Helium CEO: Crypto-Powered IoT Network Strategies

[00:00:00]vinod khosla who was our series a investor with Coastal Adventures like he once said to me like startups just have to stay alive long enough to get lucky and I think roughly that's true like if you're working towards a problem that's worth solving and you have a good team eventually you will kind of figure out the right way and assuming that those two first things are correct and I think that's kind of where we were is that we knew that this was a big thing to solve and if we could solve it it would be valuable but we also don't know how to solve it and so we we spent a lot of time going around in circles you know with different approaches until we sort of ended up where we are now welcome to Good Game a podcast for crypto insiders with your host Imran and ciao welcome to Good Game episode 13. uh we have a very notable guest Amir halim the founder of helium who is now part of Nova Labs that is building out a bunch of other products on top of helium such as the helium phone service there's decentralizedn decentralized VPN and many other products but yeah I'm pretty excited about this because proof of physical work is an area that is probably overlooked and is starting to gain traction we're starting to see more and more startups that are building in the space that are applying to our program what are you looking forward to when when talking Amir we spoke a lot about uh helium on episode 10 or something like that but for those who haven't watched that episode helium you know you might describe what helium is but also what's interesting is Amir spent five years with helium before started using crypto so he started working on helium in 2012 and it wasn't until 2017 that uh he started thinking about using token incentives in bootstrap the network inside so I want to talk to him to learn about this whole journey how he stumbled upon crypto and token incentives and how things were like in the pre-crypto world it must have been really difficult because bootstrapping the the network effect solving the co-star problem it was impossible without crypto so that would be my main area of Interest all right let's dive deep with Amir at Helium Amir welcome to uh good game today we're going to be talking about really the framework around if you're found in a space how to build a proof of physical work startup as one and then the other area that I want to carve out some time for is thinking about like your journey and like how did you find product Market fit and obviously this is like a long-term Journey especially given the fact that you started helium before Bitcoin right so there's some interesting elements of how your path as you think about building this kind of iot network then shifted to being more crypto native so there's just a lot of gems that I think Founders in the space can learn about and use that as a way to infuse when the ideas to build their own startup moving forward but before we get started ciao do you have anything that you like that's it really excited just to learn about your journey a mirror and I'm sure we'll have a great time awesome yeah thanks for having me so maybe Amir like I watch some of your podcasts and I know you're a big gamer like number one like Quake gamer in the world I think it was I was yeah there was a that's a long time ago this was like late 90s or like mid to late 90s but yeah I I haven't played any video games in like the longest time but uh during college that's basically all I did you know like late High School in college I just was playing uh I was obsessed with Quake I couldn't believe that you could play other people on the internet which was like and it was just doing such an incredible thing at the time I remember that I remember downloading the demo when it first came out and it was pretty eye-opening yeah no I mean we were playing Doom before that yeah that wasn't TCP it enabled so like you yeah I remember literally like you would go into a bulletin board No One's Gonna know what a bulletin board is but you would go there and you would like literally do you just post your phone number right and you would then go back into doom and then you would load doom and you would sit there in like receive mode basically just like waiting for someone to call you that's how online gaming worked in you know 1994 or whatever and then so Quake was like the first tcpip game that you know worked over the internet rather than using you know dial-up or ipx lands and stuff it was super cool it was such an amazing time it was a great time but soon after you met uh was it Sean Fanning from Napster yeah he was also doing video game stuff he had a company called rupture that was doing things around World of Warcraft so it was it's sort of like a really early precursor to like what Discord and stuff became you know like just way too early but like in-game messaging and you know recording and like some parts of steam were sort of built in there and so really really early but we had a lot of like mutual friends in the Video Game World and that's that's how we met probably like early 2000s and uh there was a uh an element of like you wanted to build a startup around that time and then I think you came to the idea of realization of helium why did you want to build in the space like what was the opportunity that you saw and what clicked for you internally where you wanted to spend the next decade or two building in this area I think it was about 2012 at this time so it was actually after Bitcoin was was around ah okay I was too dumb I didn't pay any attention to bitcoin at all like I didn't understand it I didn't some people were like I sometimes read people's old stuff about Bitcoin and I'm just kind of Blown Away that they figured it out so quickly but for me I just I didn't understand what was so special about it and it didn't click for me for like the longest time so helium though in 2012 we have we still do I mean I still have a lot of like entrepreneur friends and what was going on there at that time was that people were really interested in building Hardware startups like nowadays it's like no one really tries to build a hardware startup everyone knows that it's it's hard and painful but there was this period of time where like iot had just become internet of things had just become like a a term that was cool and there was a lot of VC funding going into that space and so we we knew a lot of people a lot of friends of ours that were building hardware-based startups right so like one friend of mine uh Chris Bruce he was building this like a baby monitor kind of like a Fitbit for babies you know like there was you wanted to know about SIDS and like wanted to prevent these like early deaths and like also just know when your baby's asleep he had that idea another friend of ours was doing people counting in like retail spaces you know it's way before like this was just a thing built into Google Maps which it is now but kind of like web analytics applied to the real world right and the problem for both of them and everyone else we talked to was that it's incredibly expensive to do that at the time like you had to use cellular radios and it was you know 40 50 a month in plans and no way you could use batteries on a cellular radio so you had to figure out how to power the thing and just deploying a lot of sensors was incredibly expensive and incredibly hard and physically impossible in a lot of cases right like if you didn't want to use Bluetooth there was kind of no way to do it so that was sort of the Genesis point was like we had a lot of people saying like it would be really cool if you had like Bluetooth that could connect to the internet you know just wherever you were and that was sort of the start of an idea of like well why isn't there something like why isn't there like a a sub voice you know protocol for this is the way we were thinking about it or network for this and that was sort of the Genesis idea of like it would be cool if there were just like a Public Access you know low power wireless network that was the start and it was sort of driven by friends of ours and people we know just kind of wanting it I would say like um this was a journey right you know at first you decided like you know okay we're gonna build in the space and we're gonna start with like you know problem right this problem is like we think internet's gonna be very big there's a huge tab for it there's all these things that are happening already on the internet but Internet devices and specifically like smaller sensors will need longevity in terms of battery and then obviously like internet connectivity 24 7. and so you went on to kind of like figure out what the problem is and like really figure out how to solve it and I was reading through the history and and I think there was like a few times where you changed the model completely so maybe we can start with so Founders that are listening you know we first went over kind of like the area of opportunity the problem and now it's like figuring out the product Market fit and I think this is where you've evolved over the course of like let's say 10 years yeah yeah yeah let's talk to the first the first product and then how did you change from that first product to the second funnily enough the first product looks exactly like this product except we just didn't know how to do it properly we knew that it like a a big sort of public access Internet of Things Network would be useful somehow right and we had ideas around what the most valuable applications were and what the best use cases were in startup terms generally building platforms is really hard and generally a really bad idea right like if you if you think about some of the most successful platforms like AWS they came out of an internal need the desire wasn't like I'm just going to build a platform right it was internally we need the ability to like start and stop servers you know whenever we want right and we want to be able to scale those up and down and so that was an internal Amazon problem that they solved by building this set of products that then became AWS and so generally like I think if I were to advise anyone on like startups and founding it would be starting with a platform first is really really difficult and we had that problem like so we had lots of different problems first of all neither owner I had any idea about Wireless technology or Hardware in general right like we were always curious about it and you know we had like some rudimentary understanding