Swarm Summer School ‘21, Day 2: Elad Verbin - The paradigm shifts of Web3

for now i'd like to um welcome elab verdan if you have any questions during this talk during the both of the talks by the way feel free to ask them either in the q a section of this zoom webinar or in our discord channel of which the link has just been shared in the chat okay and then i would like to give the floor to uh welcome perfect thank you very much robert and thank you for to robert for for organizing uh the summer school and for uh the swamp team for the entire event um you can all see my screen as soon robert can you verify yeah yeah it's good perfect so hi everyone uh in this talk i will tell you um some of my um learnings and and opinions and positions about how one would build the startup for a small power world right so it's um it's i guess it's in the business uh aspect but it's also in the how does one build a project that becomes a big influential project and as a vc uh in my my day job which is something i'll get into in a bit i am a bit biased towards big uh profit generating uh um you know commercial projects so take this with a grain take some of the parts of it will kind of stop because there are many other things that one can build that are just as important but i because of my background will focus on the you know how to make something really big that you know me tens of millions or hundreds of millions of people use and write to how to end them so here's the outline of the talk so and that's why i'm going to keep using the word startup because really i'm talking about startup that employs a bunch of people and builds something big um that's not the only thing to to that one one can do on the swamp um but if you are building yourself for some powered world uh my main takeaway from this talk is what you're going to want is to plan for the swamp project of 2024 not for the swarm of 2021. so you don't look at the a book of swarm and the orange papers and say let me build something that uses this right now and that would be great what you want to do is as i tease a little bit but we will get into this uh very uh in depth um right we'll get into this very in-depth uh what you um want to do is you want to um you want to figure out in advance what swarm will give you and then build for that and if swam doesn't give you that thing you want to account for it in some way uh after telling you uh that part i will um and i will tell you what are we what is this almost 2024 so i'll tell you what swarm uh does for the world in 2024 for the internet in general this is mostly going to be about economics and economics ecosystem and so on and then we're going to talk in the third part about what a swarm-powered world looks like for an engineer so you know your engineer you're coming to build an app what does one do that you don't get currently um and if we have some time we'll talk about the swarm as a database but we will see that and there's plenty of room for q and i hope to finish early uh but you know we'll see good so uh robert please stop me if there's any problems with connectivity or anything else because i'm i'm on vacation so i'm speaking for my world informing airbnb um so good so a few words about me so in my uh in my training i'm a computer scientist i did the phd in tel aviv university in data structures which is uh coming kind of coming back with swarm i'll tell you about this uh later but as a computer scientist doing algorithms data such as a bit of cryptography a bit of game theory that of everything else um and then i moved to industry i worked on a bunch of machine learning and saw victor trump present in berlin in 2016 and that kind of changed my life i was totally hooked on swarm um and wanted to do everything uh about it this is the center picture is me presenting back then in 2016 and i you know i was a swamp fan for a long long time and at some point they did a transition actually to um to investing in startups uh we realized me and several colleagues that europe really lacks a big economy for investing in algorithm making computer science startups so we started that kind of fund that takes me a little bit away from some but i think it's worth it um and uh in my day job what i do is i run a vc fund called low ventures that invest in precedent european computer science startups so if you're building a startup that relies on computer science algorithms cryptography uh swarm decentralization in some way uh based in europe in any kind of early stage please come talk to me my email is everywhere.gmail or maybe using all of my social media or approach me here or with many social media um good so uh so so let me tell you my main question that i want to answer uh is what superpowers get swam give for your project because you could build on anything like why build on small so in general how do you build a startup in general like how do you build a project that is going to be big and the point of it is that either you execute super super fast either you build something in three months and you deploy it and you gain a million you know a million users and everything is big but that is pretty uncommon uh in reality things take time and gaining users take time and you sometimes make a mistake and you're often making many mistakes in the beginning and you iterate and iterate and iterate and by the time you're finished with this is three years and the world has moved on so what you want to is you want to look at the world um you you want to um you want to look at the world as it is um as it is today and then you want to have to make a guess about how the world would look like in three years i think it is a place placeholder and you want to have some sort of educated guess about how the world that you're building towards uh looks like and we're going to use the examples of building for a world of cell phones in a few slides to uh to give an example for that so uh what you want is not to do this a squiggly arrow right here that says on the left that says uh you built for the world of today but you you'll take the three years to build the project