Space Warp Summit 🛸 The FVM Imaginarium Part 1: Request for Startups - JV , Matt Hamilton

foreign [Applause] [Music] and up next uh we're gonna have two different back to back talks so part one part two uh called The fvm Imaginarium um so it's gonna be focused on everything uh kind of what's upcoming and what can be built and what can we do and what is already being built so uh up first we're gonna have JV and Matt from the protocol Labs team uh who are going to join us and Shout about uh the different requests for startups and different ideas that uh that are kind of out there for things that can be built with dmvm so I welcome both JV and maps to the stage feel free to unmute and share your uh your videos there we go hey Matt ajb and uh yeah I'll get out of the way here and let you guys take over awesome thanks for that um let me share my screen awesome well thanks for joining us everybody uh I know there's been a lot of amazing talks and there's a couple more to come um but yeah super excited to have Matt with me here today um I don't know if folks have seen uh there's a website that we'll talk about in a second um but as the fvm is launching there's an amazing amount of opportunities for businesses to be built we the protocol lab side and the Falcon Foundation side have brainstormed a bunch of ideas this is no means by no means exhaustive but would love to share at least to inspires and builders on what are different opportunities that exist and of course as you're going through this process uh if you reach out to the teams I'm sure there's folks that have individual thoughts that might be able to help advice um so maybe to kick things off um I wanted to start with just a macro frame I think one thing that's really interesting is just when we zoom out and think about the file coin economy what we have right now if you think of all coin pre-fem it's sort of this like very basic Loop of clients and storage providers that can find each other they can make deals the follow coin blockchain is really settling that and the thing that the fem really gives us is now the ability to be much more expressive in what we can do with that economy so this is just a very basic map showing how much more complicated uh this economic Loop can actually be when we start doing more and more things on chain but it gives us the ability to build Primitives and businesses that may not look like things that you would see elsewhere so rather than just having a client and storage provider make a deal if you want to do something that says maybe builds a Perpetual storage protocol which we'll talk about some ideas there that actually implies a bunch of other services and businesses that might exist on the fem and so when we start looking more holistically at like what this economy could look like it really becomes interesting just in terms of the broad split swath of opportunities um so yeah uh at the end of this talk we will also share this link again there is a much longer list of ideas that people have brainstormed together really just pulling on the threads of for the different user groups and communities inside of the falcorn economy how might one build businesses around the services for those folks um but we'll talk about a few of them today um so yeah with my co-mc here Matt maybe Matt do you want to kick us off I know you've thought a lot about under collateralized lending but I think this is maybe one of the most important things for the storage providers in our economy sure yeah sure thing hey folks uh so my name is Matt Hamilton I'm a developer Advocate at protocol Labs working on the on the FM team and uh yeah as JB said we've got a number of different use cases here so the first one I'm going to mention is this idea of under collateralized lending now if you've come from say the web 3 defy space you're probably familiar with the pattern of over collateralized lending which is where for example you put up some uh some some collateral in one token and you can borrow against it in some other funds so maybe you put up some Bitcoin or eth or xrp and you can borrow something like usdc against that now typically you have to put up above the value you're lending so sometimes it's 150 or 200 or whatever it might be in collateral against that well one of the nice things about the falcoin network is there are real world businesses behind this with data centers and storage and everything and one thing you can do is under collateralized lending and that is because when a storage deal is made on the network the storage provider has a pretty reasonable uh income stream or a reasonably guaranteed income stream over the lifetime of that storage deal so if I as a storage provider agree to store some data for you then the fee for that storage gets put ineffectively in escrow and then gets drip fed to me over the lifetime of that storage deal now I can actually use that uh that that fact that you know future income as a uh as collateral for Lending so I could actually put that up and say hey well you know I'm guaranteed that this amount coming forth so please let me borrow against that future kind of guaranteed income uh you know now in some other token so that's what kind of under collateralized lending is and that's uh I think a fairly novel thing that is is probably not so possible on many other kind of blockchain economic models because uh within like I said within Falcon you've actually got those income streams coming in uh storage providers have committed for for storage yeah and I think like one interesting I think way of looking at this is Falcon has all of these on-chain businesses all the storage providers your on-chain businesses we'll talk about another set in a little bit about like data Dows as other types of on-chain businesses but when you have