but we didn't know how to like do the things right like how would you actually go about you know building a wireless network like how would the protocol work like what Hardware do you use so that was one problem the other was then how to like incentivize the creation of this network right because if you if you have to build the network yourself I don't think it really solves much of a problem right because you you can't get the economics right like building networks is really really expensive and part of what was really important for these iot devices is that the network is really cheap to use right so if you have one company like building the network they're going to incur a lot of costs and there's going to be a lot of pressure on that company to like recoup the cost right and the only way to really recoup it is to charge your network users a lot of money and there were lots of startups that tried this you know Sig Fox was probably the biggest one it was this French company raised hundreds of millions of dollars I actually did a really good job building you know like a global iot network but they they ran into that problem where they were spending hundreds of millions of dollars and the kinds of sensors that want to use a network like that were interested in paying you know like a dollar a year or something right and so it was impossible for them to recoup their hundreds of millions of dollars of investment as well as making the network viable for the users so that's what we kept seeing over and over again and so we didn't want to we didn't want to copy that model because we knew it wouldn't work you know like it would just be like being a t but poorer you know and not knowing anything about wireless networks which is like the worst version of at T so we went through lots of different iterations about around like how we might do that you know we tried to incentivize developers to do it by you know giving them access to the network early and cheaply and but we didn't really know like maybe just on that note um the iteration process the feedback loop process how did you realize like some of the ideas were working weren't working I think you've got to talk to your users a lot so one thing I I think we got right is that we focus very much on developers as our first set of users right we knew that that was going to be the most important group and so we gave them what were at the time called Bridges there like little hot spot type devices you know that look kind of like a Wi-Fi access point but we're for for iot and we also gave them development kits right so so stuff that you could build sensors out of that would use these access points and so we got a lot of feedback about like what was good and bad and whether it was even interesting at all and you know like what some of the problems were and most of the problems were oriented around range you know like the thing needed to like kind of work everywhere otherwise it wasn't particularly useful to anyone and so you needed extremely long range from the access points and then you needed a lot of access points you know like you needed hundreds of them per City in order for like a city to be you know broadly covered so we kind of were stuck in this place for a long time where we thought we had interesting technology we thought the developer experience was quite good so people that wanted to build sensors like had good tools to do that which was the other part was one other part of the problem but the network coverage was a real problem right there were not that many applications that were very interesting quite honestly if they only worked like in your house or in your office or in your factory like there are some but we thought the biggest and most interesting applications would involve Mobility really right would it would involve being able to like move around the way you do with a cell phone and have network access so that was the thing that we kept running into over and over again and we pivoted like several times trying different versions of like iot products that didn't need full network coverage because we didn't know how to get it yeah yeah we went back and forth many many times and luckily like all of those ideas were kind of good ideas and they allowed us to keep trying and iterating and fundraising because there was something there like there was an interesting idea there and there was an interesting problem to solve and it was just it was sort of like this matter of time like vinod khosla who was our series a investor with Coastal Adventures like he once said to me like startups just have to stay alive long enough to get lucky and I think roughly that's true I if you're if you're working towards a problem that's worth solving and you have a good team like eventually you will kind of figure out the right way and assuming that those two first things are correct and I think that's kind of where we were is that we knew there was this was a big thing to solve and if we could solve it it would be valuable but we also don't know how to solve it and so we we spent a lot of time going around in circles you know with different approaches uh until we sort of ended up where we are now I mean you said earlier that you identified developers as your main group of users right and you also said you want to talk to users all the time in practice how often were you talking to users and in what in what way were you like interviewing them over Zoom well a zoom didn't really exist but like when you're talking to them in person like did you do user service how did you gather feedback from from users we did a lot of different ways most of it was in person honestly like we would just have conversations with users like pretty much every day we also used I mean before there was slack there was a atlassian product called hipchat and that was kind of you know IRC was kind of the original like Chat thing and then hip chat was kind of like an evolution of IRC and so we had this sort of like hip chat server and we had some of our developers in there and we would just talk to them that way you know so it was like slack or Discord is today and so that was the fastest way to get some of it was just email but you know doing it in in person or in real time I think was the most helpful because you could kind of go back and forth and just ask a lot of different questions and some of these were you know individual developers just sort of random people and some of them were companies you know that we were talking to lots and lots of different companies early on some of them became big and successful like ring I remember when we were talking to when they were called doorbot originally you know and they were exploring you know like could they use helium like really really early on and so some of them were like startups some of them were big companies you know and some of them were you just just individual developers and so we we had to talk to them lots of different ways but I think it was just important to talk to them often because it helped us really understand what was good or bad yeah just uh on that question I feel like like especially early on you were tackling like a really big problem right I mean like you have Hardware as one then you have software as to and then over time there's like developers and then protocol and then we can dive into like the moments where you kind of changed over time where you started to take on additional networks as an example but let's let's just touch on like hardware and software I feel like that's generally a very hard problem like usually investors don't invest in Hardware nowadays because of the capital expenditure Etc so maybe walk us through the hardware implementation like how did you come up with the hardware component and were there challenges around that because I would assume you know getting resources getting it manufactured and shipped and their product ties in such a way where customers have like an apple ask like experience I mean that all comes you know together into like one bow so like I'm curious on like how did that Journey come come together yeah I mean I think yeah Hardware is incredibly difficult not because Hardware itself is necessarily hard although I think there is a component to that it's just that the iteration cycle is so slow yeah that's what's really so painful about it I mean when you're when you're in software you make a change and you recompile or you refresh the browser or whatever and you know you see the change like immediately and you can you can iterate extremely quickly that way with Hardware there's like some element of that like you can use some development kits and you can do stuff you know in this sort of hacky way but when it comes to like trying to actually build a product the Cycles are very very long like you design it you prototype it and then you you know one group of people or one company manufactures like the circuit board and another group of people do the actual like placement of chips on the board and if anything goes wrong in that process then it has to go back and you know so every every cycle of iteration would take two to three months that's why it's so Capital intensive is that you you know it takes so long to like go through these cycles and it's starting to improve like there are new companies that aim to solve that problem more quickly using 3D printing and like doing other stuff that allows you to get to you know prototypes faster uh but it's still it's still really slow and so it took us a quite a long time like that you know there wasn't anything that resembled like an established standard for iot devices at the time right like there were at least three or four different competing ideas or Technologies you know one of them was this this company called ingenu uh which was a bunch of ex-clock on people that had their own protocol sigfox had its own protocol there was a new technology called Laura that had come out in 2012 or something and then there was a bunch of other stuff and you know and so it wasn't obvious what to do they all had massive drawbacks in some way like engineered stuff was proprietary Laura was proprietary Sig Fox's thing was proprietary everything had a a challenge when it came to like figuring out how to use it and cost and everything else so it was it took a lot of iteration I think we went down the wrong path several times where we focused on things like throughput rather than range which was you know we thought throughput was more important like people would be able to send more and