you want to enter for for the project 1024 and when you look at successful companies successful projects that succeed they often have exactly this kind of thing like linus stovall's i don't know if but on purpose or not but was building uh the most amazing operating system for the world we're living in right now you know in the 2000s amazon jeff bezos built uh built a giant starter because he was aiming for the world after the internet takes hold you know facebook google the whole thing is very common um we'll take the example of uber so it is so that's how to build a complex startup in general now how do you build a startup on uh swarm uh you have two factors that are changing the world is changing and also swarm is changing so what you want to do is make an educated guess about how swarm we look in 2024 and about how the world will look in 2024 and aim to get there roughly as the as the world is maturing towards that now to exemplify this let's take a trip back 12 years i am very old some of you might not so i'll tell you a little bit about how smartphones develop how the ecosystem will start to develop but basically suppose you started 2008 and you're trying to figure out there's this new thing it's a smartphone the iphone just came out i think 2008 or so um and there's this new thing which is called an iphone and it's like a pc in your pocket and you feel this is a really good thing to build on because it's going to be big but you don't know exactly how what you don't want is to build for the iphone of 2008 what you do want is to figure out how because it's a flip phone that can run problems like it's nice it's exciting but that's not that's how you get an alarm clock right what you do want is to figure out what happens in 2011 once suppose smartphones get very very big what happens and here's the first thing that happens there's a pc in everyone's pocket and you get network effects right and the uh child on the right is the growth of this is the number of smartphones sold per year in millions of units from 2007 onwards you see exponential increase in the number of of smartphones so right by 2013 uh a billion startups are sold each year oh i i saved at one of the digits and basically most people in the world have a smartphone in their pocket a big probability of them um so what is it um so beyond having a piece in everyone's pocket what does uh you're sitting in 2008 you're trying to predict what the smartphone gives you so there's a pc netherlands pocket it's connected to the internet so everyone has the internet in their pocket it has gps which i mean we know it the benefit of hindsight was a big deal um a lot of people knew back then it was a big deal and it has a camera in a sense already so what do you build right if you're sitting in 2008 and you have all of this knowledge you have the time machine to 2011 what do you build you don't build a an alarm clock app right you don't build something that can paint you don't copy a pc at all what do you do you build an uber or you build an instagram right everyone has a camera or you build a twitter uh you build something that um takes advantage of all of these new things that come one smartphone are available and wasn't the case at all before now for some of you this might be there was no world before smartphones but uh if you read one of the things that insults me a lot is to read about uh times before big revolutions before the internet before smartphones before electricity by the way before steam uh you learn a lot by looking at um what you would say in hindsight like what was the ideas what was the way to hit on something really influential with the benefit of hindsight and then to say how would i come up with this hindsight in advance right and what can we learn from this today so let's talk slow okay so in 2021 what does swarm uh give you in 2020 in 2021 swam gives you uh decentralized storage chunk store which is operational you know it works you can get chunks you can get files you can draw some stuff and you get a few more features feeds and so on and these are developing as we go don't build for that right don't build for a chance though you end up building a dropbox um you'll end up building something like a dropbox um dropbox on swarm which is a very nice idea but phil drive is doing that already started three years ago for the storm of today and um you you build something that by the time it's ready it's already outdated and it's tv on it's ready on the tv like it doesn't serve it doesn't take advantage of the benefits that swarm gives you by the way this is true for any infrastructure that's nascent that is coming up um uh you want to take advantage of the the benefits of that infrastructure in three years so uh what does one give you in 2024 well it gives you amazing thing it gives you a wall database right i will exactly explain what i mean like we'll have a whole part of the talk about that but it gives you a database that anyone can access very easily a via an api or you know a call to the blockchain and so on and um it doesn't um it doesn't need any robustness it's not like you need to manage your own servers or whatever just a a database that is available anywhere it gives you it that database is fully decentralized and central censorship resistant right so um that database what you want to look at when you look at these boxes is not to say oh it's censorship that's great you want to ask yourself in the year 2024 and 2030 for that matter what does censorship resistance really means for example it means that if parts of the internet are cut off they would work just as well right um it means that if you're offline from the internet but you're connected on a wi-fi to another computer that does it is going to indeed internet you will have access it gives all kinds of emergent properties like if you if you read about emergent behavior these um superpowers of storm will give you all kinds of emerging properties that are how to predict just like gps gave us uber right you wouldn't necessarily see that and say oh gps is such a big thing but once there's you know