these on-chain businesses that have revenue streams that are getting paid on chain we can use sort of like the laws of smart contracts to start writing these more interesting economic Primitives that can unlock Capital efficiency in the economy um I'll actually touch on maybe another sort of like Twist on this which is once you have these under clouded wise lending protocols a pretty common thing that you might want to do is be able to get like early liquidity um especially if folks are familiar with sort of like the flow of capital inside of the Falcon economy if you are a token holder and you happen to give a storage provider your file coin in order to like put it to work that's going to have a duration over the course of a deal but maybe you want to be able to like repurpose that collateral in other ways liquid staking protocols ways of tokenizing these assets can give you early liquidity or can be an opportunity to provide early liquidity meaning that people can use their capital and the locked collateral in other ways so folks have looked at things like rocket pool if folks have looked at things like eigenlayer there's interesting elements that might be worth drawing on as we think about pack when you use the core filecoin like collateral like economic Loop to start funding and even being used to secure other protocols that could be built on top and maybe actually to add on top of that if you're going to build those two things so under collateralized lending as well as the liquid staking then it sort of also implies the use case for dexes and exchanges places where people can actually take those tokenized assets and then trade them into the other protocol or other tokens that they may be looking for um so I think it's an interesting frame if we really look at these economic models some of the business models behind them as well as what is the core infrastructure that maybe just is more the bread and butter of how people think about D5 today but combining it with some of the unique opportunities inside of Falcon uh Matt was their comment there that you're gonna toss in no I was just gonna put it in this you know is one of those things that kind of helps expands there's certain like businesses and I think we've got a slide later on this that kind of add in almost like supporting infrastructure for some of the storage uh systems as well so even things like a decks for example might not seem you know a decks itself doesn't have anything to do with storage but it enables a number of other use cases on top of that uh with regards to storage and the general economic ecosystem around storage um as well so yeah thanks 100 percent uh well then maybe we can zoom out a little bit um so we talked a little bit about Capital flows and some of the use cases and business models um that might make sense uh that help improve the capital efficiency of the Falcon economy I think it's also interesting to sort of like bridge that into some of the functional flows where I think one of the unique opportunities in the Falcon ecosystem is being able to tie especially with the fem uh programmable Capital flows to programmable services do you want to talk a little bit about Perpetual storage and what that might look like sure so with falcoin the model is that a client and a storage provider negotiate a storage deal for a certain amount of time in a certain cost now that might be for say 100 days or 200 days or whatever it might be um there for that storage deal now Perpetual storage is the idea that a storage deal could then be continually renewed thereafter and that might be automated uh by tools enabled by the fvm right so you might be able to incentivize say bounty hunters or whatever to create additional storage deals beyond the length of the initial storage deal uh to continue this forward now how might this be paid you might have um for example people contributing in and we've probably got a we might have a slide separately on on datadel specifically but you might have um people contributing funds in together to fund that uh Perpetual storage or it might be paid out from some kind of endowment um for example or some other mechanism by which it could generate the the funds could generate some economic reward that could then be used uh to fund that and there's some projects we have in the the next talk I believe we will have one of the builders that's building something based upon this idea a company called Lighthouse that are doing something along these lines yeah I think like one of the interesting places where this really intersects is If you're sort of looking at Perpetual storage as saying we have all these Primitives on the filecoin Chain where we can see like what is the state of like specific deals that have landed on the network like what's the health you can then combine that with sort of incentives to say I want to make sure there is always a minimum number of X copies and I can draw down from some static pool of capital whether that's coming from a data dial whether that's coming from other sources in order to like perpetually make that condition true and when you start combining that with some of the financial Primitives you were talking about before that's when it gets really exciting where we can actually use the yield from the filecoin network to help ongoing perpetually store data sets which is pretty exciting um which sort of goes into data does um so Matt had touched on this briefly I think there's a number of ways of thinking about what data dials could be I think maybe the simplest form of a data dial might be something that's very similar to the internet archive where it's just look our mission is to preserve this public open data sets it's really about just making sure that this stays accessible and effectively what this might