Saving Web3 Protocols from the fate of Web2 Companies with Maggie Love
Ethereum Denver·Mon, Oct 9, 2023, 12:00 AM
This talk will focus on showcasing the challenges of relying on centralized cloud providers, and the promise of decentralized cloud infrastructure for protocols. It will have charts, numbers and analysis to back up all claims. It's important that Web3 protocols are not just decentralized in name but decentralized in infrastructure for costs, resiliency and performance. Summary below: Cloud is one of the greatest shifts in the history of computing. For the first time in history, organizations didn’t have to deal with the technical burden and costs of running their own infrastructure. Instead, they could rely entirely on a cloud that offered infrastructure on-demand, at any scale, to organizations of any size, driving efficiencies and economics. The cloud industry is growing rapidly growing rapidly on a base of over $100B of annual public cloud spend. Yet, what started out as the most efficient and economical way to scale has become the largest area of spend for many Web2, and dare I say Web3 companies operating at scale. We will get to Web3, but first, focusing on Web2. The longterm cost implications of shifting entirely to cloud is costing Web2 companies billions of dollars each year. “Across 50 of the top public software companies, an estimated $100B of market value is being lost among them due to cloud impact on margins.” a16z. They could be saving billions of dollars they had run their own infrastructure or taken a hybrid approach to their infrastructure. Not only is it a challenge when it comes to costs, but these organizations get cloud vendor lock-in, and it’s difficult to move their tech stack or start from scratch once they’ve reached scale. Web3 protocols are in their early days, and the decisions they make when it comes to infrastructure is critical. Many protocols today are dealing with the same challenges that Web2 companies face with large cloud providers, and they are looking for ways to decentralize while they try to scale or die. There are challenges with operating costs, lock-in and an inability to operate efficiently with the hardware that cloud providers offer. Web3 protocols and their infrastructure ecosystem (node runners, operators, shakers, future zk provers, etc.) have to consider not only the short term ease of spinning up on a cloud provider, but the longterm implications for their protocol. This talk will start off by covering the above, and then it will go into an analysis of different infrastructure options and their associated challenges, benefits, costs. In contrast to the centralized cloud providers Web2 companies and many Web3 companies rely on to start running infrastructure, there is are infrastructure options that are decentralized, align with the values of Web3 while saving protocols money now and in the future. This will not highlight any specific Web3 company -- just compare and contrast data and analysis of different infrastructure options. Maggie Love
Transcript
I'm giving a talk called saving web 3 protocols from the fate of web 2 companies about me my name is Maggie love I'm the co-founder and director of ecosystem and Partnerships for web 3 Cloud spelled w3b clo d and a bit about me I started um I went to Northwestern University and I actually started out as a journalism major and all of my teachers and peers in the field seemed really unhappy and I was like well that doesn't make sense to me I want to be happy if I'm spending all this time and money getting an education so decided that maybe journalism and that type of Storytelling wasn't for me and then I ended up at IBM I did Consulting for a couple years once again looked around and my peers and uh those a couple years older than me had families never saw them weren't that happy so I decided again maybe that's not for me then I jumped ship in 2015 to start doing product strategy for IBM Watson and although Watson is not here today ai's really come a long way so that was a great experience and specifically I was focused on Watson in financial services so building out Financial Services products leveraging AI technology in 2016 I was sitting in a corporate strategy meeting and I heard a guy on the phone who wasn't in person saying blockchain blockchain blockchain we must be doing something with blockchain at IBM and everyone's like oh mute him he's so he's so annoying he won't stop talking about blockchain and I was like wait someone's excited about something in IBM I want to go learn about that so I went after work that day to Barnes Noble and Union Square New York picked up the blockchain revolution read it and decided I had to be working in this ecosystem hi Kirk um and then so luckily enough I spent all my time trying to find blockchain companies to be at and found consensus in 2017 through a LinkedIn search a friend of a friend met me for coffee took me in the flat in Bushwick and unlike all of my other you know previous careers and even college degrees seeing people unhappy everyone at consensus was happy bright-eyed passionate about the future Reinventing things wearing street clothes I was like I have to be working here so joined consensus in 2017 and shortly thereafter Joe Lubin asked me to put together a presentation for Advanced Micro Devices the chip maker on ethereum the history the timeline to Serenity proof of stake what was going on in the infrastructure space and uh you know for the CEO of AMD Dr Lisa Sue an amazing woman and I had to go look on Reddit because vitalik was basically posting all of his updates on Reddit and there was no ethereum.org Joe thought I did such a good job on the presentation he asked me to close the deal so web3 Cloud what I'm talking a bit about today and what we do is a joint venture between Advanced Micro Devices and consensus and I'm not only a co-founder but I'm director of ecosystem and Partnerships and then in general I'm really passionate about permissionlessness and the inclusion that it can provide in the web 3 space but I noticed that we do need more communal based learning and true onboarding into web 3 and D5 so I also founded shifi 2019 do not ask me about White Lotus I have not watched TV in a