a billion devices in everyone's pocket that all of them have gps you get this emergent behavior that's what censorship resistance is really important for it's not because someone will the us government will crack down on swamp it's because all of the things that you get once you have a billion swarm clients all over the world that all of them are able to access a network and serve content and so on in a censorship persistent way you get an entirely open source ecosystem we'll talk about that in part two of the talk um that means that um that means that unlike the current cloud services that we have that are right now stronger than so on from any application uh applications one will move faster and will be more adaptable to the engineers using it which is you i will get excellent service level agreements right now awa using aws is probably faster than using this one but as swamp develops and gets more prevalent in the engineering gets more robust you'll get better service level agreements because the content will serve some closer there will be many optimizations we'll get to that as well you'll get native integration smart contacts and you'll get generally an unstoppable machine powered by microtransactions we'll talk all about that in the rest of the talk but what i'm trying to impress is that you want to build for these properties you want to ask yourself and make up your own opinion don't listen to me look at swan say what will this look like in five years what do the people in the ecosystem are going to build on it what libraries are going to exist on it and build for that okay now i'm doing something a bit disingenuous but because i'm telling you built for that but um but how do you build for something that doesn't exist yet so i'll tell you exactly again i mean this is how i'll tell you exactly um how i think of it which is you plan right so this red dot is the world the swarm of the world in 2024 the rest of this talk we'll talk about this red dot specifically um what you do you um plan to end up at the dock you plan to build a red dot and remember this red dot is part of a cone it's this relativistic alright you can you might find yourself uh anywhere in this um anywhere in this cone right so um but but uh make make your opinion about how this cone right how it's shaped roughly looks plan for the red cone architect for the red dot right so build something that would make perfect sense once this red dot exists right whatever you can build in this one right now build this one if you're developing use the gps right now it's in any phone it's fine right um but some things you can't build in swamp right now um so what you want to do is i don't postpone the ones that don't matter that much or you augment it with non-zone services if your app if you have the perfect app that would be perfect for swarm in 2024 and will be swam native and all of that but right now in order to serve it you have to use a aws use aws right i mean build the thing that's correct that's advanced and whatever you can't get from swarm negatively use other pieces that can be architected in a way that it can be replaceable right twitter was an amazing um thing that is um built for smartphones like built twitter is specifically built for phones and smartphones and it's very native in there and most people that use twitter use it from their phone but it still has a website right um airbnb still has a website like a lot of these things uh you have to have full backs you have to have the pulse um detailed engineer before before the technology that they're based on comes to full fruition so um don't be a huge purist like if you look at the cryptokitties they are hosted um they are hosted centrally and it's fine you know it's not the best thing ever but they did what they have to do so that's the big thing about startups like built for for the thing that will exist um robert are we asking for uh topic for questions in the middle or should i go all the way people are if people want to ask questions feel free to send them in currently via discord would be the best way and then or for the q a panel in zoom and i'll make sure to uh forward them to you perfect let's do that so robert just i'll stop a couple more times doing the talk for questions and just ask me any questions uh that happened to be when that happens so we'll move now too so that's the first part of the talk bottom line built for the red dot now what does it say can i ask a question then yes please go go ahead maybe you're going to talk about it how do you make sure the dot is not too big yeah that's a great question um you need to be a future startup i mean historically yes historically the people we see succeed a big huge successful influential project are the people who had um good guesses about what the future would look like either they got lucky right survivorship buyers like we don't we don't hear about the people who guessed wrong but it's people that typically make a lot of guesses and most of them will correct um i know futurism is hard um come talk to some smart people that are uh opinionated about the future ask a lot of them get contradicting answers and try to make up your own mind you know um it's very very hard to build for the future but it's kind of something we have to do and most of us will fail and build you know when you aim for something that could be hugely successful and world changing then a failure is still is often still pretty successful so um yes and and listen and listen and don't be dogmatic listen to what people are building listen to many opinions and and figure out what the future really looks like oh you know what maybe the one big um i don't want to get too much into philosophy this is philosophically enough but maybe the biggest trick that i know about um about how to predict the future is a try to think about the future as you think mental and i is you think intellectually it will be not for the future you hope for so don't say uh i hope that the future looks like x everyone is free data is in silo the governments kumbaya everyone seems compare crypto wins by the