BuildrSpace
Find out what crypto native builders are building next.
BuildrSpace
#33.Digital surveillance and how to defend against it.
Harry Halpin CEO of NYM talks about how mixnets can help individuals overcome digital surveillance.
Prepare to embark on a riveting exploration of digital privacy with Harry Halpin, CEO and co-founder of NIM Technologies, who joins us to unravel the complex world of internet surveillance and the breakthrough technologies at the forefront of the privacy battle. Harry's transformation from a climate activist under the magnifying glass of government scrutiny to a pioneer in privacy technology is a testament to the urgent need for secure digital communication. As we venture through the episode, we discuss the innovative decentralized mixnet designed by NIM Technologies to encrypt and randomly release data, offering a robust shield against prying eyes. Our conversation also extends to the integration of this mixnet with blockchain, promising a new realm of anonymity for users across diverse platforms.
Get Tangem Hardware Wallet:
Use discount code "BUIDL" to get 10% off!! on your online order.
https://go.tangem.com/Tangem_cards
Tangem Wallet a swiss based hardware wallet company . Learn more
Harry Halpin: 0:00
I've always been very interested in privacy, Though I worked around issues on climate change when I was living in Scotland. While I was doing so, I became a target of surveillance Police surveillance on me, people trying to read my phone, trying to read my email all because I basically thought climate change was real induction music.
Host: 0:47
Before starting today's show, I want to mention that we have partnered with Tangem Wallet, a Swiss-based hardware wallet company. All have heard the phrase not your keys, not your coins. It wasn't until FTX and other exchanges collapsed that we understood how important self-custody is for your crypto. That's where hardware wallets come in. Tangem is an amazing hardware wallet that makes storing your crypto super easy with a completely new and secure approach. Tangem is a credit card style wallet with a chip that works with your mobile device. The amazing thing is that the private keys generated on the chip itself, tangem app itself, doesn't store any assets and everything is on the hardware wallet and it kind of interacts with the app via NFC, which is the same tech as Apple Pay. There are independent audits of the firmware that is conducted by the Swiss auditing companies. If you are interested in grabbing one, you can use my discount code that I will mention in my details of this podcast. Now to the show. Hello guys, welcome to Biddle Crypto. Today I have Harry Halpin. He's the CEO and co-founder of NIM technologies. Nim technologies is startup building, decentralized mix net to end mass surveillance. Prior to finding a NIM, he was a senior researcher at MIT, where he led the standardization of web cryptography API across all browsers. So welcome Harry, welcome to the show. Harry Halpin: 2:45
Thanks for having me Turen. I'd love to hear more about what you're thinking and what people are thinking about privacy in general.
Host: 2:52
Yeah before we get there. Harry, I wanted to ask you how did you get into privacy in general, into this arena? What inspired you? What life experiences got you into this?
Harry Halpin: 3:06
Yeah, I've always been very interested in privacy. I was always the kind of person that tried to use weird things like PGP on their email. And then what happened? That got me? I did my PhD, actually in artificial intelligence, in particular looking at things which eventually now we call large language models. So these are kind of the foundations of chat, gpt and other apps like that. Now that seems very distant from privacy, but it's kind of not, because what I realized when studying artificial intelligence is that a very little bit of data or metadata can be used to discover or infer, let's say, lots of data about someone. So by looking at, for example, your friends, I can detect what you might buy. If I'm looking at your patterns of communication, I can tell where you live. I can really predict your behavior I mean, I'd say, 80% accuracy by just looking at, for example, social network data. Now, this is something that concerned me and it wasn't like a personal concern until I became involved in climate change activism. So I worked around issues on climate change when I was living in Scotland and while I was doing so I became a target of surveillance. I had kind of police surveillance on me, people trying to read my phone, trying to read my email, all because I basically thought climate change was real and, yeah, that was really disturbing. We eventually beat these kind of British police officers in court. But having that experience made me realize how easy it is for both cyber criminals, unjust governments, organizations like Google or Microsoft or Facebook how easy it is for them to use data against people. And so when I was looking at possible solutions to this problem because I'm a researcher I was a researcher at MIT and other people I knew activists in Syria and other places who have even worse problems it was really life or death. What became clear is that VPNs were not enough, that a VPN a virtual private network which just hid your IP address but really didn't defend you against the kinds of surveillance that I was targeted with or the kinds of surveillance we saw other people target at us VPNs would still kind of reveal a lot about a person, could be used to track the person and weren't very resistant to things like internet shutdowns, which I know we have a lot of in some countries. So I began researching privacy and it's been full on ever since.
