Decentralized Web3 Infrastructure Karen Scarbrough mod; Nick Fett Tellor; Evgeny Ponomarev Fluence;

through another full day of east denver and we are at our last panel so i'm karen scarborough and this panel is on decentralized web 3 infrastructure so i'm going to give the panelists an opportunity to introduce themselves and talk a bit about what of each of their individual projects are working on so if you'd like to kick it off nick sure my name is nick fett i'm the cto of teller teller is a decentralized oracle we bring off chain data and we we put it on chain so if you need a price feed you need information about another chain or any real world data we can put it on for you hey hello everyone i'm evgeny from fluence we call fluence peer-to-peer application platform but basically this is the off chain compute layer that allows to create any kind of applications or off-chain protocols that complementary to on-chain and to other decentralized protocols like storage identity and things like that yeah my name is david vorak i'm at skynet we're a decentralized file storage and web application hosting platform so we focus on developers and we enable them to build applications like a decentralized dropbox or a decentralized like facebook or youtube and one of the big things that we've been doing recently is decentralized d5 front ends so users have control over the code that they're running and they can downgrade upgrade use a fork they don't have to use the dev team's front end great the whole week we've been talking about decentralization in the forms of dows and core blockchain applications uh i want to hear a little bit about from each of you why decentralization is important in the infrastructure that surrounds that sure i can go first so um i think that decentralization is important because it is more powerful than centralization when you think about or when when i think about the peak of the web i think back to around 2010 which is when facebook and twitter and linkedin had uh these like robust or just just really rich apis that exposed all of their data and so entrepreneurs and developers could come in and make applications that had access to the full facebook social graph and all the posts and all the groups and so it was just really easy to innovate if you had a crazy idea you could have a new application running by the end of the day with a billion potential users and data from a billion users and of course this was perceived as a competitive threat facebook etc realized that these applications were actually outgrowing them and potentially in a position to you know replace facebook and so these apis got turned off um and so i think if we can go back to [Music] having these open apis we use decentralization to ensure that the apis cannot be turned off and so we can go back to 2011 and kind of lock that in and make it so that you know things remain open and i think that's important because you just get more innovation better user experience you can't you know user lock-in doesn't happen anymore users have the choice to switch away from an abusive platform that is say pushing really vile privacy policy updates yeah i agree with this point this is great point about like um composability and um apis cutting access from whatever whoever built on top of apis that's the biggest like one of the biggest issues of web 2 and this is what web3 kind of um tries to solve this like in in general decentralization is great like for privacy you know censorship resistance user control data but actually like uh when you think of uh like were the real value for developers where the real value is like actual composibility actual like being safe when you build on top of something that other people created um and you're sure that this service or data will be there tomorrow so you will not lose the access you will not lose your product and your users because like you're investing your time and money and resources into building new stuff so you want to be sure that the infrastructure the foundation that you chose for your application will still be there tomorrow and decentralization allows you to achieve that yeah just kind of those are both good answers i think decentralization is sort of kind of the whole point of it all um you know that's that's why we're all here if you look at just why didn't you build this in a web 2 manner it's because well you wanted some sort of censorship resistant piece it's obviously most of the time whenever you move towards the web 3 version you have to use a blockchain you have to pay gas things get expensive it's probably less efficient might be slower than just using a single database so you have to think about like okay well why do we want to do that and then if you have a centralized infrastructure piece or you know you have admin keys running any sort of piece of your apis or your oracles or something like that it just it it shows that the whole thing probably didn't need to happen right and and on that um like you mentioned people already using elements of centralization outside to make the experience maybe easier to build or faster or whatnot what is your biggest challenges that you see in keeping 3 infrastructure decentralized yeah i think it's it's just harder um you know usually like people will say like oh i want to build this you know whether it's a decentralized facebook or they want to build this d5 protocol and they're like oh we need for us it's an oracle so i need off chain data well it's super easy if you just do it yourself and are centralized um it you can get it up and running tomorrow but then you if you want to use a decentralized oracle you want to have some sort of way to vote on it some sort of fallback mechanism uh it adds a lot of complexity it it makes it slower um maybe not quite as you know easy to use and um those are still some of the challenges that we're doing and just making sure that we educate people on okay