way bitcoin becomes the strongest currency in the world don't think of the world you want to live in but rather the world that you do live in and how it might look in three years or in ten years for that matter there's a there's a really small technique called back casting back is in back and casting and testing that is a pretty good futurism technique and i recommend it but in general just you know not being dogmatic and trying to think by yourself what things will look like and paying attention is pretty much the best you can do thank you um good so if there's no more questions let's go to part number two and the rest of this talk we'll just explore this red dot and see and form some opinions about what it might look like so i will split this to two parts the first part is for the internet and that will be mostly about economics and ecosystem and the second part sorry so tattoo is about the internet and health play is about engineering it will be very short because um i'll direct you to other resources as well so the world right now is i'm a lot of claiming people are playing the internet is broken but i clearly it can be fixed and let's talk about white spoken and how swamp specifically fixes it or the swarm-based system so the internet you know i'm old as i said uh from the 90s the internet had the big promise of forming a world government getting people from different nations to talk to each other it was supposed to let us share information better and to collaborate better um and that's for you know the work we do and that's true for our fun uh but it's failing so um there's we we we got locked into silos each of them separate each of them only allows to talk to people that are part of it to talk to each other so if you're old enough to remember ilc ilc was amazing people from all over the world came and talked to each other um it was replaced by slack which is uh if i subscribe and you subscribe and we're part of the same organization then we can talk to each other and um you know this course is uh so this code is is you know better arguably but it's not that much better it's all blocked in silos that are of people that are already connected to each other facebook is very nice it allows you to connect with anyone in the world but in reality you connect um you know it doesn't serve the purpose of the early internet stuff so what is swarming how uh will swam help inside of this this situation so you can think of some as decentralized hosting you can think of it as bitcoin 20 incentives but really the best way to think of swamp is a system that stores and serves data on demand any data is accessible from any device right and it's auto scaling it's like the light kind of aws and you'll get to that later aws is also broken um all of these kind of again siloed services that are only accessible from single endpoints run by one commercial company uh relating all of its code by itself or stealing some of it from open source project it doesn't make any the internet as it is right now doesn't make any sense it didn't live to its promise but it also doesn't it's not efficiently in order to iterate fast so let's talk about that the key insight behind swarm as far as i'm concerned the economical insight is that the cost of using an application of the swamp is paid by the user and not by the host that sounds like a trivial thing right i mean why is this really who pays is this really the important thing you know the thing or building in a so on uh but this is the most important part that will allow swarm um and other decentralized hosting and i will centralize projects in in general to uh win and win big and take over the internet i personally believe that in 10 years give or take 20 at most decentralized storage will be pretty much the only thing that exists like there's still tv in the world right some people still use tv but really most people just use the internet similarly decentralized hosting will be the internet of the future like some people will still use centralized hosting but but um but decentralized will win and the reason it will win is exactly because of this ability it's because developers can build software without raising vc funding without uh putting millions of dollars behind for the hosting uh it's a simple economic change which will have a huge outcome so let's talk about that uh let's take a time machine like this is related to backtesting let's all of us pretend that we're going into the time machine we image at the year 2030 and look around us and let's do interviews with the people we're talking to and let's see what they tell us so let's talk to a software developer uh which is not you it's a software developer after this one becomes very common after all the pain goes away and so on uh how does the software developers look like in the year 2030 after swarm has won um so dana built software herself and not part of the company which is a stark difference to how things look right now her salary is through something that is related to swarm to to smart contracts uh it could be um inflation-based funding it's more likely some sort of revenue sharing of network fees data creates features in interoperable software so she's not the sole owner of the software that she develops no one is a solo owner it's an open source software and um she adds features to it with super fast iterations that users immediately use um this is if you use open source software right now it's it's you know this but right now opens up software is experiencing a tragedy of the comments no one is paying only a small number of people are paying for it and it becomes a huge mess but if you're doing revenue sharing through network phase or in patient-based funding with some caveats or anything like that then you can actually have people that make their livelihoods by by developing by developing for this um now she collaborates with other coders just like in open source software and features interface with standard data formats right so everyone uh like dana and you know there are millions of people in the world like them and they write features that