Are you a DePIN? A Decision Tree to classify Decentralized Physical Infrastructure Networks

Transcript

[Music] okay um before I start like so I Mark basically I'm involved with deepen on different levels like Academia but also practical so I'm trying to show you a little bit um from from that space my quick question is who knows what deep in is who knows and who would say that they are an expert in deepin they already know everything and kind of okay so like so I mean I I designed this talk under this expectation so it will be like I see it as an intro to deepen so like I I will try to show you what is about and for those who maybe have one or two opinions there will be some nuggets uh

on which we can later discuss or something like that so that's the idea so let's start are you a deep in so what is deep in so it's one of these uh verticals in blockchain which is uh I would say emerging and quite thriving also so they are like over 500 systems already building in this base so it's like kind of already a small industry within blockchain where people doing stuff so and um basically the core idea is that you build decentralized physical infrastructure networks deepen so the most famous example which also like kind of started the narrative in the space is like helium you might have heard of

helium so what they did they incentivized people to place physical assets physical device devices so called laavan gateways and they buil these network of devices so you see they covered quite some part of Europe and the US and basically each dot is a physical device so basically they give they gave people or they still mean they still do it they give people tokens and they maintain these devices yeah so in this they kind of build a network um The Core Concepts which Deens are using are basically from crypto economics so as you heard token incentives but also decentralized governance it's like part of it and also other properties of the blockchain like the immutable data storage and some other aspects so it's

like kind of um and I mean it's having real world impact as you see helium also went to cooperate in a cooperation with T-Mobile in the US so kind of some of these networks they they really go out and they manifest the presence out there and uh so still in the end what is the deepen so let me briefly walk you through how we see this a little bit um so here's an example from unoy so basically unoy has this solution for Global navigation Satellite Systems so Global navigation satellite systems like GPS yeah which you have in your phone like to know your positioning basically they have a problem because they atmospheric delays up there so basically when you get the signal there's some

error in this signal so your positioning is only accurate up to some meters and if you want to go to subcentimeter accuracy which you require for some applications out there it's not good enough and there's an established technology which you can use it basically requires you to place an antenna for example on the roof of your house which knows exactly where it is it gets the same signal can calculate a correction term and share it with nearby devices so it's like kind of a stablished this technology exists and now what unoo is doing they incentivize people basically with tokens to place every 20 kilom such a station on the earth to kind of provide um yeah this coverage and uh this service basically so like

this kind of the first take and I think probably the most famous take in deepen it's like about these physical assets yeah so it's like defi is like kind of lot of virtual like virtual um space and here blockchain kind of comes out and um penetrates so to say like the physical world so physical assets placed so deepens are kind of concerned uh with this aspect so it's the first take on what the deepen is a second take that actually deepens they have these three sided platforms yeah so like kind of makes them most complex in the sense you have these infrastructure providers so those who kind of place these antennas yeah then you have sometimes a community sometimes a company like a DA

could also be sitting there who kind of takes this infrastructure and kind of creates a product out of it and then you have a demand side yeah and when you design a deepen system you need to take care of all of these three aspects so one concrete example for example what we're doing in W high it's like the same idea with unoy just for weather data now yeah so basically you incentivize people to place these weather stations you collect weather data so kind of these are usually like hobbyists Hardware enthusiasts kind of a very particular community and then on the service s you kind of have experts people working at metrological organizations and universities who can use this data to actually nowadays like use AI to do forecasts yeah so kind of an expert community and then you need to package

this into a product for the customers who want to have intelligent Services yeah and you already see is they kind of quite um different requirements you need for each of them to satisfy like this so it makes it um yeah interdependent so and this is a second take on deepend so usually you have these three-sided platforms yeah so that's like what deeply characterizes deepend and also in the beginning I mean still gives a lot of challenges to satisfy all the demands um and the third of course we wouldn't be here kind of deepens they use tokens as a coordination tool yeah so like it wouldn't be a deepen if it doesn't have a token and um so why do you use these tokens so one is like I mean obvious maybe maybe obvious for