more data but I think that was in incorrect like range mattered a lot more being able to send less data over like massive distances was much more valuable to that group of developers than being able to send a lot of data how did you come to that realization mostly just talking to them you know like talk talking to developers that were building various things I think everyone would love the idea of being able to send a lot of data over long distances and do it power efficiently and cheaply but you have to take some set of trade-offs and so I think the trade-off we went for was was speed but it it was the wrong trade-off right and so it turned out that it was a much better idea to like allow people to send data over really really long distances because it meant you needed less you know access points in a city for example and you know the end devices as a result were able to do stuff that you couldn't do otherwise right like if it was tracking packages or or whatever and so so that was like one of the important realizations and that caused us to change direction significantly away from the technology we're using and towards at the time Laura which is what we ended up using in the in the eventual crypto implementation of helium but that was you know that was an example of like one of the learnings where there were like so many different options and you just you didn't know what to do at the time and you just had to like try a bunch of stuff basically and talk to developers and sort of understand and the hard part about it is that the developers don't know either for the most part because they're also like just inventing new products and they also don't know exactly what they want and so in the ring case like with with doorbot the speed was kind of interesting right because they were trying to stream video and they were trying to do other stuff but then it became like why do why even use this why not just use Wi-Fi right and so they that there was always this sort of like no one was really sure how how to do it and so you were learning while they were learning and that was what you know you had sort of two sets of things moving and that's what that that made it much harder to figure out it's interesting that it just has a sign out um that you pointed out that one of the difficulties is the fact that Hardware iteration is just so much slower than software but it's actually very analogous to crypto software development because crypto software Rama is also very slow because you have to do very lengthy and expensive audits between upgrades between contract upgrades right anyway just an interesting side note side observation yeah no it's very very true and also when you your network starts to succeed or get big whether it's a network like helium or any other protocol this also becomes governance questions right because it's like the decentralized protocol that you can no longer just like unilaterally change right and so it does slow down and become like Hardware where like every change now takes a long time because you've got to get consensus from the community and the technical changes are complicated and so yeah it's very similar I kind of thought of that of that until you said it but it it does feel exactly like that and all these learnings that you talked about on the hardware set right all that was pre-crypto right and at some point you realize crypto played a huge role in the product when was that and how did you come to that realization early on we were screwing around with like what how to incentivize people to install these access points right so like imagine that the problem is you want to build a network and ideally you don't want to pay for the network right like you just want the network to be there so that developers can use it but you don't want to pay for building it that was like really the problem and it's not because the hardware is expensive by the way like the hardware can be quite cheap it's like where you put it is expensive right you need real estate and you need internet back holding need power and that's what's expensive about building wireless networks whether it's for a startup or whether it's for Verizon right like it's expensive to find real estate and so we were screwing around in the early days I remember it was Mark Phillips who's our are still our VP of Business Development he was like our first employee he was like we should put Bitcoin Asics into the access points and I was again I had no idea what Bitcoin was like he was much smarter than I was like he was in the ethereum Ico and like he had been paying attention to the crypto space closely and I hadn't and so you know we kind of joked about it we thought like haha that would be a funny idea kind of thing thing but but didn't take it seriously and and also realize I think that it was not cost effective and it wouldn't really have worked and you know Bitcoin Asic mining was was evolving very rapidly at that time and so the hardware would immediately have been out of date and would have been useless it wasn't until probably 2017 or or 2018 when this came around again so during this time like we were still iterating and trying to figure stuff out and like again feeling pretty confident that there was something to do there was a problem to solve and like building lots of different product variants to try and solve it and then it was sometime in 17 or 18 where we started to like get a little bit more serious but still kind of joking about the about using crypto as a as a solution and part of why at least for me was the filecoin white paper was the first time that I took the idea seriously because I was like here's a credible team that had built ipfs which was an excellent piece of technology so it was the first time like in this Ico bubble that I had seen something that looked real right like this was a good a very good team trying to like use crypto economics to solve like a very specific problem right and building a consensus protocol and a civil resistance mechanism and like everything geared around the problem they were trying to solve so rather than just like slapping proof of work onto onto everything which was sort of the way other file storage platforms had been working up until that time they were trying to like create a proof of something that was related to the problem they were solving like you know proof of space time or whatever it ended up being as a way to solve the problem of storing files and that for me was like a big like light bulb moment I was like okay you could do you could kind of do proof of anything right and is there is there a proof of existence that would that would make sense in a wireless network that would incentivize people to start building the network because you had this massive chicken egg problem that every Marketplace has right where you can't have sensors until there's a network and you can't have Network until Those sensors and so you needed some way to like break that tie and incentivizing people to start in the same way that Bitcoin and filecoin and others had done was what I think we figured out as as a thing that needed to happen and we figured out an interesting way to do that wirelessly that ended up being called proof of coverage and so the crypto component the the token launch itself did you have to do anything special prior to the launch to make sure that it would really work and get the effect that you hoped for and really you know get the coverage and the token into the hands of enough people because for example in the early days of Bitcoin like what Satoshi did was he he told his friend helfini about Bitcoin and then helfini told a bunch of his friends and then they all started mining right like it and then they built this bitcontalk.org forum where all the early Bitcoin is already on you know hang out and then the words basically started spreading did you do anything Brute Force to put Force the network first of all for the the thing that was harder about helium than something like Bitcoin or you know software-based networks was that there was the hardware part there was a physical part of it right like proof of coverage only worked if one hotspot as they ended up being called would sort of transmit this encrypted packet and it would only work if there was another hotspot that could hear it right otherwise you could fake the whole thing unless there was an actual you know Hardware component to it so one part of what we needed to do was make sure that the hardware worked and the range worked so we spent a lot of time while the software part of the protocol was being developed just making sure that the range was good enough you know so we had all these like handheld devices like literally just walking around with them in backpacks and In Pockets just like constantly checking would this physically work so that was one thing and then the next thing was you know just trying to get friends and family into like an alpha test State this was like before the network launched before any tokens existed and again just seeing like okay if we gave these to like people that weren't necessarily technically Savvy or crypto people at all would they put them in places that actually could talk to each other right because not everyone is going to be a telecom engineer and go put the thing on the roof with a big antenna like most people are just going to stick it in their window or even worse like stick it in their kitchen counter or something so there was that kind of thing that happened and so we spent a lot of time testing that and then eventually when it came to be ready to launch because it's not like a software Network like you need geographical like diversity right like it can't be 10 000 Asics in one room right like it has to be like spread out across the country or the city or the world we had to just go kind of old school with like digital marketing like we were marketing heavily on like Facebook and Instagram really in those two places like Google to some degree as well like really sort of old-fashioned you know like buying ad banners and buying you know placements uh was how the word initially got out and of course everyone thought it was a scam right I mean you see one of these you see one of these ads on especially in 2017. you think we probably started in like eight eight late 18 which is even worse because all the icos had just like imploded right and it was clear that everything was a scam right and except except for like three things or whatever and so everyone was right to think it was a scam