interface with standard data formats and allow you to make really powerful software products that are all um interfacing with one another and augmenting one another and the point of all of this is that all of them can develop features separately independently of one another right so that's dana she develops open source software but she's paid not like an open source software model and at that point and the exact reason this is a possible is because of this impact because they never needed to raise funding for uh for help software because she never needed to host it herself her users are paying their hosting fees so if every night they number the number of users of dana's features go from 100 to 100 million um she didn't need to prepare for it at all there's no uh um effect or one of those there's it's just uh the users pay for uh for for the hosting and dana just gets a chunk of the network fees you should you get some sort of um darwinist behavior of features you you just the good ones rise to the top because all of the cost of a single developer is writing the lines of code is the coffee you know it's the coffee and internet connection that's all um now from the other side of this let's talk about how this looks for the end user so for the end user the entire digital life happens on top of the thrombic so right now my life happens on top of all the cloud services that don't really talk to each other and the cloud flare and like my data is split into silos and each company separately keeps very strong struggle in a stronghold of these silos because that's the way of keeping making money but this as soon as you move to a model where uh where the company doesn't have to pay for uh for hosting this completely changes because this open source facilitation model of software development will be much faster than this centralized each company say developed separately and needs to pay for hosting if it succeed right so if a company succeeds then it's penalized by having to pay a lot of money um this is going to disappear so your entire digital life will happen on top of this open ecosystem uh you're working your social your gaming all of those are going to be open i think voting in texas a bit later um and you can because all of these systems are interoperable you'll get a modular experience your people choose right and it's very nice to stack bots or trello butlers but this will be truly modular experiences where you pick and choose your pick and choose how your software looks and there's high competition and it's frictionless because all of the payment is done um for the for the hosting itself right so so that's what it will look like for joe it will be similar to now but with really good software because his data would not be stuck inside so the bottom line of this is that swarm and other decentralized hosting projects at all but swami is very special because it integrates really nicely with smart content but regardless the centralized hosting does a really small change to the economic model it allows users to pay themselves uh rather than the developer pays like you don't have to pay if you succeed and that creates really large outcomes because anyone can try starting over the week and developing features and integrating them into existing products and actually getting making a living focus so the bottom line is that swami writes the last 20 years of cloud cloud is uh centralized and that was very efficient in the beginning of cloud it was amazing we've gone through an amazing 20 years but it's not it's not cutting it anymore uh the network fees will be paid by the user the developer incurs zero cost and uh this open ecosystem will allow faster interactions so you'll get the web 2.0 capabilities of having airbnb you know having apps that all talk to each other and across the globe and everyone interacts with each other but you'll get a freedom like in web one meaning um meaning uh you know if any of your result is me uh having downloading an exe file and having software really runs on your machine um either the local right now they call it local first but back then it was local only and um and single developers that are able to serve this software because they don't incur a cost per user um so that's the big change for um respect to this uh that's the big um change for um for there's a big change for the internet uh that will occur from using swamp and when you look at that that's like the red dot like that's what it looks like the life of dana and the life of joe and what the red dog looks like and now the question for you is how do you develop amazing apps that feed into this world that become a huge value generator or a hugely popular in the world that i'm uh if you agree with me that i'm painting here oh alternatively here are some ideas make your own view of the world and figure out what what the red what your red dot is and how you can build something that will be very influential in the predator so that's um that's about how swarm you change the internet and i'll just say a few words uh next about how swami will change engineering but maybe first if there are any questions about that can be happy uh cancel them now or we can do q and again well one of the questions given the um the future you described and again i'd like to uh point out that you can ask questions by discord as well or via the q a option in zoom in this webinar the future you describe over the past years we've seen other vc funds either start a subsidiary or invest in let's say decentralized focused vc funds socially enterprise focused or not in your opinion uh do they recognize and um well with funding support this future in a sense that you see that there is an uh expectancy there from from there because they most likely will need to uh partly fuel this transition yeah so in theory if you can do if if this there are some theories about how this model of software development will make this funding obsolete at least for software development right right now there's many many of my colleagues that invest all the time in investing in sas like software as a service