A Playbook for Middleware Protocols | Michael O'Rourke - Grove Redefines Our Very Existence Inc

hello can everyone hear me okay awesome hi everyone uh my name is Michael oror and uh I am going to be talking about A playbook for uh what was previously what I would previously call uh middleware protocols but now I think is officially uh deep in uh so to move forward with this talk uh what am I talking about uh I'm going to be mostly focusing on the economic perspective on these kind of middleware protocols and I'm going to be leaning heavily on the Lessons Learned in founding and and launching pocket network uh some of the key takeaways in kind of the Arc of the talk is really going to be talking about what's it mean to launch your protocol what's it mean to grow your protocol and then what's it mean to get to sustainability uh a little bit about me um co-founder of pocket Network I'm the CEO of Grove Incorporated formerly known as pocket Network Inc I am a Florida man uh I studied international studies in college and I eventually learned a code before uh founding pocket and was an iOS engineer for a few years uh real quick to to talk about what is pocket Network solving um so the key thing here is is the RPC trilemma uh hat tip to olansky for coining this uh but it's really about uh every every middleware deepin protocol is is effectively solving the same problem here uh we all want to uh uh kind of find the middle between reliability performance and cost Effectiveness and as an RPC provider right we want to make sure that we meet the quality of service slas and everything while having the cheapest and fastest service possible H and and why listen to us uh uh well uh we've got pretty serious experience in terms of theory and practice uh We've overcome some pretty serious uh Technical toonomic and uh social hurdles online and we are the creators of the first uh decentralized RPC network uh we've been online on mainnet for 3 years uh we just had a uh a protocol upgrade last week that we coordinated over a thousand validators to to upgrade uh the protocol uh we peaked at about two billion daily relays and we've served uh over 700 billion relays so while this isn't Web Two scale I think it's a meaningful enough scale to to kind of hopefully impart some of our lessons so what's it mean to to launch your protocol um as this talk goes on you'll see this kind of solar Punk uh imagery uh uh and and that's just kind of to highlight the optimistic nature of what I think we're building I really think that we have the opportunity to uh create new institutions and uh through these through these protocols uh so so that's kind of uh why why I like to have this kind of imagery behind it um so launching a blockchain is is really hard um this isn't a not to KN web2 SAS Services uh but uh these have been uh really tried and true for the last 20 years so not only do you have to deal with the front end to backend uh your infrastructure devops team Business Development uh you also have to build a protocol you also have to deal with the community uh in most cases you have a dow uh you have to design some coke and economics and honestly if you're in infrastructure uh marketing is really hard in this space so um what I'm hoping to do is uh uh since we've seen so many Innovations on the tooling to actually be able to launch these protocols with Comet bft with uh rollups with arbit arbitrum optimism and these different protocols what I want to do is really kind of frame how you can think about launching these things in uh in a practical way and give you at least kind of a path forward uh but first I kind of want to frame how I think about blockchains um if you draw a triangle between governments religions and corporations uh blockchains sit right in the middle uh these are really kind of the first and and only three major human coordination mechanisms that we've seen in the last you know 6,000 or so years and that's why I think this is such a really exciting time because uh blockchains are really the first uh internet native human coordination mechanism so when we talk about tokens and and these sorts of things the important thing here is that uh uh uh we're really coordinating people uh across the internet uh without knowing who they are where they're from and I think that's just incredibly powerful uh so what do you have to think about when you're launching your protocol um uh uh you've got the supply side uh that you have to they they have to think about so you have to bootstrap the supply side um I think it's really important to just experiment and iterate uh I'm a big fan of don't let the perfect be the enemy of good uh that said you obviously want your protocol to be incredbly secure and you don't want to deal with uh any major hacks or anything like that uh we think about just one major kpi uh obviously the simplest is is the simpler the better uh in our case it is requests going on the network or relays as we call them um when we think about the demand uh uh the founding team should generally be the people driving the demand of the product uh at least initially uh because generally in our experience uh garnering the supply side is a little bit is a little bit easier uh also when you think about Capital markets uh you know uh things we have to you have to think about things like listing your token and and and what are the economics of the token and and whether you're on evm