like you see some Facebook ad for like you know magic internet tokens flying out of radio tokens or whatever it sounds ridiculous so getting you know the initial batch of users was incredibly difficult you know and it was also difficult from the point of view of how many hot spots should we build right like you're building a new thing which has no like product comparison that you could really look at right except for maybe like Bitcoin Asics but Bitcoin was Bitcoin and you know it's different and so we didn't know how many to build right so we built I think 3 000 of these things initially and that's by the way I think the other really hard thing about Hardware startups is Inventory management like you have no idea how many to build of these things and like you know you can really you can really put yourself in problem in trouble by either building too many or too little and so we built like 3 000 of these things and it was a real struggle you know I think on our launch day there was a good TechCrunch article about it I think we sold 150 hot spots on that day and then after that we were lucky if we saw you know single digit a day was kind of the going you know that's how it went for the longest time is that every day we would sell very very few of them so yeah that was but it was tough going at the start you really really were pushing a narrative that no one really wanted to hear at that time did you manufacture your own Hardware because today you have like a bunch of third-party Hardware providers like the network really has decentralized itself but back then when you first started who took the ownership of Hardware manufacturing and quality and so at the start we we wanted to do it not because we thought there was a bunch of money to be made there like we were certain in fact there was no money to be made there and like we actually lost money on every hot spot that we made I think part of what caused helium to succeed was that we took two things that were typically quite difficult for a consumer to do and made them easy to use right like there was the Telecom side of it right like setting up an Internet of Things Gateway which was not easy to do right like it involved SSH and clis and you know just it wasn't designed for consumers right it was designed for like Telecom Network operators and then crypto mining was the other thing like if you ever use like an any kind of Asic like an ant Miner or something like a shoebox full of wires and like none of it is designed for consumers right it's like really for professionals so part of what was what we thought was super important was making it super easy to use setting it up with like a mobile phone app you know making the hardware look somewhat like respectable you know so it wasn't the shoebox full of wires and that was you know that was why we did it was to sort of establish the sort of reference implementation basically of what we thought a hotspot should look like we open sourced all the software and like everything everything to do with it and then we quickly started talking to third parties now they're 60 or 60 odd or something like that but again it was very slow going at the start because like who's going to build a hot spot right right like the network doesn't exist the token is does no value doesn't exist either really and so at the start it's kind of impossible and you've got to sort of just force it to occur because it's not like anyone's going to take any serious like no one's going to take the risk of building 50 or 100 000 hot spots for this new you know helium Network that you know barely exists and could be a scam up until that point so you launch the token 2018 you really got into the crypto part of the product in 2017-ish that means that prior to that you have spent already five years before crypto like just entirely focusing on the non-crypto native product help us maybe visualize how the runway you know the team size you know team organization culture how those things evolved in those five years because it must have been really difficult to iterate to having to deal with hardware and not being able to hear it quickly it was there was at least one large layoff that we did before the crypto pivot when it was clear that the path that we were on wasn't going to work so I think we had you know 50 or 60 people at that time and it became clear to me at some point that the product that we were building was just not going to work right and so it wasn't sustainable to have 50 or 60 people this is pretty crypto so there was a layoff there and layoffs incredibly difficult right because it's not only a bunch of people that you like and you know but the remaining team are also friends and you know so it's very difficult to to see your friends have to leave and it I think calls into question like why are you still here and like what are you doing kind of existential questions so it's incredibly difficult to survive through that kind of a pivot and that kind of a layoff and I think it's very important to have a very strong team culture or team Bond during those periods otherwise I don't think you can make it if people just view it as a job then they should at that point go find another job but if it's you know of if they really feel like they're part of a team that's like trying to do something and they enjoy they enjoy the process then I think there's a chance of you of you making it so I think we were fortunate to have built a good team culture and a good Bond I think nearly all of the people that were there during that time are still at the company now but yeah I mean it's incredibly hard to not succeed right like to just keep trying to do something and like not and it not work is is incredibly difficult and I I think it's one of those things you know like Steve Jobs had that here's this little video quote where you have to really love what you're you're gonna do and the reason why is because it's so hard right and if you if you don't really love it and really enjoy it and really want to solve it it becomes impossible to continue because it's just so difficult at every level you know like every aspect of like trying to build a business from scratch is so difficult that if you don't really love it and your heart's really not in it it's impossible to like continue right it's impossible to just keep stay motivated to want to keep doing it and and it never goes away either like as the company gets bigger and bigger or more valuable like it just gets harder in a way because it's before if you fail it's like not a big deal almost right it's like sad and look everyone's out of a job but there wasn't you know no one was depending on it anyway right like it just didn't it didn't work once it gets big the pressure becomes really really different and just keeps increasing it at some level so you have to really I think really want to do it and that would be my biggest sort of caution to like new Founders is like it's not easy and I think the best businesses take a decade or more and if you're you know if you're not ready for that then you know don't bother I feel the same as I'm sure Imran feels the same as well as as Alliance becomes better and better I we feel more pressure to succeed even more as we get bigger you get more as Founders you have more responsibility so things actually do get harder over time not not easier so definitely feel feel the same way yeah I mean they're different problems right like we don't have to worry about like can we make payroll right we've got you know we've got a bunch of cash in the bank and so like that problem goes away but other like every other problem is still there or worse you know than it than it was before and so that's you know I think everyone just should be extremely cognizant of like what they're going to get into before they really take it on I want to talk a little bit about talking design because tokens are in general a very difficult Topic in crypto but specifically for proof of physical work it's such a critical component of the network because there's no way you can bootstrap the the network effect without the token so you launched the token 2018 and it's been almost five years now have there been points where you realize that original token design didn't work and you had to Pivot I'll have your you know thinking evolved over time regarding your own token design we had lots of different iterations like that the first token design was very much like filecoin right which was very much like Bitcoin right where there is a here's a single token and there's an emission schedule and there's havings you know every few years and so the scarcity built into the design and it's from heavily front loaded which is how you sort of start the Flywheel spinning I still think that was a Fine Design and that was what we were going to go with but then I think having spent some time talking with the guys from multi-coin before we launched you know it's a multi-coin participated in the the financing round we did just before the network launched and spending time with like Kyle and Tushar and the team there at the time they convinced me that there was a better way to do it which was to have this burn and mint equilibrium design right which is which is kind of what helium ended up with where there's two tokens there's one token that everyone kind of earns and another token that you use to to use the network and the the second token is kind of like a stable coin and the reason that was such a smart design was that we knew that the potentially biggest customers of use of the network were going to be companies like big corporations that would never hold like a volatile crypto asset as as like something that they would use to pay for a network like it would be ridiculous if one day it cost you know 10 cents to use the network the next day it was like eight dollars right just because the market had moved around the burn and mint equilibrium model was a very clever design for how to solve for that problem where to use the network you basically use a stable coin and how many of the stable coin you get by converting one you know coin into the other is variable uh that way you know users of the network have always the realization of the expectation or the understanding that it's going to cost a fixed amount and that was that was critical so that was the Mage that was the biggest pivot before launch was the change to this burn and mint equilibrium model and then after we launched lots of other things changed initially for example we had gone for this model