some like to kind of reduce capital investment cost yeah so like for unoy we calculated you would need to have $80 million if you give it to me I could build this infrastructure out yeah covering the whole world the problem is you can't really make a sustainable business model out of that yeah too many up front cost not much not enough demand at the moment so kind of what we do you can kind of crowdsource crowd Finance the investment in such a network through tokens the second which is actually much more interesting is like you it can lead to self regulation like kind of optimization you can set incentives to the Token so that your network learns so for example what happens in unoy someone wants to join they buy an antenna they place it in the garden under the tree

and they're happy they're waiting for the rewards and then they get literally no rewards and then they come to Discord and they ask oh I I bought a device I don't get any tokens why and then someone from the community jumps in and said ah where is it placed you should put it on your roof you know more coverage of the sky you get better rewards so a person does it and then still bet rewards and then Community gives support again and ask is antenna still does it move have you touch the cable what type of cable are you using kind of and kind of we as a DA or something like this we don't need to micromanage and everyone how to do it kind of a continuous learning through token incentives which makes it quite lean um yeah there different models single token models up to multi- token

models the most famous token model in deep in deepen is like this burn and mint token model some of you might have heard of it I will show you how it can look like so you have like kind of two tokens here in the example Uno and data credits and uh the Uno token is like kind of what you expect a blockchain token to be yeah value appreciating transferable all of these aspects and you have a data credits token and the Credit Data credits token is basically more seen like as I said in the name a credit so it's non-transferable you can have it you can spend it on system service but you cannot transfer it to somewhere else and

through this we can actually make the data credit to be a stable coin within our ecosystem without it being a stable coin yeah because it's non- transferable no third party Market can form it's just within our ecosystem but we kind of can bind the price to dollar for example and as you see in this uh three sided Market on the demand side usually you have very traditional players like conservative industry like this and they don't know much about tokens and the data credit is very easily explainable it's like just a credit on AWS or something like that and um yes and the onos kind of um pack I kind of bound to it you need the owner to get this data credit and this token is rewarded to the Reference Station

owners gives access to governance and in this way you have a burn and mint token model because the Ono token actually gets burned in order for the data credit to emerge um it's a very quick intro I know I but I don't want to overload also so I will stop here we have a very technical paper on it um if you're interested in some uh end term state of the economic model then you can uh read it or ask me and I will tell you more about it but otherwise that's this third take on um on a deepen so what is a deepen basically it places physical assets yeah it is a three sided platform and basically the placement of these physical assets are incentivized with

tokens and basically this led us to the creation of this design guideline and not design gu I'm sorry a classification Tree in collaboration with Andrew um who is a I mean I will later have a slide on him like featuring his project and basically he needs he needed such a classification tree and we developed it together um basic you see these three core questions on the top is it a three-sided platform are there token incentives and are there physical assets if you kind of fulfill this you are a deepen if not there might be you might be a project who kind of offers a service kind of to deepen projects because like it's also like a little bit narrative you wanted to include those like kind of you call deepen support

systems and then you can also of course not be a deepen and to give you an example for example in this deepen space something I have not told you about uh there's like still like one important differentiation which you can do so there's kind of the physical assets which you place are they fible in the sense of is it important that this device is placed here or could it be simply replaced with another device somewhere else in the network so for example for weather when you have a weather station having it placed in my garden gives different readings than having it placed in your garden or on the roof or below a tree so kind of each data point is kind of um unique and you cannot just say oh this

one is dead I take another point so kind of it's non fungible in this sense like this fungibility is important and in unoo and whyi for example we have seen it's um this basically location dependent deepens you can also have location independent ones so for example when you think about file coin decentral storage I mean there for redundancy purposes there might be some was a little bit important where they place but in general you have storage here and you can just replace it with a storage somewhere else it doesn't matter too much if it's in your house or in my house it's like kind of the idea behind on the Other Extreme we have like examples which are not deepen so for example Unis swap it doesn't Place physical assets so it's not a deepen um you have like um for example Uber you could say it's a three-sided platform

yeah they have like a they have a Supply they have a demand and they have kind of a centralized company kind of which uh um makes a service but they don't use token incentives yeah so they are also not to deepen you can have participatory sensing communities or citizen science which kind of links it to research which also build such infrastructure networks you could say but they usually don't use token incentives and we also have a paper on basically that actually deepen is token incentivized participatory sensing so kind of adding to participatory sensing the these token aspects and then you have projects kind of who particular build for this deepin ecosystem and which we're also including like iotex and Peak um yeah so