whether it be things like slack or things like facebook which i think of this as well it's web applications that have multiple users right um and sas investing is a bit is is like the bad part i mean sas investors are doing important work because they're working as part of the current ecosystem to um to build software that people will use and that will create value so uh something it's a bad thing but it's a bit outdated uh i don't think that sas investing as it is will continue to exist 10 15 20 years from now once the centralized hosting uh takes hold um so what will investing looks like so there are some projects that are super risky super high value projects right i mean spacex if you're building rockets that go to mars investors are very important if you're building something like swarm or something like um you know invite has some vital infrastructure for the world that is very complicated to build and it's not clear whether it's the right goal and so on like um like ethereum itself originally and so on uh you maybe you need some external funding i think that for stuff that is built in swarm if you need funding to get started come talk to someone like me i think funding is way less needed because uh again the only cost is coffee and rent and uh computers and i mean you don't need to pay for hosting so overall i think for this world that we're seeing here vc funding is barely for if at all needed and i am thankful for that i think that the humanities allocation of a risk of of risk money should be spent on stuff like biopharma that has a super very high chance to succeed and big capital costs and so on uh if you can do without vc funding do without visiting if you need vc funding for the priscilla that's great uh i i think there are very few projects that need that in this kind of world need um and sorry in the same ecosystem in this kind of world with vc funding beyond that and um about what investors uh [Music] like to invest in um like it's it's a very it's a very traditional industry it looks at stuff that has already worked and it's trying to do the same thing i think that the ethereum ecosystem in general and the small ecosystem in particular has a lot of money floating around for investing in things that will define the future and there's a lot of angel money around with people that can actually understand uh what you're doing and help you and i think that um that looking at vc to invest in uh swan there are many visas i i really like but generally decide what you're building and for most of these things that we'll talk about here i think you don't you don't necessarily need it okay thank you it's good that you went to the slide about dana yeah just go back to that one because we have a question which is what are some of the pain points for data you see in the future so i'm sitting right now in airbnb and it's a bit like that it's this kind of model where you don't have perfect visibility on where your incoming is coming from it's a bit of a um it's a bit of a either you become either you make it into a business and you make it quite successful and then it's quite dependable or you try to make it like a youtube business hydena dana's life looks like a youtuber's life a bit some of them make it really big and the other like always like making something and making augmenting their living in some way so i think reliability is very hard then i might have another day job at the same time she's trying to make it in this this is a passion project that might turn into a career that's on the income side on the kind of in the life side in the engineering side honestly this is a much better world for a developer than anything i can think of uh right now there are open source developers that are having a hard time getting paid so we've talked about that and there are commercial developers that are working for a fee for salary and um and they use the same dana will be able to use the same tools as them exactly but do the things that she has a passion for so um i don't really see any on the engineering side any pain points for dana um of course the field of decentralized system needs to be a bit more advanced by then so we need to we need to get there this needs to be the year 2030 like i don't think this is the year 2023. thank you no further questions at the moment okay so last part of the talk is um is what does a swarm-powered world looks like for an engineer and for that uh right so we're talking about this red dot right so as an engineer you're sitting right now saying uh i want to build a small song is very exciting censorship resistance is amazing what can i build um so the thing to start understanding is that swam right now gives you stuff that you can get that if you don't care about censorship resistance and so on you can get from centralized platforms uh but these centralized platforms are inherently broken and swarm will give you so much more than these in the year 2013 and i think in the year 2024 it will start giving you things that are more powerful than what you get on aws with cloudflare and so on so this is what azure looks like for storage right look at the these are the um azure is like aws is like a platform for uh for cloud services and these are the storage options you get on a show and i uh i i screen capped as you s or entire services until i ran out of room and that's what i get these are monstrous um platforms each of them in its own silo can't interact with the other ones each of them i mean by each of them i mean azure aws google cloud all of these um each of them is is this is the kind of the main way to build the web application and to host something oh an app oh when i android that i am you know smartphone and these are gigantic systems built by developers all getting paid by uh the same company in this case microsoft in the good case or stolen with code stolen from open source suppositories in the fairly common but bad case uh and this is how the life of the web developer whole web developer looks like right now and this is um so this is i mean this doesn't make any sense uh for anyone except these