where there were only 50 000 tokens created every month and then one thing that we very quickly realize which I'm sure lots of people already understand is that people like large numbers especially when there isn't a value ascribed which is kind of the way it is at the start it's more fun to earn a hundred thousand of something than like point two right and point two works for Bitcoin because Bitcoin you know is worth 60 000 a piece or whatever at the time but it doesn't work for anything else in in my opinion right and so that was uh one of the challenges so like early on there was this sort of governance process where it switched from 50 000 a month to five million a month so that was one of the changes was just like observing the fact that people liked large numbers right and that was again potentially kind of obvious how did you get that idea like where did You observe that people I mean it's obvious in hindsight but was it like you know just spending time on Twitter and see what people said about helium or other tokens how did that realization came about yeah at that time the network was like small enough that you could talk to all the participants on slack right so we had this like public slack server and channel and all the network participants were in there right it was like again it was very small at this time like hundreds of people it was very easy to get feedback in fact most of the time like we would get the feedback whether we wanted it or not like people people would just you know tell us what they were thinking and that they were mad about something and the thing that I think was most maddening for people and probably still is today actually the sort of the biggest complaint is that like as the network gets bigger everyone earns less right it's kind of like Bitcoin getting harder to mine like you can no longer just fire up a laptop and you know mine millions of Bitcoin like Satoshi did and so that was the hardest thing is that the early participants were like mining thousands of h t a day in a lot of cases like there were single hot spots that mined you know hundreds of thousands of hnt just a single hotspot there was this pair of brothers that mined like five million ATT and then as the network started to get bigger like that kept dropping off that was the most sort of like drastic thing you know so we kept getting feedback all the time from people about stuff like that and it just sort of became obvious at some point that it was going to get really really difficult for people to stay excited if they were dealing in fractions right and and it just it's harder to come to comprehend like you know I earned 0.017 today right and it's like it's just not as clean sounding or as easy to like think about as I earned 170 and so you know that was where that came from that was one of the early changes and then the next thing that changed was to add a having schedule when we switched to the burn and mint equilibrium model we did away with the having and the initial reaction to that when I think helium started to get popular Market participants were the most vocal about this being a problem right they were like if it's sort of like this Unlimited Supply and the only way the supply goes down is if like lots and lots of sensors use the network it's going to take too long in the burn and mint equilibrium model like you destroy one token to end up with another so as the amount of usage increases in theory you're destroying h t and so the way eth today has become sort of deflationary that was sort of the design right and the problem that I think everyone realized quickly was that it's going to take a long time before there's like sufficient sensor usage to really put a dent in that equilibrium and so we introduced the having or actually the community introduced the having model again where you know every two years the amount of ATT that gets created would cut in half until eventually it was it was zero to just just like Bitcoin so that was kind of the evolution and then recently there's another one with you know multiple tokens but I think that was those were probably the most significant changes in the early stages follow-up question to that is if I am a Founder looking to build a proof of physical workspace and I'm working on token design what are some of the things that you would tell them to avoid it's so difficult to say because you know because one of the things that worked really well for us for example by accident was the fact that there was scarcity of Hardware like this happened because of covid and it wasn't designed at all right but the fact that like there were these people on the network earning a ton of ATT and this whole group of other people like on the outside looking in Wishing there's like we did not design that at all right like that was a disaster as far as we were concerned because it was you know everything was like six to 12 months behind because of because of covert so I think you know there's something about that Supply demand balance that is is very difficult to get right because you need to sufficiently incentivize the early users otherwise nothing happens but then you also have to like be cognizant of the fact that if you succeed you're going to have all these people join in and it has to be interesting for them too and so the demand side of the network has to pick up much faster in the proof of physical work models I think in order for it to be to be successful the other thing is that having the sort of continuous emissions model is difficult in a different way there's always selling pressure right like if you look at some of these networks that are out there today like all the tokens are in circulation already right and that they just are sort of there but with models like like Bitcoin and like helium like new coins are created all the time and that creates a different kind of Market Force you know where there are constantly sellers effectively right and so that's just the thing to be sort of again to be cognizant of that is ideally you you design your token model such that there's organic demand for the service that sort of matches emissions to some degree and I think that's incredibly hard to figure out and do and I think Hive mapper has kind of an interesting model that they've gone after where the emissions are tied to like the amount of mapping that gets done on the network and I think that's directionally I think is probably the right way to head rather than just a fixed or time-based schedule because that doesn't account for like how big the network is how much demand there is you know et cetera Etc so I I guess one of the key things you pointed out is that you know Supply demand is very hard to calibrate beforehand things will change and there will be exogenous factors like covid and so it's almost inevitable that you're going to have to change the token Supply or demand a little bit in production right and you did that by uh adding this uh having this change how many months after the initial launch did this happen I don't remember honestly I think the hip that introduced it was so helium has this like helium Improvement proposal model that we copied from from Bitcoin and ethereum and it was hip 20 I think that introduced the model of a fixed Supply right so that and I I forget who it was in the community who introduced it by the way that's perhaps the most important point is that like having a community with a governance process even if it's imperfect is incredibly valuable right because a lot of the best proposals and changes in helium have come from the community and so giving them a way to like participate and make serious changes is incredibly important and it's not going to be perfect and some people are going to be complained that it should be you know they're going to complain that it should be weighted by this or it should be weighted by that or it's unfair and you're never going to get it right but having some process where the community is empowered to make changes I think is incredibly important and I think we've we as the helium Community have been fortunate to have participants that really put a lot of thought into like how to improve things and hip 20 was one of those and early to mid 2020 would be my guess as to when as to when that came out I forget after like more than a year of observing data from production getting feedback from actual users was this token design change made exactly yeah and it was again it came from the community for for like observing everything going on and those honestly have always been the best proposals of the ones that came from the community who are actually like observing things that were were happening rather than just here's our point of view on on how it should work shifting gears a bit on the technology and infrastructure side you mentioned Laura Wan which is long-range wide area network right yep and is that something that you developed your team no so so Laura Wan is this open standard that came to be probably in like I don't know 2014 or 2015 or something or something like that and so yeah we or the helium network has keep trying to remember that we're not helium anymore I have like came to adopt you know an open standard which I think has come with a lot of of benefits and a lot of value like there's a lot of sensors and there's a lot of you know a lot of stuff out there that already use lorawan so you don't have to build you know helium-specific things it just has to be lorawan you know compatible got it so you had this and then you have something called the open LNS maybe we could talk a bit about that but that ultimately enabled devices that is on the helium Network to only authenticate once on the network and then they can use it across any hot spot is that correct uh sort of so like if you think of okay like the networks work like this so you have a sensor and it transmits some data a hotspot receives that transmission and then what the hotspot does is take the transmission and then deliver it to a server that's roughly the flow so those are the three components is a sensor hotspot and a server and one of the things that we had learned over the years of being in the iot space and dealing with Enterprises and and larger customers was that they ultimately wanted control of that data right even though it was encrypted and blah blah blah like the fact that it went to like a server that wasn't there in a lot of applications was a serious problem right and it doesn't necessarily mean it has to be on their premise but it has to be