that's the input I mean there are some challenges I want to give you some notion in the space like of some info so I mean there's this huge promise of providing real world utility also got me excited so kind of I think these applications are very easily to be explained to my my father or my mom like you get a better weather weather forecast you kind of have more accurate GPS for your Rover and the Garden or something like this so kind of I think it's a really cool story to tell for policy makers for Layman people why crypto makes sense it kind of brings these values of co-ownership Co coordination cooperation kind of into the real world and um kind of it's very

promising I think um but there's of course also problems so like uh very often I would say these deepens like all over crypto again they just focus on the supply not really having demand in in in their mind and you see all of these sensor networks for example who collect a lot of data and um and the problem is here that in the end data really has no value like in most cases so no one is really willing to pay for it and um and I think it can it could result in this disillusionment so now there's some excitement people people come in they think they're joining a cool project and in the end it will be just yet another failed Rock pool project so I think we need to be a little bit careful here so

I think what to watch out for is like in this data domain for example for projects who go into existing markets but with curated high quality data exactly knowing that this data is actually needed because there's a market or otherwise my maybe if it's more explorative which we do with why high so it's not no guarantee but I think in the end you need to put the data into usage so kind of doing some intelligence service on top how many three yeah perfect I'm I'm done I'm done uh that's very good okay so that's um basically here so if you want to if I now got you excited like it's cool like deepen is the thing you want or you want to learn more I don't know um there's like one cool initiative currently forming at at

the trust square like uh it's basically it's called public infrastructure networks community so like it's not deep in community but for you it's deepen community and basically we're trying to bring different stakeholders together like kind of existing industry players like uh mobile like mobile pH carrier companies like energy grid providers like these different stakeholders and connect them with adoption Partners like um startups like deepen startups have this academic touch in it and also the broader ecosystem of blockchain and in the real world like kind of bring them together to work on concrete use cases yeah so to kind of give um startups maybe a channel to have a customer corporates to learn Academia to study

some nice topics all with a goal not to burn the technology because usually what you notice is like there a hype all the people jump in everyone tries everything and after half a year the technology is dead and no one's touching it again and we're kind of trying to build like kind of a sustainable thing around it with a medium term goal of unconference in Q3 so stay tuned if you want to join let me know I'm very curious and last slide before the end this my collaborator on this work it's he's Andrew he has a very cool website which called who loves burrito. comom need to check it out to know why it's called like that but actually he's a very intrinsic motivated person open- source project and he basically crowdsourced all the deepen

projects which are out there and I would say in a neutral form he has classified them and it's really unbiased so he has no I mean I don't know like but it's my my view on it and um I think he's also very happy to share this data so if you're resarch or something like this um I think you could always collaborate with him having this said this was it so thank you very much and um yeah thank you very much we can get only one question thank you thank you very much for this nice presentation um I have a question a bit

more General regarding deepin applications and if it's too technical or just too broad I'm happy to take it in later I'm very curious on your position of looking at deepin compared to the problem of having an economy of scale that a normal serice side farm would have compared to for example the cost of equipment that individual peoples will need to carry for the same amount of computational power not fully like um what's a question like uh excuse me so um if you look at deepen and the cost of the infrastructure that is required for individual people to carry yes compare that to the economy of scale of a centralized location yeah I mean okay

yeah good question I would actually argue like in this setup that actually deepen setup like if you if you take also the community approach on the on the platform side like kind of a service side like building it actually um reduces costs because like usually when you build these cross um country networks like you have different regulation in different places um I mean sometimes it's diff difficult to maintain them in particular areas and like this so go going this community route so to say kind of reduces these operational costs you kind of Outsource it also to to the community and I think what deepen tries to do is like kind of create the synergies of the economics of scale on the service side so very often

at the moment you have more or less centralized parties they sitting usually uh corporates like kind of TR pushing it with of course goals to decentralize I would still argue a dow setup could also make sense on this end but it's I think deep is actually very pragmatic in this sense and they're probably seeing what you're seeing so they kind of decentralizing very often on the infrastructure side but then on the service side they are more centralized and kind of trying to yeah be more efficient and these things yeah yeah thanks thank you very much Mark for your talk and presentation [Music]