big companies and the reason it is looks like this is historical so we need to uh like why is it okay so it's very nice that geologically i say this is bad company silos yeah they had all of these nice ones but what does it really mean it means that it's super super slow and super super corporate uh and and super hard to get into a news and super locks you into one venue all of these are terrible things all of them i won't get into but let me to say why is it slow because that's the most important thing people getting the salaries from microsoft sit and build this stuff this is a political nightmare they're sitting there saying how how do you get promoted in in a microsoft azure you build another web service so they add each of them your biggest achievement is to add another login and if that happens you get promoted to i don't know this is inherently broken they're not working on interoperability even inside of each of these companies they're silent like this is a silo yeah in itself that yes can talk to each other to the others um and each of these companies are building separately they steal from the same open source projects sometimes sometimes it pokes out um it's it's it's maddening how the internet became the you know the basic statue of the internet became so maybe let's say see what's what is the basic structure of the internet supposed to look like right uh or what would we never mind suppose that's ideology what is the right way for the future of the internet for efficiency uh to build this basic level of the internet so it will look like tcp i think of tcp it's a beautiful um stack of connectivity protocols so when i think of swarm i always think of tcp because it's almost the same thing the swarm uh you know tcp started in the 80s 70s 80s whatever and took 50 years until now to reach um or until the 2000s let's say they it keep kept being optimized and kept being added to and kept being improved and made more efficient until it got to where we are we all are streaming videos some people choose entirely not trivial in the 90s all our streaming videos and talking to each other on zoom and all of the people corona hit and everyone started using the internet even more and the internet persisted there was no hiccups it was kind of amazing uh why does this happen because tcpap was built correctly it was built by a beautiful system of lens of abstractions that was defined but then optimized so swarm is a bit like what you see on the right it's defined but it's not entirely optimized yet so swarming its current state is a chunkster with basic key value store right and a few more features that we won't get into right this is like the tcp conceptual layers on the right uh it is the perfect abstraction for achieving the thing that you need to achieve so participate it's to get packets from one side of the internet the other side of the internet in a robust way and and for some it is to allow you to retrieve chunks in a robust way that's all tcpip however was very very inefficient i don't know if you remember this but tcpip is built using dns you find the path to the internet and the servers on the way tell you where to go this vip in its basis is designed to be very very robust but not necessarily efficient and what happens from the 90s until the 2010s give or take is that tcpap becomes way more efficient because of various optimizations right so the way that routing right now happens in the internet is not um a true to the origin of tcp at all but it's on top there are various optimizations on top of it and that's how companies like cloudflare make their money that's why how you get them um what are these called uh and forget but these these companies that host all the video that you can download from and so on so the basis of swarm is like what i want to take you take from this is the basic of swam is a level zero infrastructure similar to networking and there will be many many optimizations that are coming in the coming years to make it more and more and more efficient and kind of drive it away from its abstractions but uh but keep it full keep these abstractions and fall back so what will you get right what is the red dot of what you will get from swamp you won't get simply a chunk stone that is kind of that stable stakes uh that's a you you'll uh you'll be able to implement dropbox always right but that's not the only thing you get what you'll get is a world database so a entire database that is able to give you all kinds of sophisticated queries database services which is if you think of databases it's there it's the right usage of a chunk store it's not a dropbox it's not a file file store is is too basic file store only allows you to get a file or not get the file that you retrieve it and then you think what to do a database is what you really want in order to build um to build applications for the world um and if you want to read about that uh about how swami will access the world database i don't have time for this right now but check out my talk from um say devcon the one in prague i think it's 2018.

i gave a talk based on work a joint with victor called a swiss which kind of shows how this database and will we look and this is available on youtube um maybe you know this is too complicated um so you'll get one thing in the red dot you'll get the world database what else will you get you'll get integration with smart contracts which actually serves a lot of the of the paying problems right so uh smart contacts will give you a lot of um governance right uh a lot of the exercise a lot of this stuff will come from smart contracts uh you'll get a lot of flexibility in accounting so the payment for all of these services will be can be as complicated as you want to including a smart contact which is quite quite complicated and now with ethereum 2.0 you'll get more scalability for this payment so um and finally it will allow essential censorship free or censorship resistant applications overall as an engineer these four bullets give you a lot to work with because uh you know you can plan for in 2024 having a database that