this is why it's important this is why we need to do it and try not to take shortcuts well i guess the biggest challenge is you know most of technologies we work with are not mature enough so nothing works really and uh um obviously there are some pieces of infrastructure already built like like we have pretty solid storage foundations like from from skynet or like protocol output protocol labs is doing or like rv uh we have like you know teams building oracles we have like you know some ceramic which is like a great piece also like a database uh part uh but some some pieces are just missing like for example what we see like every defy or dao project that have some on chain code and want to create the front end that would have like a proper you know user experience with you know when you can click the button and immediately see the response or like you can understand what is happening there or you want to see some aggregated numbers coming from blockchain um this forces teams to to use centralized pieces of the solution so they basically beside smart contract and the front-end people doing the centralized back-ends sitting in the amazon so in general their application depends on this central piece uh and this kind of like and there is no way to do it better because there is no good solution we're kind of working on the solution right now at fluence but uh you know i would say like from the challengers perspective i see this is this is the biggest like one of the biggest challenges in in the industry how we transition from totally centralized to totally decentralized solutions yeah i think i think i would go in kind of the same vein uh at maturity i think web 3 is actually easier than web 2. like for example if you want to deploy a website you don't need a server like you don't you don't need to make an account on aws you don't have to pay anyone and so like development is faster deploying is faster growing users i think if we get the equation right you don't you don't need to pay the infrastructure bills i think web3 kind of perpetuates itself and and takes care of the funding itself and so you don't have to fear as a developer like getting to you know 10 million users tomorrow watching a video because uh that's not going to cost you any money whereas if you're using aws that could be you know a ten thousand dollar bill that shows up if you end up being too successful too fast um but the challenge is that web 2 we've been building libraries we've been building infrastructure we've been building all the tooling around it when you think about aws in 2010 it was like really crappy and difficult there's been 10 years of polish and like billions of dollars of engineering going into making the ux better making the tooling stronger solving every conceivable problem in web 3 we just don't have that maturity yet and so things are more difficult not because web3 is worse but because it's different and all of the solutions that we made for web 2 don't apply and so we have to we have to rebuild the solutions and resolve all the same problems but when when we get to the end the result will be fundamentally more powerful all right good points uh so david you mentioned something along the lines of getting the equation right and i think uh with the recent events in d5 we can see that some of those applications did get some very important equations uh quote unquote right or at least scaling and very notable so in the decentralized infrastructure here what kind of incentivization schemes or mechanisms are you working on to promote that same kind of uh growth yeah so we have been focusing more on like social media and so i like to use the example of reddit because it's this like crazy parasite organization where all of the value from reddit comes from users writing interesting comments making interesting memes like finding interesting links and posting them everything that reddit does is users and then users get rewarded in like this worthless karma and reddit takes all the ad revenue and all the value generation and they sort of value capture and all of that goes to reddit and so we're working on a tokenization framework where you can attach a token to everything every post every comment even like every upvote could have a token attached to it and then you can make social media that is rewarding the value creators rather than being fundamentally parasitic and reddit reddit's my favorite example because they literally pay nobody anything but i think all the social media giants are heavily parasitic and that the power dynamic and that the revenue dynamic is way unfairly in favor of say like the facebook or the youtube uh or like spotify is another one where it's like the the artists just don't get what they deserve and so we we hope that a token framework can address this and have the money flow outward instead of upward um actually we are doing something similar for focused on the code and on on the developers so the idea is like if you have the decentralized network for running a code like a decentralized analogy to the cloud um it's actually pretty transparent which applications consist of which modules or like components and it's also transparent who is the author of the components and if you have the revenue flow coming for hosting of this application and components then it's actually pretty easy to see like what which component was popular which component was used across many applications and uh it's really easy to drive the part of revenue uh of this decentralized cloud to author of the component so the idea is we are sort of can tokenize the code and we can drive the um revenue for open source creators which is not possible in the traditional cloud like in amazon uh they do like they earn billions of dollars on these popular database instances like mysql or like mariodb things like that and they share nothing with the authors like really nothing and the like there is there