while but you can ask me a lot about web3 just a bit about me and so the goal of this talk today is to convince web3 to make decentralized infrastructure a priority in first class metric yes decentralized infrastructure is not culture it's not defy it's not that sexy but it's really the base in what is necessary to have these Global resilient networks that we are trying to build and all that fun stuff on top doesn't really work if we are centralized so maybe I'm going to make decentralization sexy again so in 2011 Mark Andreessen wrote a very prolific piece called software is eating the world how many of you have read this remember this piece awesome and then about 10 years later his firm a16z put out a piece called the cost of cloud a trillion dollar Paradox so looking back on yes software is eating the world but now software can be costing companies trillions of dollars and so we all know that the move to Cloud was one of the most important Innovations in the web 2 world that we've seen today that allowed all these engineers and software developers with great ideas not have to worry about learning Hardware infrastructure running it and maintaining it and it is what has allowed all these companies to build uh to be to scale and be very large and not actually own any of the hardware but that now is coming at a cost and so an estimated of a hundred billion dollars of market value is being lost across the top 50 Public software companies due to Cloud impact on margins and so although going to AWS going to a cloud provider is easy and accessible and helps you scale in the beginning now when you're at the scale of some of these companies like Dropbox they're paying 75 to 80 percent of their revenues to AWS and even Dropbox has decided to do something about it so web3 protocols mainly focusing on software also need to be thinking about the long term let's dive a bit more into the billion dollar cost of a cloud and it's actually billions and we know as I said the cloud is one of the most significant platform shifts in history I.T spending on public cloud is growing on a base of a hundred billion dollars annually the cloud Market is not slowing down all I moderated a panel this week with Microsoft Oracle and Google they're here for early stage company Cloud drives efficiencies and operations and economics and skills you don't have to worry about hiring someone to maintain your on-prem infrastructure yet when companies mature the cost of cloud drives lower profit margins outweighing the benefits and then they're at a point of scale where they're really not sure what to do how are you going to move all that infrastructure back onto your premises and an estimated eight billion dollars in aggregate is spent on Cloud for the 50 50 of the top publicly traded software companies so this is just setting some context of what's happening in the web 2 space what their current fate is being very tied and very financially dependent on AWS no matter what the cost so some solutions that have come up from this article and then we'll take it a step further with web3 is cloud repatriation that just means bringing your Cloud infrastructure back on premise or Cloud optimization using a mix of cloud offerings including third-party third-party Cloud efficiency of solutions or moving workloads to special purpose Hardware can save companies billions of dollars definitely we want you to remember this special purpose Hardware as we go into the web 3 portion of this and so for the 50 top publicly traded companies software companies if they did move their infrastructure on premise they could drive a 50 reduction in Cloud spend or a net Savings of four billion dollars and public Cloud list prices can be 10 to 12x the cost of running one's own data centers that's because we know AWS offers all these great features billing accounting you really don't have to worry about anything they take a lot of the software pieces off your plate too and so too long didn't read tldr infrastructure spend should be a first class metric not only for web 2 companies but also web 3 companies and web3 protocols need to optimize early and often outside of the cloud and dhh do you know who he is he invented Ruby on Rails uh he was just recently released a port about their Cloud spend and they are going to build all of their own infrastructure in-house so repatriated all away from AWS and estimate saving seven million dollars over the next five years so that's not a small line item that's not something that's you know just a couple zeros it's many many zeros and I can share the link to that in this tweet after so what has occurred today and actually something that I think many of us are mission Envision aligned to get rid of is something called the cloud oligarchy you have Azure you have AWS and you have Google cloud not only is our software on them our traffic is on them they are closed gardened clouds and they really control how things work they drive the costs the margins and also have extreme power over many different parts of our internet experience and they have a five trillion dollar market cap combined and so right we're here this week and this weekend and we've actually been here with the Genesis of Bitcoin because we believe that crypto can reinvent the internet and even the rails of the internet not just applications not just Dows but starting from the very way internet is constructed and McCormick said that web3 is the internet owned by the users and Builders orchestrated with tokens so if we compare web 2 to web 3 web 2 is known as the platform economy the closed Garden economy and we know that web 3 has been called the ownership economy and why web3 is called the ownership economy is because now we can send value across the internet atomically with bits so in web 2 we could read and write we could send bits of information we could interact with each other but now in web3 we can read write and own that's also from Chris Dixon I won't pretend that one's for me and this ability to own things on the internet from the infrastructure to the organizations you participate Aiden to D5 protocols is what has inspired all of us to be here I think there's 20 000 people here and there's many many more of us building all