something they control so their own AWS cloud or like some something right and so one of the things that we thought was important about designing helium was that there was multi-tenancy on the server side so that a sensor could belong to a server and the hotspot would know which server the sensor belonged to right so only like almost like DNS built into the network so rather than having to be delivered to like a server that we control like anyone could run one of these servers and register it on the network and then every hotspot would know okay like okay I see a I see a transmission from sensor a I'm going to query the blockchain and I know that you know sensor a belongs to like server B and I'm going to deliver it to server B so we use the blockchain as sort of a DNS like the routing system basically which allowed Enterprises to run their own servers and control all of the data flow themselves other other than the hotspot open LNS is sort of like an extension of that which is in the Laura Wan ecosystem the server that I talked about is called an LNS a lorawan network server and there's a there's a universe of them like there are several different ones the chirp stack is one AWS has an LNS you know there's several different LNS providers and and products and so open LNS is is the idea and the implementation that you should be able to use any off-the-shelf LNS with helium like you shouldn't have to use like the helium-specific one that we built and it just you know it makes it easier for people to participate in healing because like some people already have existing workflows or they trust AWS more or they have the agreement with X Y and Z to use a specific LNS for some reason some of them have different features different functionalities so open LNS is basically like bring your own LNF and make it work with with helium and that's what that because up until then you know you had to use the LNS that we we had built which was open source and everything but it in some cases was lacking features or people had a preference to to use something else so that's that's what openlms is so you were very deliberate on many things it seems like from a tech and infrastructure perspective right when it's like maybe adopting open LNS right that's more developer friendly two is like you know open sourcing or Hardware design so that everyone can build your Hardware that can support the ecosystem and then you know having more of a kind of community governance approach to making decisions long term was this planned or did this just happen over time or like were you deliberate about this I'm just curious on how you thought about just like just opening everything up and becoming very open sourced and like supporting developers and things like that yeah I mean we thought it was critically important I think back then in 17 and 18 it was arguably more important than it is now you know like I think people at that time were I mean certainly I was and we were like over indexing on maybe not over indexing but certainly indexing on being open and decentralized as much as possible and so everything we ever did was it was open source right from the start and we took it to an extreme even where there were no validators and no servers running the network like the whole consensus protocol ran among hotspots which was kind of an amazing design but it was impossible to scale it like as the network got as big as it did like the hardware just wasn't powerful enough in the hot spots to do this but as far as I'm aware like that was the largest peer-to-peer Network that has ever been built of any kind and we had like the entire like blockchain consensus protocol like running on these like Raspberry Pi based devices but right in this like kind of almost ludicrous to think that we even tried that and it did work up until 50 000 hotspots or something and then it really started to struggle so yeah I mean we just thought that was really important like we we always wanted it to be that if we disappeared or died or something it would continue onwards and that nothing could ever stop it whether that was an asteroid hitting us or a regulatory thing or like who knows right like if something happened to the company that the network would still have a way to live like we thought that was the most critical like Baseline of like what what decentralization means and I still think that that's true like it it must be possible for the company the founding team to like exit and the network to continue to function like to me that's sort of the the Baseline stakes and just on that note how far do you think helium is from that ideal you know ultimate level of decentralization that you hope to achieve I think it's quite close now you know there's a independent Foundation there are committees inside the foundation that are all community members that make all the decisions now I think with every crypto project you wish there were more developers it's tough right there there are only so many developers that that are available to sort of work on open source things and I think every project suffers that way like Bitcoin I think is has a little bit of a crisis of this right where there are almost no core developers really left in bitcoin's case maybe that's okay because it's sort of you know ossified and this just sort of is what it is now and it doesn't change very much but I think all of these things otherwise are still are still in motion right they still have still evolving whether it's ethereum or Solana or helium they're still evolving and there's still improvements that can be made and work that can be done and so I think that's probably the next Frontier of things that good crypto projects gonna have to tackle is like how do you fund the developer ecosystem outside of the core team and that's you know that that's kind of critical and I I still don't I still don't think I've seen great models for that from from anyone including helium by the way you mentioned Solana so helium is moving to Solana either have or or soon but probably this month what was the um the thinking behind it because originally you were building a sort of your own chain right yeah so the the migration happens on April 18th that's the date that the foundation is put out we built our own chain initially not not out of any desire like I think it's a terrible idea to build your own chain quite frankly but at the time in 17 or 18 you kind of had to depend on ethereum and we we weren't sure how to do it on ethereum in a way that would be cost effective or that could work I'm still not sure how to do that but by the way even with the l2s and everything else not clear to me like how to do something like helium in a way that that doesn't incur insane transaction fees for the for the users so we had known the Solana team for a very long time Anatoly who's the the founder of Solana Labs had worked with our CTO at Qualcomm and so we we knew him well we actually tried to hire him before he started Solana and so we knew those guys and we were very close we still are very close and we had always stayed in touch and I think now you know it's gotten to the point where those platforms are so good you know like Solana Avalanche you know like all these things are like excellent compared to what was available in 2017 or 2018 and we've been spending so much time as the core team but also as a community just you know making sure that the blockchain works when really what the most important thing is is like is there a good wireless network and does the wireless network works and the you know the blockchain part is almost I don't want to say relevant it's obviously critically important in terms of how the economic part of the network functions and but whether it's something we built or whether it's something someone else build is less important and to me it feels a little bit like the cloud compute shift that that happened where some people were obsessed with like it's got to be on premise it's got to be bare metal it's got to be blah blah blah and then over time it just became easier and more efficient to just you know use AWS or use Google cloud or Azure or whatever and I kind of view l1s like that now like if you're building a decentralized project especially a physical world one the last thing you want to be doing is thinking about consensus protocols and peer-to-peer networking and you know it's it's a nightmare and if someone else is already doing that well and the economics work like I think they do on Solana then great use that if they don't then you've got to build it yourself but I think the the cases where you have to build it yourself are are probably very very small at this point so uh you probably looked at a bunch of options Solana ethereum l2s you said ethereum l2s are just not cheap enough compared to Solana which I agree with in your use case and he also said that rolling your own chain is a terrible idea but just to play Devil's Advocate I mean today the SDK is for rolling your own chain is probably a lot better than it was five years ago with like Cosmos or even some of the roll up as a service on ethereum have you looked into those options by any chance yeah I think if you have to do something yourself I I would do it on Cosmos you know like that I'm super impressed with what they with what they've built there's some parts of it that are still a challenge right so for example one of the things I love about Solana but also you know ethereum and others is the sort of budding like defy ecosystem that you have there like I I'm a massive supporter of the idea of like decentralized finance and decentralized exchanges and I think we're going to see those become critically important and if you do something on Cosmos you're still a little bit in your own Island you know there's you're at least on a familiar protocol but that works a little bit more clunkily than I hoped it would right and so you're still a little bit on your own Island I know there's IBC and there's all these you know there's interconnection between the islands but it doesn't you know it's not quite as seamless as it is if you're just sort of a an asset almost on Salon or ethereum or something so when I say like rolling your own chain I literally mean you know building it from scratch the way we did where there it wasn't a clone of anything it wasn't a fork of anything like we wrote every single line of that thing ourselves For Better or For Worse but if you do find yourself in that position I absolutely would advocate for using Cosmos or something like like