or are you on IBC or are you on an island and uh particularly uh when it comes to the economics uh uh you don't want to over complicate your design of your token um as it heads up I will be talking a little bit about token price uh but this is purely in the context of having an actual sustainable protocol uh that you want to actually exist for for over 100 years uh but to talk a little bit about economics um I really think that that we over complicate our designs a lot um this is you know uh this is not very different from video game design um if you're if you're if you've ever played any MMO RPGs ultimate online was the largest in the 90s uh and in fact they wrote a paper uh about the economics of their MMO RPG and this is a picture on the right of their actual economic system um and at the end of the day all you really need to think about is what's creating the token and what are the syncs for the token right so In Pockets use case uh to be a node you have to buy and stake pocked to have the right to do work that's great because in a permissionless system now the protocol can actually slash you or uh punish you if you're not applying uh adhering to the rules of the protocol itself um in terms of our our faucet for the for the for the protocol it's every time a request is created uh uh Pock is also created once that is verified by the protocol and you can think about how these things can evolve for example uh you can think about and various lending lend borrow desks and these sorts of things as different kinds of syncs for your token um and at the end of the day what you really want is an equilibrium do you uh do you have the amount of pocked in our case pocked being emitted equivalent to the amount being burned or uh being sunk into the system uh uh so you have kind of permanent burn and then you have things like staking and and cryptoeconomic Designs but fundamentally uh I actually don't think it's that complicated as long as you uh uh really think about kind of the outcome of what you want your token to coordinate right and I think that's really really important to think about uh I'm a big fan of how placeholder uh Capital FL uh frames uh kind of decentralized networks uh when you're launching your protocol you really kind of have four vectors of decentralization uh Capital markets governance supply and demand uh I'm not talking too much about Pockets governance in this talk but uh we went fullon full decentralization on our supply and governance side of the protoc uh whereas on the demand and capital markets for us was a big fat zero uh that's because we had to learn what it looks like to have worldclass quality of service on a network of people that are running nodes uh uh that have all very different setups right so in our supply side we had permissionless Supply uh our governance model is one person one vote as long as you participate in the protocol you can actually claim your vote in a in a bit of a game but up until relatively recently uh Grove has been the only Gateway on top of the Network and uh it's taken some time to get our Capital markets mature so these are things like you know exchanges defi um lend borrow like there's a whole ecosystem of of capital markets that I think is actually quite critical for tokens despite honestly being a a product person and really wanting to focus uh mostly on the product uh so we're evolving a little bit you've launched it uh uh now you're starting to think about uh growing uh your protocol uh so so so let's look a little bit about uh let's look at what uh let's see what that looks like um so you're starting to get traction on your main kpi in our case that's uh relays um just to give a little story we uh uh the first year of pocket uh was really just fixing things uh we were maybe doing 2 million requests a day for the first few months and we actually had uh one of our first users uh uh 5x that traffic effectively overnight and that was actually our first chain halt uh we call that the uh uh fondly the the dark blocks phase uh of of pocket uh but what I mean is here is is really focus on your main kpi it's inevitable that things will break um you know we are not near web 2 scale where you're seeing you know trillions and trillions of requests every day uh and as uh apps uh start to come on the market it's going to continue to push the infrastructure and then we have to learn to figure out how to really uh improve it to be able to maintain that scale um you know as product people it's really important to to to see our people actually paying for your product uh is your token actually serving a real use case and within that context it's really important to see to think about kind of the right incentives um I think a lot about how people launch their their their tokens and and their networks and really if you want these things to last um you really can't frontload incentives um you really want people to continually earn and see opportunity over the course of three four 5 10 20 years uh and we kind of don't see much of that these days um uh the other things are how is your supply and demand scaling um it's always going to be a push and pull uh but as I'll talk about in a moment uh you'll see how uh an extreme version of one side or the other can impact your protocol um and of course you know down the line you want to think about decentralizing supply and demand uh when you're bootstrapping your ecosystem