throughout the world and we know with web 2 the key aspects are centralization the cloud oligarchy it becomes expensive for developers and companies over time and corporations retain all of the value generated by all of us Builders and users of the internet and we know with web3 we're trying to invert that we're decentralized we have crypto incentivized protocols and because they are using tokens we're actually giving value back to the users and Builders and we have a native value in this internet and that is what's truly new we did not have value uh from web one to web 2 and now with web3 we actually have an internet of value and you'll see on the bottom the ecosystem participants are some of the Big Two web big web to Cloud providers I'm sure many of you in your head are feeling hopefully thinking like oh no I'm on a web 2 cloud provider um and then on the web 3 side we have all these internet native infrastructure protocols like ethereum like file coin rweave live peer that are trying to incentivize at the infrastructure layer to build a more Equitable and value-driven internet and I want to once I want to again step back and look at our matriarch the Bitcoin Network and what Bitcoin did was just not invent a Global Network payment system and actually invented a global distributed internet infrastructure Network Bitcoin pioneered proof of work I know that a lot of us want to move away from that for many reasons but by pioneering this it the software itself distributed token rewards to coordinate and incentivize individuals to deploy and operate physical infrastructure and networks I think we forget about that in the software world but if Bitcoin hadn't used code to distribute rewards to coordinate physical and to coordinate physical infrastructure it wouldn't be this Global distributed highly resilient Network and so that was really the magic of Bitcoin how can we build resilient censorship resistant performant Networks and so now what blockchain networks offer and what many of you are tinkering with I'm sure is this blank canvas for token distribution and crypto economic models to facilitate coordination blockchains are often called the coordination machine and so what Bitcoin did is proved with the right incentives one can coordinate Capital machines and labor without a centralized entity Google is not paying people to go stand up infrastructure data centers it's not a salary based thing code is incentivizing people from all over the world to go buy Hardware do the labor plug it in negotiate electricity contracts all of that just done through code and I think that is what is so inspiring right we like this idea that code can replace the intermediaries and make much more efficient markets for this and so in the space now that we're heading towards uh one classification of protocols is decentralized physical infrastructure they notice that the right token rewards allows a network to bootstrap the supply side to the point where end users find Services attractive to use so they want to bootstrap people deploying the physical infrastructure and then it gets to a point of scale where people using that infrastructure find it find it a good experience it's not always that way in the beginning but this was really The Genius of Bitcoin outside of this global payment Network and what we're seeing in the regulatory space is because Bitcoin is truly decentralized and global they are in the United States allowing it to be a commodity so it's something to think about and so really quickly let's talk about the evolution of the cloud today we're talking about AWS Google Cloud Microsoft really that next wave in Computing that got us all off of running our own computers in in our own premises right so clouds are actually just running a bunch of computers I call them computer aggregators and then we have this idea of hybrid Cloud when I was at IBM we talked about this a lot banks will never go into the cloud they will never let other people run their cloud and they realize that they couldn't compete with AWS toe-to-toe so they were like let's do a hybrid Cloud let's try to keep some of our clients happy and then you're also seeing some players in this space and now we have web 3 with web3 Cloud we actually called it web3 cloud in 2018 before web3 became the popularized term and then you have a lot of these protocols uh still Distributing tokens for infrastructure providers and this is what is being called decentralized physical infrastructure networks and then you also have the public Bitcoin miners so in the web 3 case the difference here is that all these infrastructure providers are examining protocols understanding that the network is trying to bootstrap the supply inside going out and putting huge investment in the infrastructure and creating actually decentralized Network Works your network is not decentralized if it's running entirely in AWS I'm sorry to let you know that today and so the bet that we took in 2018 with web3 Cloud was that the cloud Market is going to be or the compute footprint is going to be just as big or bigger in web 3 than it was in web 2. and it's important that web 3 native organizations or projects or protocols actually own that piece of the Internet it's important that we you know in this world that the infrastructure who owns it where it is really matters can't be shut down if it's Global and distributed and our thesis was web3 software we'll need web3 Hardware and really quickly this is just an example of all the use cases you might be aware of that actually need massive infrastructure so defy even nfts where they're stored matters if they're not stored in a decentralized cloud what happens to our nfts the metaverse uh decentralized metaverse because AR and VR already needs massive compute we'll also need compute and then the Enterprise space will as well and it's important to note that there are protocols working on decentralized rendering we know there's tons of nfts existing on many of the protocols in D5 as well and it's important that that infrastructure continues to scale with the demand from these different use cases and so what is the value of decentralized infrastructure why do companies need to be thinking about this today when