are some established open source models already but if you're the open source developer in web to space and amazon starts providing your uh your software as a service at their platform you can compete by creating your own cloud but it's it's really hard like you you are you have no chances their ecosystem is so higher and like if developers go there they see like they have the whole set of services they can use your service together with others it's like easily composable versus using like very separate cloud that uses open source um developer running so it really creates a you know hurdles hurdles in terms of earning money as as an open source developer and we just see that with this code tokenization we can really enable and drive the new revenue model for for open source and like really accelerate the software uh innovation which is pretty cool yeah i just wanted to add on that real fast when you when you look at the biggest companies in web 2 and just like in general uh none of them are open source uh the the the models that have been super successful in the web 2 world have all been centralized closed source you know company control but when you switch over to the token world it's the exact opposite uh companies that keep their source closed tend to struggle um and in the tokenized world the most successful companies are the ones that are the most open and like user defending and so that that's something that's really exciting to me because i think cryptocurrency has the potential to completely shift the dynamic there and you know make it so that doing doing the most beneficial thing for humanity also gives you the most upside as a founder what about coinbase so you know step by step yeah i mean to quickly kind of answer the the incentives question i think we just do kind of what every crypto company does and we print money um it's i mean it's sort of like the big revolution in crypto is that we all realize that we could mint our own token and use inflationary rewards to to make people do things um you know bitcoin started with incentivizing incentivizing security and now you're seeing people just really experiment with okay well we can put these token rewards towards security or towards paying developers or towards getting people to actually use the thing with yield farming so yeah i think just different models around that and it's it's definitely not perfect quite yet but we're moving there and as all of this develops there's a lot of further decentralization or fractionalization and what's happening for for developers just building decentralized applications now you've got different l2s competing layer ones tons of different ways developers are approaching this space now so probably looks a little bit different for each of your solutions but how are you thinking about providing decentralized decentralized infrastructure around all these new different ways developers can build yeah i mean i think david and both these guys were talking about just kind of composability just the ability you know now you're you're really stuck like in two years ago like everyone was designing their applications where you would only be on ethereum for instance but now everyone knows like you have to build it so you can go on every different chain out there or different roll-ups just because users are on more trains now so um we're just trying to build it so you you can sort of in a modular fashion go over there and it doesn't break any of your governance doesn't break the inflationary aspects that you need so it's definitely tougher but yeah i would say just making it easier to plug and play related to l2s and like many chains i think like you know i expect the most innovation to happen actually on the wallet side because like wallets should abstract out all of this complexity from users users should not know which chain they're using or like when they they should just click a button you know create this uh depth position or like a take a credit or like a you know take a loan or like something like that and under the hood it should be like moving tokens between chain between different l2s from evm to other virtual machine users shouldn't care really about that and for like from our perspective our network is like off chain network but we're going to have the on chain economics layer like with all the payments and transactions we're probably gonna go with evm and letter two um but you know um we will see how it goes like we will see where the the market goes like currently it seems like evm gonna be winner for a longer for like a big period of time uh so everyone will support it and it sort of will be the uh you know universal virtual machine for every chain with different security models but who knows yeah i think um you know first for skynet we've been doing we've been working as hard as possible to make sure that skynet can just slip into any stack it's a data storage layer it's like a application front end um and so we've been looking at ways to and we're hopefully going to be releasing something soon that you could slip into any wallet so you could make metamask a fully skynet compatible wallet and so anyone with metamask can now use the full skynet ecosystem without needing another seed um the same applies to tally and to the cosmos wallet and so we're trying to make it as easy as possible for developers who are you know in multiple ecosystems to just be able to plug skynet in as a module of their application and not have to worry about compatibility as much and then same same thing as if jenny was saying on the wallet side i think you know really all of this complicated machinery of moving between chains and like bridging and cetera and really in the end state users will never see any of that just like uh you know when you log on to the internet you don't realize that you're going from tcp to dns to you know and then and then you're wrapping an encryption and then