation portfolio and methodology at the moment but it's something that we'll be upgrading in 2020 we'll be incorporating best practices from each of the other layers to assess the different redistribution groups or dows as well as the matching programs like I said we're going to remain on the opinion unup enya nated because that's vital to ensuring that these other groups can provide the diversity of allocations that really maximize their impact but we're still going to apply a lot of the concepts from the ecosystem support program and delegated domain allocators to review the process that these groups go through to make their decisions and to review their overall decision-making capabilities so again I'm personally really excited to see what happens as we begin to solicit

the formation of more layer two allocation groups and matching programs that are targeted towards different problem domains may be different geographies and may be different other ways to slice up this huge ecosystem now I'll do a quick review of the strengths and weaknesses of each layer layer 0 is the only layer where financial support is by default recurring that's obviously a pretty big deal it's great because it allows teams to comfortably focus on longer-term roadmaps but it might not be as great in the sense that it changes the accountability dynamics and impacts the longer-term flexibility of the EF as for layer 1 which is traditional grants those are nice and that they're kind of the Swiss

Army knife of allocations and when in doubt this is what we use of course the Grants Program is still largely driven by a single in-house team and subject to its blind spots there's another challenge which is that the EF has a minimum fixed quantity of downside risk attached to just about every grant that it gives this means that you have probably can't experiment as aggressively as might be ideal because of the scrutiny that comes with each grant now you might be able to say hey we we thought that there was only a 25% chance this would work but that would be valuable so we allocated a big chunk of money towards it anyway and that might sound reasonable but I promise you that when talking with prospective grantees the conversation often involves the applicant pointing out a past grantee that they think they're better than and

who therefore deserve more money then just at the Stanford blockchain conference a few weeks ago I'm not going to name the project but they cited a grant from mid 2018 as proof that the EF must fund their project since their project is so much better than that other grant so clearly this layer can't handle everything now layer 1.5 that delegated domain allocators could be the best of both worlds but if we don't do it right it could be the worst of both worlds I'm optimistic but it's challenging to ensure allocator accountability to the same degree then when the alligators are in house I think this is worth the trade-off or at least the bet but it's no sure thing that our efforts will pan out and lastly layer two will I think consist of increasingly significant

allocations over time now one challenge is that it can devolve into popularity contests and related issues but I think by diversifying the set of dows and matching programs and possibly having them feed funds into each other we'll be able to reach a healthy equilibrium so to wrap up is one of my final slides I just wanted to show an approximate history of how these different allocation layers have been used note that distributions and allocations aren't exactly the same as these numbers are based more on cash flow note also that 2018 is kind of anomalous because many major allocations were made near the peak of the market and it's reflective of how you can't just spin up a bunch of team of teams overnight but you can make some bigger bets again these are pretty crude numbers and 2020 is all projections they might not look

like a lot is going towards layer 1.5 or layer 2 right now but in both cases we're talking a pretty significant ramp up compared to past years and if we see quick success with these layers that's where excess funds are going to go in the future so in closing as we all know there's a lot to be done but over the course of this year we're going to share increasingly detailed breakdowns of how the EF allocates by category and by layer all in service of this simple and important goal of supporting the etherium ecosystem and while in this talk I focused on allocation layers the F is also working on a lot of other exciting things you're going to want to keep an eye out for these are some of the things that we think the EF is best suited to do on behalf of aetherium and

as always we try to capture those opportunities so yeah as always the EF will explore word to doing its part to supporting the etherium ecosystem and the growth of this vast and vibrant ecosystem in the years to come always like this graphic and thank you thank you very much questions I'll move back to this slide because it's cool yeah thank you thank you very much for your talk if you could actually go back to one of the first slides where you show the different meanings to people over helium I wanna remember sure which which one

the one were there is the tearoom symbol in the middle and all the different meanings that people give to ah yeah yeah I just want to take advantage of a representative of the Team Foundation to be here to ask you the following question yesterday there was an interesting talk about this American guy who was saying that actually having different meanings of etherium is actually it's not helping the community so he was arguing that we should have only one explanation of what the idiom means you know so could you please kind of share with me your opinion on this you know I definitely understand the temptation to want to have one explanation of what aetherium is I think

in practice it's probably unrealistic at this point and so that's what we have to work with it's not that I'm necessarily for or against it but even this is barely a sample of what aetherium means to a lot of people and then when you factor in the challenges of communicating about it to people around the world in different languages and with different perspectives I've generally thought that it's unlikely you're going to be able to do that unless you really really really narrow down like overly I think narrow down the scope of what you wanted them to be that said it is an impediment to people understanding aetherium and getting more interested in it my hope is that there's middle grounds I don't think we've like fully explored the space of how to describe or talk about aetherium and with therefore we should give up on

there being multiple concepts but I do definitely understand the frustration I mean I certainly experienced it myself when I try to talk about what if the M is to some people they love that it can be a lot of things to other people they're like I don't want to hear that it can be a lot of things I want to hear what it is and when I encounter people like that I do wish that there was a crisper way to describe it I think that's all the time we have cool thanks again everybody [Applause] okay I'm you see you never