they're starting to go and run their infrastructure somewhere else or they're thinking about how can I incentivize decentralized infrastructure it's highly resilient we know this no one at this point can shut down the Bitcoin Network it is probably the most resilient internet infrastructure out there ethereum too but we're going to look at that in a minute performance so when you are incentivizing multiple providers to run your internet infrastructure they're starting to compete You're Building these Cooperative yet competitive clouds that's also the genius of how blockchain works it can coordinate competitors to cooperate to scale This Global infrastructure so many times you can get performance benefits from providers trying to compete for your network to earn the most Rewards from your network cost so you yourself can drive the cost down if you're not running all of your infrastructure and you're Outsourcing it through these token distributions to many providers you're going to run down your costs and they're going to think about what is the cheapest most performant hardware and how quickly can we get it stand up to start uh participating in these networks and then there's no vendor lock-in so in the web 2 world it's very very hard and even the web 3 protocols running on AWS I won't say any names are so tied into the AWS API that they could not even begin to think to get out of AWS and we know the points of centralization in the space they are just so tied in and vendor locked into these clouds with web3 there's many ways in many software Technologies being built where you can stake with different providers but keep your keys and your wallet so if you want to get out of staking with that whether it's an RPC staking or a traditional proof of stake staking you can just move to the fast Cloud so once again this competition in November lock-in really benefits the software companies protocols and the users and then many times different bare metal infrastructures are working with smaller providers offers greater control than the cloud because the cloud has all this software on top it's actually gives you more control if you're a software developer to be building on different types of Bare Metal Solutions and so what's the reality for web 3 and why is this happening I have to point out to my beloved ethereum community that the number of ethereum nodes hosted on AWS went up from 53 in August when I first tweeted this out and got a lot of blowback from the community that I very much believe in and it's even increased now to 70.9 percent if you're not aware uh the ethereum network doesn't really penalize individual uh validators for your node going offline but if more than one-third of the nodes goes offline at one time so if U.
S Virginia East gets shut down then all of those nodes will be slashed until they are less than one third of the ethereum network so the ethereum network can start working again so if you're staking with the provider if you're staking with a large institution you might want to figure out where they're hosting their notes because if it's on Amazon and for some reason there's an outage your tokens against those validators will start getting slashed and so why is this happening in web 3 well the same reason it happened in web 2. wab 3C has seen such explosive growth that the software developers building these different applications on top are trying to scale with the demand they want to be first to mainnet they want to be the one winning there's many competing services in this space so they're like oh we'll go on Amazon now and figure out decentralization later but then they get so tied into Amazon they leave that in the back burner but yeah it's a I do believe this is a problem in the space in that if something happens to AWS we're going to see a pretty negative events on the performance of the ethereum chain and so then what are the other risks of not having decentralized infrastructure like I said I keep joking in outage of U.S Virginia East just because that is a very popular AWS site so it will cause performance issues not only in the validator space but also in the RPC and Gateway space so how your wallet actually interacts with the blockchain you'll notice like I can't get my transactions through I won't name any names or what if it's turned off what if there is a catastrophic incident uh you know data centers uh have a lot of protections but what if there's some cataclysmic physical event that we can't imagine once again it would cause uh challenges in many of the blockchain networks and not just ethereum many of the other proof of stake layer ones really sacrifice decentralization for performance censorship so we've heard about this a few times in the past year uh different service providers on the network were censoring certain transactions or not allowing certain geographies to have transactions and without optionality without multiple infrastructure providers and service providers in the infrastructure space offering access to these communities then there can be censorship as I mentioned there will be the slashings of nodes in a group that are more than one-third of the network if they go offline until they are less than one-third of the network be a lot of financial penalties and I Joe we joked today that Jeff Bezos owns your nfts so I'm sure as you know an nft is actually just a certificate it's not the image and likely your image your video your clip is stored on a centralized storage or a cloud provider and the only thing on the blockchain network just says you own it but their certificate is not going to do you much good if you can't send your picture your digital good whatever it is to somebody else and we have a lot of regulatory like I don't know Scare Tactics in the US I feel like at this point but it's true Gary Gensler made a statement earlier this year that since 70 of ethereum validators are in Amazon in the United States that technically the us would have jurisdiction over the ethereum network so whether or not that's just him performing or you know saying he's going to actually take an action it is true that this can happen and this is why Bitcoin gets away with being uh you know this Global distributed Network and