Host: 6:18
My next question to you actually is so yeah, that's quite an experience privacy and how you got there, how you actually focused on this area. Can you expand a little bit on what you're currently doing with the NIM platform Very unique name, by the way. Harry
Halpin: 6:41
Yes, you know, when I left MIT and Enria part of the French government I was working at we decided to build a platform that could actually resist actual censorship and surveillance. And at the time, my friends folks I knew, like Adam Back from Blockstream, who invented proof of work, or Gav Wood from Polkadot, so they were all friends of mine and they were basically raising money to work in cryptocurrency I said, well, we should raise money to work on privacy. So we raised, you know, a grand total of probably about 15 million and we decided to build what's called a mix net. So a mix net is something we've been working on for five years prior and a mix net what it does is it mixes packets together, so it takes packets, like when I communicate to you, when I'm talking to you, for example, in Zoom, my computer all the way over here currently in France is sending you packets and you're sending me back packets and Zoom sees the packets going back and forth and basically knows to help someone in you know, NDS talking to someone in France and they more or less know what neighborhood we're at. They could maybe identify us. And so, the way, if you think of these packets as, for example, like a card and a deck of cards, what a mix net does is it mixes the packets. That's why it's called a mix net. They were invented by David Chom, a very famous genius scientist and like, I think, 79, but they never really had implemented to scale. And so what we said is we said we can take these mix nets, they bring the packets in and, rather, and they just mix them up and they get released a different order than they came in, and they're also encrypted and they're made the same size, and it's kind of shuffling, just like shuffling a deck of cards and having the packets come in and out in different orders. When combined with the power of blockchain and decentralized network, where different nodes are mixing the packets differently, this really creates something which even an agency as powerful as the NSA can't easily track. And if it's powerful enough to be the NSA, it can also be, you know, the Russian government, it can be most likely any cyber criminals and you know it can be almost anyone, and it makes it very hard to see huge communicating with who, and that, we think is a real breakthrough. When we founded the company, we called it NIM, and NIM is from this. You may have heard of NIM. It comes from a non-NIM-ness or pseudo-NIM fake name, because NIM means name in ancient Greek, and so we said, well, that's a cool word. Called email servers that did things like mixed nets were called NIM servers. So we wanted to keep that kind of cypherpunk traditional live and we named our project NIM.
Host: 9:39
Thanks for that. So mixed net, you know, I remember Chomian sort of mixed net, so it's kind of an extension of that, and so basically what it's doing, from what I understand, is it's scrambling the traffic and so it's basically hiding the fact that how an individual is using who the person is talking to, like communicating with, and so it kind of mixes it up, shuffles it up, right, and that way, yeah.
Harry Halpin: 10:12
It's a little bit different, actually, than a Chomian mixed net although we love David Chom where what's called a continuous time mix net. So what this means is you send a packet into the mix and it gets released at a random time. That's sort of like shuffling the packets when you send a lot of packets in, they get released randomly at different times. That's based on a design called Loupix, which is continuous time or sometimes called stop-and-go mix net, and these mix nets are sort of interesting, because the problem is what if it never? What if your packet never comes back? Right, how do you lose your packets? So what we do is we basically create something like a send acknowledgement packet. So if your packet gets to whoever's supposed to receive it, they send you back an acknowledgement which is also mixed up. But if you don't get acknowledgement back, you just send your packet again. And this is really useful, particularly on blockchain systems, because blockchain systems are all public and they do appear-to-peer broadcast. And these peer-to-peer broadcasts are kind of old fashioned. They're nice, they were invented in the 90s. So I want to send something to say Bitcoins or, let's say, ethereum, and I basically talked to RPC, a relayer, and they send my packet out in a peer-to-peer broadcast. But the problem is that peer-to-peer broadcast again reveals who I'm talking to, when and where. So what we can do, which is really nice, is we can use the mix net to defend the user against the RPC provider or the relayer, so they don't know who they are. So let's say there's a big scandal where Consensus, who makes Infura, which I think a lot of people use via MetaMask, was gathering everyone's IP addresses and data. So we can talk to MetaMask via NIM by just changing the RPC provider and then, all of a sudden, you can still use Ethereum, you can still use MetaMask, but Consensus doesn't know exactly what you're doing, and that's great. We think that gives people more freedom. It's dangerous in a decentralized system that have a single company like Consensus know everything everyone's doing, and we can do the same in Cosmos. We can do the same with Polygon. We can do the same, and it's really important. While Polygon's been a good actor, polygon hasn't been recording people's IP addresses. For example, your IP address is a unique identifier that's given to your number every time you connect to the internet. While Polygon hasn't been copying IP addresses, zk-sync, which is another L2 on Ethereum which is about to come out, has been apparently recording everyone's IP addresses, so we think this is a really powerful technology whose time has come you mentioned that there is an acknowledgement that goes when there is that the packets have reached wherever they have to reach.