you can just retrieve and start start working with you can start in advertising i know advertising is not popular again you can do an advertising marketplace on it you can do inflation-based fun you can do whatever you want these four bullets i think serve you a lot they will all be i think available around 20 20 24 in a mature way um and if you mature and scalable important to say mature scalability with the service level guarantees and so on uh way uh you should add more bullets to this like make up your own mind about what this red dot in the engineering side looks and build for that and whatever you don't get right now for example a wall database use whatever database you can i mean it's uh either either use swarm and make it in and make something inefficient um or host a database and swap or alternatively use whatever you can and gradually by the time the project is finished and you have a lot of users and so on you will be able to migrate it to swarm as well i think i'll just say one thing if you want a thought experiment that will allow you to start as an engineer experimenting with what swamp will give you think about decentralizing twitter think about what you need in order to decentralize twitter and um you can look at my talk from from a from a theme um but basically what you want is a is an advanced level somewhat advanced level of song which is i think what you have in a few years with swiss um that's it i'll leave you to to look at the other stuff offline um feel free to text me on me at twitter and so on if there's any any more questions or check out my homepage a lot of this is uh already exists inside as someone who's developing who's going to develop an app apps and is considering doing it or either doing it or say you do it in so what i would very much encourage you to do is build for the swarm of the future uh make up your mind about what the world will look like what swarm will look like and build for that architect for it planned for it and whatever you can build right now build on it and whatever not to spawn augment with non-storm resources swami change the world this is an incredibly exciting time to be getting or to be using uh decentralized storage uh this is going to be one of the most important skill sets to have in in five or ten years this is like it's like getting into the internet in the late 80s or early 90s uh it's like the potential set is uh astounding and all you need to do is to build stuff that is ambitious enough so that it might succeed big and that you get the sort of experience that you need in order to be part of changing the world in this way so on that uh thank you i'm happy to take any questions and also take them offline thank you again illa for your inspirational speech um again i'd like to open the floor for questions via discord or via the q a and i have one myself over the years i've also talked to some of your let's say colleagues from other firms and most you'll see that um uh the concept of decentralization in the form of um building your application on a blockchain and offering it to the world is often combined with another current cutting-edge technology to find a new niche um as an inspiration for the people watching as well what are some inspirational projects you've seen recently uh come through at lunar ventures with that you don't have to mention the names but at least what combinations did they try to make so i really like so decentralization in general i really love molecule which is a drug development marketplace for early-stage drug development so it's a they launched vitadao which is a dow for longevity drugs that is exactly using the power blockchain to avoid the blockages in current flow of resources so drug development requires a lot of money especially clinical pricing pre-clinical price and investors are not doing a very good job at the funneling that money today like their projects because of all kinds of commercial concerns that are uh pretty pretty myopic um uh so what they did is they launched a dao called vitado for investing yeah or for um for finding money into that project right and i think that if we manage to funnel billions of dollars right now better that was around five or ten million uh raised um if we managed to um if we managed to funnel billions of dollars over the next few years to pharma to early stage farmer we will do a very good things for the world and make a killing by the way and this these investment projects will make killing also financially um i believe so so that's one that i really love that we invested one that we uh i didn't invest in anything that i truly love is any type which is um a very much decentralized storage project which is a little bit it a little bit inspired the story of dana and uh dana and joe so what i described here is sort of the thing that i believe any type is um is working towards let me let me write any type so we can know how it's written alpha test right yeah i know the project and and um and check it out because it's exactly the missing piece i think between um it's exactly the missing piece between swarm which is a very basic basic infrastructure project and and go and and test for um for apps right this is the way that you get these uh you remember when i talked about um features that are interoperable and data data types that are interoperable like uh um it was somewhat anyway so these interpol types are exactly the stuff that any type is building okay thank you um if there are no more questions i'd like to uh thank you for your time unfortunately for your presentation i'd like to welcome you just like the other speakers on friday when we have the pictures and presentations by the participants of the week uh friday afternoon i hope you can be there i'm exactly an education in croatia so i suspect i'll miss it but uh i do um scott me exactly on vacation but i would love for anyone to you know anyone can please please reach out for any kind of discussion and i'd love to to see everyone and if if i can be of help in any way just reach out and my thank you we'll make sure to forward thank you very much