ethereum and other blockchain Protocols are getting uh you know threatened or told that they're going to be coming under U.S jurisdiction and once again just to point out that even though AWS is really easy and it presents is this one click button to start running your nodes start running your infrastructure our Kevin owaki the founder of get coin and super module started staking on AWS and it was costing him a hundred dollars a month plus a thousand dollars a month to host on AWS so he took it to start running his own ethereum nodes in home and if you're curious what that's like I have a different presentation on that from Devcon but this is where web3 is now and I think the biggest challenge in this space is running your own note at home and money is of your own infrastructure at home does take technical chops and for a lot of us who are new it takes 32 eth which is not incredibly accessible it will become more accessible with distributed validator technology but then we're all running inside of AWS because it's easy and our uh large institutions that are offering staking or even more decentralized aligned ones are still inside of AWS and it will start to cause these problems so what I really want is if you're working on a protocol and you're calling yourself decentralized I want you to think about maybe how you can actually have decentralization go all the way down to the infrastructure layer and these are some of the protocols and use cases that we think are best set up for decentralized infrastructure so access infrastructure this is known as RPC our Gateway so this is how you read and write to the blockchain uh pocket network was one of the ones to First spearhead decentralizing this and then there's many others along the way compute infrastructure we're really excited about the zero knowledge space because as I mentioned in the beginning that application specific Hardware is a good choice for these for protocols or companies and zero knowledge technology is actually going to Spur innovation in application specific hardware and none of that Hardware will actually likely run in AWS in the beginning it might because they're using traditional servers but in the future we will see more of those decentralized Cooperative networks of provers keeping it both decentralized and your information private storage infrastructure has really been spearheading this so filecoin is actually made up entirely of decentralized entities all purchasing hardware and deploying it against the network and then you know serving the network to store files so there's a couple others in this space as well and then finally staking infrastructure there's a lot of players staking is definitely a bit more centralized right now but it could really benefit from the provider's offering staking thinking about how to decentralize their infrastructure for resilience and so what can we do like how can we change this how can we save web 3 protocols from the fate of web 2 companies I would really implore all of the protocols listening to this to incentivize infrastructure providers to support your network so if you're a brand new protocol don't leave your tokenomics or crypto economic incentives uh to the last minute and don't think that your infrastructure providers don't matter there can be an outage there can be censorship and then you're not running into centralized Network you could lose customer funds there is just decentralization really matters there's no such thing as true ownership and financial sovereignty on non-decentralized networks right when Solana goes down people can't access their tokens that is not the financial sovereignty that we're trying to build and get those tokenomics and crypto economic incentives correct talk to infrastructure providers understand those who are doing potentially ethereum proof of work mining who have infrastructure and why they were doing it why would infrastructure providers from all over the world going from hobbyists to now professionalized organizations run decentralized Internet infrastructure and figure out what protocols who have tried to do that again might have gotten wrong and potentially what you can do right I didn't finish this bullet because I did this presentation this morning work with multiple providers for your infrastructure if you're on AWS now and you're sort of panicking you can talk to me and talk to other infrastructure providers get a team that can be analyzing different options that can liaison and understand what is the art of the possible think about leveraging already deployed infrastructure maybe you don't know this but when ethereum moved to proof of stake a bunch of gpus were just left out there in Hardware so if you could figure out a way to incentivize all that infrastructure you could start really bootstrapping your network on that supply side collaborate with partners for optimal design so quick use case web3 cloud is a storage provider in the filecoin space we actually brought Advanced Micro Devices in to help the filecoin team optimize their infrastructure they had way over architected the infrastructure they wanted store storage providers to buy and we brought AMD to the table to help them optimize on cost and performance and so then the whole network and ecosystem players and that space benefit I like to say decentralized early and often we're here for decentralization at least I hope we're here for decentralization and we saw the challenges of this last year of giving in to the word web 2.5 and just bringing people in through memes and not really uh reminding them of why we're here which is to build Global resilient performance infrastructure that connects all of us around the world and that cannot be shut down by U.
S Virginia East going down and then of course you can talk to me my name is Maggie love you can find me on Twitter at Maggie love underscore I work in the infrastructure space um pretty big nerd about it and pretty passionate about the art of the possible if we go back to building these decentralized censorship-resistant resilient Networks especially if you're working in storage ZK or RPC I want to talk to you and thank you so much [Music]
Automatic transcript — names and jargon may be misspelled.