you're routing through a bgp later and like all this stuff is just happening you're not doing any configuration and i think crypto is kind of the we're not to the point yet where it's just as easy as typing an address in um but we will get there and i think crossing through the whole you know crypto sphere will be very low friction once once the wallets are more mature and and on that topic of the developer experience what are your projects doing to enable that specifically is it portability language specific what are your approaches to that yeah so we initially about two years ago were very language agnostic we tried to make our sdk available in every language and then we found that there was really only one language that mattered and that language was javascript which is not my favorite language yeah but it is it is where all the traction was and we realized that by putting piling all of our efforts into javascript we could really see like the most returns and so maybe in the future when we have a lot more dev power we can go back to supporting a bunch of languages but i think for us the strategy has been you know finding finding the focal point where all of the energy and all the development is and just really really trying to make that experience as simple and slick as possible um yeah developer experience is a huge pain actually mainly because um like we have the new language basically we have the aquap which is like we're a new unique programming language for creating peer-to-peer systems and protocols um and you know syntax is not hard but you know understanding the concepts and and writing the code it's something very new to to people the new you know uh the new uh domain area uh it's like when no one knew solidity and everyone started first time uh got into solidity so it was hard um and for sure we are like we're trying to like we also have like javascript sdks for like complimentary to aqua but this is um i think the way we think of this is that if you're as a developer understand the value of the technology that you are going to use like if you're understanding that you want to create this new internet magic money uh then you have enough motivation to learn solidity the same thing if you have the if you understand why do we want to use aqua that you really want to create a decentralized backend the decentralized api like a setup the off-chain protocol you will learn it like you will you will overcome through this uh you know complexities of learning new uh language so um you know tooling yes we we are working like constantly doing the new tooling that makes it easier but like the biggest thing is you need to learn like you need to um commit to learning the new the new stuff yeah i think there's a big thing with you you want to make your documentation easy you want to make everybody able to use your decentralized infrastructure but to be honest there's not like thousands and thousands of users who are clamoring at us to use our our protocols like if you really want to like people who reach out to me like i'll like literally like make the pull request myself and like like help you in input your code you know that's what we found like a lot of users like making sure that we like answer our discord messages with them and just help them out and it's still a super small space it's not um you know people like being able to talk to you and and put that way so it's we're still kind of growing but yeah yeah there is a lot of approachability in this ecosystem that i like um and so for our last question here um borrowing a question i saw at another panel that i liked what's your biggest prediction for your category in the future for decentralized infrastructure so what do you see happening in the coming years and uh going forward um kind of for my category so i guess i'm like a little bit different than these two um oracles um yeah i i guess my biggest prediction is i think we'll hold fingers crossed get away from a lot of these bridges that are really really really bad um you know the oracles are a big piece in making decentralized bridges so i think that's going to be really a key piece in making a lot of this interoperability that actually happens um i guess on on our side is we see a lot of complexity in building uh the new networks the new unique systems like you know the new oracle with very specific economic design or like the new bridge with very specific security model or um you know things even like uh off chain multisig uh wallets which have very specific way to create signatures or distribute the the pieces of of um uh threshold signatures so what i really want to do what i see is the um easiness of the like of the creation of such systems so like i want to see like the design space where uh it's it's easy to experiment by deploying the new oracle with the new algorithm or like the new uh replication model for the data or the new uh you know uh i don't know like the new the new off chain governance voting mechanism and and combine them together and play with them and iterate on them so this is what we are trying to enable basically so i'm looking forward to seeing it and this is what i you know this is what would be super cool i think i think we're going to see the world hyper tokenize i think every every human being is going to have one or more maybe several uh of their own tokens that they've kind of like launched and curated like all of your friends will have tokens i mean like every restaurant will have tokens and i just i think the future is billions of different tokens and people will have portfolios of like tens of thousands of tokens each it's gonna like just replace the stock market i think i think web web 3 is going to deliver that it's a very bold prediction i like it um so i i saw the other panels do questions from the audience too so i don't i don't know if we have a minute for that but so we have a minute so if there's one question out