um I think generally if uh you're really focused on building a a a technical uh community and people who really care about your vision uh friction is good um so just to give you an example uh we launched our protocol in July of 2020 uh we weren't listed until uh uh I believe it was January 2023 um uh on proper exchanges or I'm sorry 20 yeah 20 uh 22 forgive me um and for the first year or so of pocket being live you can only buy Pock through a uh telegram OTC um and this is very different from buying something on Unis swap uh there was a lot of trust involved um luckily we were introduced to some good people and and some some of our own investors just decided to to help out there uh but was really interesting about this and that has actually impacted our community to this day is that that friction to be able to trust as telegram OTC through non-traditional channels uh really made it such that when people came across pocket they had to really care about the product and the project to really kind of take that leap and actually participate in the network and uh that's actually formed the foundation of our community uh until today and I think is really uh something that that that has been a really strong impact for us since we since we've launched uh so let's talk about unsustainable economics um uh so pocket has a dynamic inflation uh it's purely based on the traffic uh going through the network I've got a picture up here at the top uh of that inflation and at the bottom I've got a picture of the price of Pock um so from 21 to 22 We Grew From about 10 million requests a day to over a billion requests a day uh and this was all providing free infrastructure uh this resulted in a uh incredible uh over uh uh inflation of the network I think we're like over a th% a year at one point and this created a really kind of perverted dynamic because we actually had a very Supply Le growth in the adoption of the protocol and the token uh so you can see here kind of the pocket price it Maps TimeWise even though the pictures aren't exactly mapped uh you'll see our kind of token price going up because people were just kind of printing money uh actually it was kind of ridiculous uh uh and uh everyone was also running on the cloud uh which is also uh uh not very good because kind of that doesn't really align with kind of this RPC trilemma uh that I mentioned before uh so you can see here uh uh around uh January of 22 uh January to April 22 you'll start to see the emissions start to go down and that's because the pocket Dow decided okay this is crazy uh we need to cut down inflation significantly uh and that resulted in a very uh reflective a reflexive cycle for for our Network and this ended up creating a ton of pain uh for for a couple reasons but also there's some Silver Linings too uh so markets uh will adjust uh one of the things that I'm most fascinated by is how the pocket infrastructure Market has actually really been Dynamic um so when people were uh make printing money no one cared they were on the cloud but over the next 12 to 18 months everyone moved to bare metal and everyone people started buying their own servers and actually getting the most efficient setups they possibly could uh which I think is really really fascinating but when we talk about an imbalance of supply and demand uh we had a massive over-provisioning of infrastructure uh not matching the uh amount of traffic that was being sent to the network much less paid traffic because at the time this was all free and we were just trying to bootstrap it to prove that this could work at scale um in hindsight I think one of the key learning lessons that I had uh was we should probably be uh much more aggressive about adjustments I think we should get our protocols used to shocks to the system uh early on uh because I think uh uh that really helps our uh uh protocols be kind of antifragile in this respect uh we ended up kind of slowly lowering inflation which caused a slow bleed um and in hindsight if we had just kind of cut inflation from you know 100% to 5% I think uh we would have been much more healthy uh at the end of the day but that's in hindsight um so just a couple key learnings from this story um Supply is easier than demand uh I think you should get your community used to shocks uh to the system early on I I mean I think we see this just in crypto Cycles uh uh over the last however many years you know it's every three or four years we see highs and lows and everything in between um I think being internet native we're better equipped to actually handle these sorts of things um Supply Le growth is obviously not sustainable you want a real balance on Demand versus Supply the market will become more efficient uh defy rails would have helped us significantly uh we had lots of node Runners having to sell their pocked because they were running full archival nodes on AWS and uh uh things like being able to l Po and lend usdc and and to pay your server bills and be able to pay that back and like your Pock emissions would have helped a lot for example um and also Community committed Community will contribute and push a protocol forward uh the last 18 months were a lot of pain was a lot of pain for the network and uh we've had some people uh most of the core Community has actually stuck around we've had some consolidation uh uh within the supply side network but at the end of the day um people uh who who really