look like is a governance structure say you have some decision making on like how this is done it could be very light it could also be basically non-existent it could just be a giant pool of capital like a pension fund where you're basically aggregating Capital that is purely funneling into something like a Perpetual storage protocol to make sure that the data that is supposed to be managed is perpetually being funded in perpetuity um you could also think down the line about uh as things like the compute networks come off the ground how do you want to handle derivative data sets how do you want to do things like tokenization so you can have value flows go back so for permissioning Access and then getting some of the retroactive rewards if someone purchased a data set um I think these are things that are interesting to think about whereas today we're talking a lot more about the storage side of the house as more and more features come to the filecoin economy these data structures can become increasingly more complex down the line as well I don't know Matt if there was other things you wanted to touch on on the data side I think you've pretty much covered it but yeah one of the things is regarding things like scientific sets or or you know machine learning for example that you wouldn't want to do the actual computation of machine learning on chain on Falcon you can do that off chain uh there's a project called bacchial that's part of the compute over data project at protocol Labs it's working on that but you could use again the Falcon Virtual Machine to automate some of that and provide some additional governance around that so you might have an example where um one person could create say a dated hour which a raw data set is uploaded there's a bounty put for somebody to run some computation against that data set the data set is downloaded by somebody they do the computation uh the end result is uploaded back to filecoin they present the deal ID back to the smart contract to say here I've done the the computation here's the deal and you might even have then a set of say Witnesses or validators that can then look at that data so they might even be you know a human process that analyze data and say yes that task has been completed they vote on the completion so that's another sort of dowel like thing they can vote on whether or not that task has been suitably completed and if it is then the Dow might then pay out and you know maybe you can imagine a scenario where 60 of the funds go out to the computation provider 30 are split amongst the validators and 10 goes to the Dow for the maintenance or something along those lines so there's a lot of different models that can be built from some of these simple uh Primitives and the demo just now by Zach the last example he showed gives you some example of how you could structure a data down and how you could incentivize the participants to um take care of the storage uh actions required for the data down and I guess like maybe one other thing before we move off of this one that might be interesting to think about is the different funding models to bootstrap some of these organizations it could purely be just donations from Individual folks it could actually also be like soliciting folks like if they're public open science data sets are there governments that might be willing to put money into this other research institutions where you can get universities to collaboratively instead of everyone individually stop trying to solve the problem on their own can we pull together the resources collectively to just de-duplicate work um or it could also be like nft holders like can you mint nfts to fund these public goods which is another interesting Avenue to think about yeah indeed and no just one more bit on that with with the um uh with the funding models you know one of the things I've heard a lot about is actually scientific data funding is really quite complex in some scenarios you might have multiple different funding bodies funding a particular particular scientific effort so you can imagine something like a dated out that could almost like pay out in response to the funding that is put in so different bodies that are investing in that funding um in that project and funding that project could then get rewards you know in proportion to maybe the way they've uh the way they've funded it so um yeah on to uh on-ramp shall I uh should I take this one as well while I'm here yeah so with the falcoin network at the moment um and again you you saw just from Zach's demo earlier he was talking about storing uh piece IDs and these are kind of large chunks generally the falcoin network itself is concerned with storing um sort of large uh sort of almost like wholesale if you can imagine sort of chunks of data 32 gigs of data so how does that help when you've got your uh 50 KB uh jpeg image for your nft for example you want to store that on on filecoin well there's a number of storage helpers that are coming out so things like nft.storage uh web3 storage um Estuary Lighthouse there's a number of them that are that are building out and they're helping sort of on-ramp data into the Falcon Network and again using something like data dowels could uh help that process because you could fund uh the um onboarding of that data collectively via that and and some of the management of that data so I think that would uh that's another use case there 100 and when you think about things like cultural archives some of the requirements are very specific or even like video data if you're optimizing for fast reads and so some of these may be really interesting opportunities both as like products or services to be funded through data Dows to basically on-ramp a bunch of data but also maybe like individually interesting business