it but I I personally still believe there is not enough upside in having to maintain that infrastructure at this point you want your network participants that receive the helium token to be able to use D5 immediately right like they want to be able to cash out for example and rolling your own chain on Cosmos would require the user to bridge to like osmosis which takes like I don't know seven seconds back and forth right which is a really painful experience I don't know if that changed but at least when I tried it like a few months ago that was the case but that was a very important experience yeah it's also you know helium introduced a lot of normies into into crypto right like a lot of people that were not D5 people or any kind of people there's like normal people and some of that is because I think it's a use case that people can understand and some of it is the way that we went to Market you know we're on Facebook and Instagram you know so you kind of got to bear that in mind like bridging like assets and stuff I mean no one has any idea what that means and like maybe you can you know wrap it up in a nice UI or user experience but it's a lot easier if the fewer steps there are the better and I think you know what we're going to see is in my opinion like just working with counterparties like centralized exchanges is just going to become riskier and riskier over time and there's going to be a few that I think do it well like I think I love what coinbase does for example and the fact that they they strive to to always be on the right side of Regulation but most don't right like most are just winging it in some unknown jurisdiction and if your funds disappear there's truly nothing you can do about it right and I think that is going to become less and less of a tenable model over time and so the I'm very bullish on defy and especially decentralized exchanges and the speed that they work at on Solana is just kind of like mind-blowing it doesn't even feel like you're using a decentralized network of any kind like it feels like a web 2 application and I think that's very special now uh helium has become what I would consider a network of networks I don't know if you've thought about it that way but Founders that are applying to our accelerator program are all building many are building proof of physical work and they're all thinking about integrate grading or using helium in some sort of fashion because obviously every device needs a internet access or Wi-Fi and so one is like was this the grand plan overall and then two is there's like four or five products that helium has and subsidiaries have now built right now you have the helium phone service you're thinking about decentralized CDN you're thinking about iot devices Etc so like maybe just like when you first started in crypto or when you first started as a Founder was this the Envision of all of what you wanted to build or is this like did this come throughout the Journey of learnings and then maybe talk about the end game here with uh helium yeah I think a lot of it has been part part of the journey I mean when we wrote the white paper we did talk in in the end there about the desire to like explore other protocols right like could you use this model for other things Beyond iot and part of it is that you didn't know if helium was going to work right like was this idea of people buying hot spots and sticking them in their house is that even going to be a thing you know like is that actually going to take off or not so we didn't invest much time into like exploring other Wireless protocols no no clue whether any of this was going to actually work or not the other part is it relates to like physical work that's like very challenging is especially for wireless networks is regulation and you know things like Spectrum access right so only after helium launched did it become possible to do open source unlicensed cellular right so there's a unlicensed frequency called cbrs in the United States that only became available after helium launch and then that you know the availability of that Spectrum then spurred a bunch of other stuff to happen you know like now every iPhone supports cbrs and you know there's a lot of things that have to come together for any of this to be viable right and not all of them were were available at the time especially in the cellular world and so then once the frequencies available then you have to figure out how to build a network core and that's an incredibly big and complicated piece of technology and no one was really Building open source Network cores because there was no unlicensed Spectrum to use so it was all owned by Nokia and Erickson and Huawei and guys like that and so there's like a sort of evolution that's happening alongside helium of you know sort of Open Source networking so I think it makes sense for helium to constantly be looking at that and taking advantage of it when it when it makes sense and cellular is one of those places with the 5G Network that it feels completely obvious to me but at the time was not obvious right there was literally no way to do it and so I think that's going to constantly happen I think it will be an evolution and there will be other networking protocols that come along and will be other types of Technology whether it's decentralized cdns or vpns or like other stuff that makes sense to fit in this sort of community-built network of nodes model and I think helium should always be open in my opinion to supporting those and the question is like how to construct the economics but I think this network of Networks approach that helium is adopted is the right way and it's it's super exciting to see you know the 5G Network coming to life uh and then you know I'm excited to see the other networks too there'll be more I'm sure ciao any final questions that's it on my end I really appreciate all the insights uh it was a lot of fun thanks Amir yeah thank you so much Amir this was an awesome chat thank you guys for having me what a chat I've learned a lot you know just talking to Amir from Helium there's just a lot of learnings and maybe we can just like debrief on some of the learnings for for the founders and others that are listening what were some some of our top takeaways I have some stuff that uh that are wrote down some of them is idiosyncratic others are applicable to pretty much all the founders in crypto but the idiosyncratic ones is this is really interesting he said the hardware business is hard I know I know it's hard we all know it's hard but I didn't know why it was hard but he said that you know uh it's hard because the iterations uh iteration Cycles are slow and that makes a lot of sense right and just like designing the thing right like designing the hardware getting Hardware manufacturers Gathering the resources it's just insane how long it takes I remember having this conversation with a friend of mine who is um a really good engineer and he told me this about the the history of engineer and software engineer back in the day like in the 1990s in the 1990s software engineering was also very difficult very lengthy iteration Cycles were a lot longer because back in the day you had to you know write something write a few software and then put it onto a freaking fizzle CD and then ship it to the customer right whereas today you write some piece of code and then you push it to the user through the internet so you can iterate a lot faster today but anyway just a really interesting piece of history but I also brought up the fact that web3 or crypto you know software development is uh very analogous to that because you need to uh spend a lot of time and if you don't you're gonna get hacked like if you make an upgrade without the audit you're gonna get but anyway so that's uh just the interesting bit of uh lesson from from Amir the other um just came to mind uh was uh he mentioned that vinod khosla who's like a probably one of one of the best investors in this in the space he said there's always an element of love you need to survive long enough to get lucky yeah yes there that's the exact comment I was looking for and I feel like that's true with everything right like so as long as you're you know focused on the journey you stay lean and you stay focused uh like reasonable in terms of your Runway you could last you could Outlast pretty much anyone so as long as you have the right you know the product or iterating over time to get to that product and maybe the Market's ready for it to take off but I've noticed that luck is important especially in the case of Amir and helium like him didn't really take off until crypto came in right until 2018. yeah and yeah even if you wanted to do crypto in 2013 right like he started helium 2012. even if he wanted to do crypto in 2013 it would not have been possible because he ethereum did not even exist and even if ethereum existed the dev tooling was not good enough until maybe 2018 but even then he launched his own chain which he then said it was a totally bad idea and so it was not until slana that which was born in uh 2020 that the current version of helium is possible so if you don't survive long enough you don't even have the enabling Technologies from the outside world to help you become successful yeah even if you think about like what's happening on those layer ones right like with D5 and and all the other products that are being built where helium the users and the helium the token will have more utility over over the long run I also think that's like probably a larger element of why he moved from like his own chain to uh to Solana the other thing he said unrelated to this was that his or initial product idea came from his friends oh this was yeah yeah because I think the vast majority of startups that we like at least are either ones that build products for themselves yeah or products that you know product ideas that that just came up organically by observing the world like you know by observing like Twitter like for example ribbon right like uh Julian from ribbon gave a talk the other day to our Founders and he said that his current product which is the options product came from just spending 24 7 on Twitter and observing what DJs think what products DJs want in the case of Amir it was an idea that came from his friends but what's also very interesting he mentions later on was uh through like governance proposals like some of the best governance proposals came from the people the community which are his users which are the people that have you know actively