care about it uh will will stick around which I think has been one of the coolest things to see over the last 18 months cool so I only have three minutes left so I'm going to talk about getting sustainability so this is what the future looks like you know how do we get to the sustainable uh uh uh kind of kind of uh future that we want um so I have two view I have I have a view on how uh these things should grow on the demand side generally we see them being either Foundation Le or Corporation for-profit Le in our case Grove is a for-profit Le in terms of how we are thinking about decentralizing demand but in uh for example our weave I believe is a foundation Le and you can kind of think about how these uh different protocols do this right um uh I believe that in our case uh uh by building out the tooling and building out the business on top of the protocol uh by fundamentally forcing ourselves to have to turn a profit uh we can build the best tooling for uh uh the right ecosystems in our in uh for for the markets that we're trying to capture right whereas I think uh uh it's a different scenario if you have a found Foundation uh kind of uh uh giving grants to different participants in in in your ecosystem uh and one thing that we've really thought about is is how do we create a moat in an open source Network when uh in about six months or so we're going to be launching the new version of protocol of the protocol which uh allows us to have permissionless gateways uh which is really exciting for us like I said before we've really been the only Gateway on top of the network until relatively recently and the last year we really spent uh uh thinking about uh what does a Moote as a for-profit company look like uh in a decentralized network um and also when I think about decentralization I think at the end of the day it depends on how SI how big your Market is and that'll allow for how many uh kind of Niche uh uh Solutions can actually be built and serving different customers I think being in the infrastructure space uh there's effectively an infinite number of permutations that web through developers need uh which means that I think there's a lot of opportunity for people to build different types of gateways within our ecosystem uh and also I think it's just really important to build really easily uh runnable software uh to just talk a bit a little bit about Grove's kind of business model within this context um we've spent the last 18 months uh just building an incredible stack uh that allows for people to access the 50 plus blockchains that we have at you know sub 150 millisecond latency um and uh uh what we want to do is is really kind of go towards this open core model um actually if you look up uh open core Ventures I think they've got some really good blog posts on like what this means um but generally uh what we plan on doing is opening up our own core apis for our core product the portal and we want to be able to enable as many other people to build these gateways as possible uh we did this first rather than open sourcing kind of the core pocket logic uh because pragmatically we wanted to be able to get to Market sooner and with the launch of the new version of pocket Shannon uh uh we want to be able to enable as many different types of services as possible uh but over time uh we're going to end up open sourcing our core pocket logic which means anyone will be able to uh run their own Sovereign Gateway and for our business we see our our our kind of opportunity as being able to sell into these Sovereign gateways through these additional apis that we've gotten really good at serving so that's things like real-time data notifications and this sort of thing so uh you can kind of think of uh kind of word WordPress plugins or or some mix of like Shopify and this sort of thing uh so at the end of the day uh every middle word protocol literally has the same same challenges you want to launch you want to fix a lot of uh you want to incentivize Supply and uh figure out what your kpi is in our case quality of service uh and and relays you want to keep fixing it's going to keep breaking uh you want to mature your Capital markets uh and you kind of want to rein in Supply Le growth like I said in hindsight um and as your Market grows you want to keep fixing uh and then you want to build out the open source infrastructure to enable uh kind of this demand Side Market to uh to flourish and and what I mean by that is is I think at the end of the day uh deepin and middleware projects are really enabling uh uh opportunity for entrepreneurs and that's really dictated by the size of your Market um at the end of the day and and and I think that's really how we should think about this I think we are in a special position as kind of broadly middleware and deepin because we are actually doing the really hard work of actually uh uh uh uh touching the real world uh in some way whether that be through AI uh infrastructure storage and these sorts of things I think we have a big opportunity to actually have real web two use cases and by enabling as much opportunity for as many people as possible uh you'll actually have uh an incredibly strong community that can withstand uh uh incredible shocks of the system and token up or down or whatever it might be in the future so uh that's uh um I got to end my presentation now so thank you for your time and uh yeah thanks [Applause] everyone