models of their own depending on the requirements on like retrieval and things like that um cool um yeah programmable storage markets I think is a really fascinating one um so filecoin itself is like a quite flexible protocol um and so the markets actors as they are today are just one implementation I think there's opportunities to think about what are different ways that someone might want to build these markets themselves this gets a little bit more into the weeds of like if people haven't divided into the guts of fog one itself um but I think it becomes interesting of where are there opportunities to build better markets and also maybe think about different ways of even setting up deals how to sort of like think about uh sort of like the incentive models when people set up bounties and things like that just for uh sort of like settling the actual transactions that might be like being struck off chain um and so yeah I think there's opportunities here just to think uh more about what are potential opportunities that yeah might exist um I don't know Matt if you have any on this one uh yeah yeah I think that's a that's a key point and this is an opportunity to hear point out a little bit about how filecoin works and to to sort of clarify that a lot of the process that happens around the file coin the entire kind of storage protocol happens off chain the idea being is it is a very flexible system and that the actual negotiation of the storage deal between the client and the storage provider is effectively left you know flexible and is not like mandated kind of on-chain a particular way but is done off chain because you can imagine if I'm trying to store a vast amount of data then again it might be a case of you know a station wagon full of hard drives going down the the highway is the easiest way for me to get my data to the storage provider to store it so a lot of the the you know the negotiation of the storage deal happens offline at the moment um and and happens separately by the like the filecoin protocol but not actually on the blockchain whereas now what we have is the ability through fvm and through the small contracts to now build some of these automated marketplaces so I can I can actually like bid for storage based upon a certain price or somebody could actually um you know offer offer prices or whatever you could have this kind of um you know a bit like a uh a limit order book type thing perhaps around storage or whatever you might want there so that can that adds a lot of flexibility on top of the network to automate some of those processes or I mean another way of looking at it is like with additional features so like as people do things like on chain credentialing for storage providers perhaps you would say like I would like to set a bounty for storing data but I only would like storage providers that are in specific geographies to be able to like bid on this data or maybe you'd like to say like I would like a Max retrieval price and then have a retrievability Oracle or something where a storage provider is locking up collateral to like guarantee that they will actually serve the requirements you basically can extend and Define broader ways of setting up these storage deals with more Primitives inside um which gets to retrievability orbitals um which I think is really fascinating like service that could be built into the network if folks haven't seen there's a great video that uh Nikola from the cryptonet lab uh he sort of put out a couple months ago but this is the idea that you could have an on-chain entity that sort of acts as like a Better Business Bureau for storage providers so imagine me and Matt are making a deal I'm client mad as a storage provider and I want to guarantee that Matt will never try to like price gouge me on retrieval and at the time when he strikes the deal he might say John I will never charge you more than five cents a gigabyte to retrieve your data and that's totally true but then some exogenous shock happens and mad like actually it's going to cost you 50. well with the retrievability Oracle the way this might work is you could have a dow where when Matt and I make this initial deal we both send in the terms to the steel to the retrievability Oracle and they say okay we will make sure that this is always enforced over the lifetime of the deal and Matt you're going to lock up a bunch of collateral so that we know that if you ever try to buy violate that there's something to be done and so now Matt and I can do our transactions if something goes wrong I can always appeal to the retrievability Oracle they can try to get the data from Matt against the terms that we had set if they can't do it they'll flash Matt if they can do it they just send me the data so this is just like another functional like here's a service that can be built on the network that provides a level of security for clients that could totally be wrapped into a broader set of storage Primitives um okay that was a lot of content um this is only even scratching the surface of like what could be built on the fem I'd encourage everyone if you have your phone scan this QR code go to this URL if you're looking for more ideas feel free to reach out to anyone on the team I think we are super excited about what this unlocks for the fem and excited for you all to uh take it up take the ball and run um so yeah thanks everybody thanks a lot and let's see what you build awesome thank you so much Matt JV uh I guess for everybody who's watching if you're part of the hackathon I mean this is like one of those spots that is absolutely pure Alpha for you uh you're gonna definitely be building um over the next couple of couple weeks and I think Matt and JV just kind of went through the gamut of uh of what's possible and and some different ideas