been mining hnt as an example and so a part of like I guess you know building your product especially in crypto is the element of like crowdsourcing through the community that you're building over time and because some of the best ideas do come out from from the people you talk to from your users Twitter and then the community that's governing your protocol I think there's a little bit more even more Nuance to those which is that what I've seen is that the best product ideas are always almost always are organic they're organic I've never almost never seen people succeed by building a product just for the sake of building it the idea always grew organically within the person themselves or their friends or observing observing Twitter Etc so the original idea is very organic but later on the specific features like minor upgrades here and there they tend to come from the community also known as the users like the our industry uses the word Community a lot but really the community is is the users right I mean it's the best feedback loop in the world I mean people that are building in in the web 2 space they don't have this interaction that we do in in the crypto space in regards to our users sometimes it takes them weeks months years to figure out exactly what's like stuff they don't use Twitter or nothing yeah yeah you have to fill out a form that's managed by someone in let's say India or wherever and they may not and they may not understand how critical these problems are if especially if it becomes a pattern right while on Twitter where you just get memed or like called out on and and maybe that's important right I think having an open discourse to us to a certain degree is good he also says something about marketing which uh really struck me so when I asked them uh did you do anything specific before the launch of the tokens you said they did a lot of Facebook ads and Instagram ads which is very interesting because yeah I almost never recommend as to to our Founders definitely not the crypto native products like if you build a product for crypto natives there's no chance that ads will work there's far better channels which is Twitter Discord Etc right but uh helium is a different story because helium's Target user segment is the mainstream is the normies how do you reach those people yeah and it appears that ads do work in that scenario and it's something I will start recommending to Founders in the proof of physical workspace Hive mapper who recently launched I'd say like three months ago they pushed you know as heavily through Instagram and Facebook I just remember scrolling into Instagram and and others and I would see an ad for hive mapper this was very early on and over time I'm like okay let me just check it out and I checked it out and then I learned all about hype mapper and what they're doing and Etc so I do think there is an element of of importance there even though you know we as a crypto native community we look down upon like you know using traditional ad networks as a way to get distribution and and Community the other thing I wrote down is your your camera's not on this will never happen in Chicago come on man move back you got to move back man I want to get the hell out of here too um all right dude the other thing is um the two token design ah the uh hnt and the data credits yes smart yes so uh basically in helium you have you have two tokens you have a token that is paid for the network contributors and then there's a token that people use to to pay in order to use the network and the second token is actually a stable point and the insight there was that uh which we talked about this in our La in the other episode that we recorded on physical work which is that in a lot of these networks the main users of the network are actually Enterprise clients and there's just no way they can deal with the volatility of a non-stable asset I mean they wouldn't buy a token to begin with um yeah so that is definitely something that I would say pretty much all uh Pro physical Network networks should do I mean even in the case of file coin I feel like it's the same story yeah and our weave Etc what's interesting is that when you buy this data credit it automatically burns the agent and so it takes away Supply from hnt which is basically like using hnt to pay for the network but instead of doing that the user pays in Fiat or or stablecoin and then the stable coin applies hnt which ultimately is used to pay for the network so basically the stable client basically creates an extra step in the mechanism yeah in order to uh improve the user experience for the end user you know what's really cool he said it towards the end about this is that he's building a network or like helium is building a network of networks right so you have helium as the infrastructure for the to Broadcast internet there's going to be other services on top so you have the uh the hot spots with 4G hotspots then now you have a cellular service right and so the cellular service will have its own like token it's called the mobile token and whether you buy if a subscription or use the the cellular network the mobile token would then be used to burn for hnt as well so it's like all the value that that helium is building on top is all going to accrue back to the hnt token and speaking of that the network of networks so helium now has a multi-token kind of design but the interesting bit there is that which we speculated for for a long time is that for proof of physical Network projects or any products with a token it's extremely hard to get the token design right on the first attempt because it's extremely hard to calibrate the supply and demand before the token launch without seeing data from production especially given some of the exogenous factors like covid in the case of helium that really disrupted the supply of Hardware the lesson there is really you're gonna have to Pivot in your token design in production and they did this at least twice like two major pivots the first time was when they introduced the uh you know the having model which uh happened a year after the initial launch so a whole year of observing data from production gave them the confidence to make the the pivot it's pretty interesting because like all the things that are happening within D5 like if you remember the our D5 episode we talked about the Curve Model with the ve CRV where if you lock up your tokens for a year you may get additional Emissions on top or incentives on top and then if you do for two years three years four years then that gives you more governance control some interesting design aspects of the token so I just found this to be very interesting was that hip 51 introduced ve hnt which ultimately allows you to do the same thing so if you lock up your tokens for a year you get voting power and you might get some aggregated voting power on top that gives you more influence in how you govern the protocol so I thought that was also very interesting because I mean although Amir has been the space for like let's say 10 years or whatever he's been started with this like iot like device now he's like getting more and more into like what's happening with D5 and introducing like D5 elements into its own token I thought that was a very it's crazy that Amir has been building this for 10 years because some of the craziest stores in crypto of or success stores is you know people who who got successful within a couple years right like there's so many stories like that but the lesson is not all of them are like that helium is is a 10-year success story in the making and even today I would now say helium has reached its full potential yet there's a lot more things they can do right yeah oh the Solana Solana decision ah yes this is a good one but but the fees uh fees are much lower on Solana Yeah by an order of magnet one or two orders magnitude on Solana then uh ethereum Layer Two for a product that has the normies as the main user segment this is a big deal and especially for a product where the user earns some money by doing something this is a big deal because if you're gonna earn some money and you're gonna pay like three dollars in fees back to the network it just sucks it reduces the financial incentives for the user right so ethereum layer twos are actually not good enough for helium I mean obviously there's also an element of him being friends with Adam Toby but certainly the fees are the biggest Factor here I mean even if you think about it like okay let's say that he also said Cosmos right was an element that that he would he could be open to you still have to like recruit your own validators there's still like work that you have to do and I think the idea of him the analogy that he gave about AWS as you mentioned makes a lot of sense like I think future Founders that want to come into space that's just like back in infrastructure they don't want to worry about that they just want to build their startup and figure out the problem that the difference between AWS Google and those crypto networks is that AWS Google they don't have Network effects in crypto you do wait what do you mean AWS I'll give you a comfort example like a network effect I mean not inserted Network effect but composability interoperability with D5 specifically ah yeah because yeah in the case of helium if you launch uh helium on Cosmos the user is going to have to bridge back and forth to a defy Hub like osmosis and the bridging will take seven seconds also a lot of that you know Gap friction does not exist yeah so like building on like layer ones like monolithic chains there is a market for this right and you know like with the you know helium building on on top of Solana now you have Hive mapper building on Solana I almost feel like there's you know Solana is becoming a home for proof of physical work and that in itself could have its own network effects right in regards to other Founders that want to build and get access to these resources and access to these like products that are being built people are still bearish now it's unbelievable but I think they're wrong yeah yeah I do too that's why the price is where it's at right and so the ones that see the opportunity will will seek it great well I think that's all from from us it was a great chat with Amir have learned a lot and if you're a Founder building a proof of physical work startup apply to our program at aligns.xyz otherwise we'll catch you later see you later thanks for listening to good game don't forget to subscribe we'll see you next week