Host: 13:03
Is that done by I'm guessing, and I wanted to get into it the blockchain aspect of NIM, which is the NIM blockchain, so it is acknowledged from validator end. Is it randomized enough to such that even the validators cannot tell where sort of the traffic is coming from? Is that the kind of philosophy behind it? Harry Halpin: 13:30
Yeah, that was a great point, so I'll deal with them one at a time. So the first question is who's sending these acknowledgments back? So you are. If I send you a packet, the receiver sends the acknowledgement back, because only they know if they receive the packet or not. Right, if you receive the packet, you send it back and it goes back through the mixed net. So by sending your acknowledgement back through the mixed net after receiving a packet through the mixed net, everything gets mixed up and the receiver and the sender cannot be linked by what's called a third party. A third party is someone who has like a God-like view of the network. Who's looking at it? We can imagine from above, looking at all the packets moving around all the servers on the internet. Now the question is second question is can the validators figure out who we are? And the answer is no. The validators are only used to make sure that people pay for the mixed net and to maintain the reputation of the mixed nodes. Right, because we're a decentralized system. So someone puts on a node, you can run mixed nodes today and you get rewarded in them. You know when you turn on your node I'm writing a mixed node, let's say on my Linux server or my Windows server. What if I turn my node off? What if I start dropping packets? And if I'm sending you packets and I'm offline? That's bad, because all those packets don't get mixed and they just disappear. So you have to send them again. So what we do is we kind of maintain the statistics if who's offline and who's online and if they're mixing properly, and those statistics then get reported to a smart contract and that smart contract rewards the nodes and NIM, kind of like how Bitcoin miners are rewarded and what's called the Coinbase reward. They're rewarded for mining Bitcoin, but rather than getting rewarded for mining, you're rewarded for mixing. So you mix the packets up and there's a kind of way that the statistics can run that makes. That keeps everyone anonymous, but develops what we call proof of mixing technique, and that's a little bit complicated but effectively we can tell if someone is mixing or not. Then higher network can tell and they get rewarded appropriately. So if someone's doing a very bad job, over time they'll mix less and less packets and get kicked out of the network. But someone that's doing a great job will get more and more. And we also allow because it's not just mixing like, there's also reputation and so we allow delegated staking right. So if I have let's say I have a good reputation. Let's say I'm a famous human rights activist or something and I run a server, people might trust me more than they trust, like some guy no one's ever met. So you can delegate to these people and then they make more rewards also the more people that delegate to them.
Host: 16:30
You mentioned the delegated proof of stake, so the consensus is proof of stake based, right, I imagine. I think I was when I was researching. You're on Cosmos, right? If I remember, you're on Cosmos ecosystem. Harry Halpin: 16:44
Yeah. So the way you wanna think about it is, nim has two components. One is the mix nodes, which run do all this mixing which actually kind of does the work that you need to keep your internet traffic private, and the second part is the validators. And these validators maintain the blockchain and the smart contracts that reward the mix nodes. And these validators are running delegated proof of stake on a really standard Cosmos chain. But the delegators, again, all they know is who's mixing packets and who's not. They don't know who's, for example, sending packets because they can't kind of break the mixing themselves. They just maintain the blockchain.
Host: 17:24
Can I ask you one thing? Why did you pick proof of stake, sort of consensus mechanisms, rather than say something like proof of work?
Harry Halpin: 17:34
Well, I think it just depends yeah, it just depends on what the part of the system is doing. So the mix nodes, they're mixing packets. That's kind of a work right, so they are rewarded in a proof of work way. But the validators, their work is signing blocks, getting consensus on blocks, and we also make them do some zero knowledge work. We make them help develop zero knowledge proofs just to prove people have paid the network and aren't doing kind of civil attacks on that work. And that's via this new technology we developed called ZKNMs. We're big fans of zero knowledge tech base and we think it's kind of grow quite rapidly. And so the way you wanna think about it is that the validators their whole work is just signing blocks and making these zero knowledge proofs. So we said, yeah, they're just. Why not use proof of stake there? And we like Cosmos because we're not trying to be a general purpose blockchain right now. Maybe in the future we might be, but we're not trying to develop our system that. We're not trying to compete with Bitcoin, we're not trying to compete with Ethereum. We're trying to do one thing which is give people privacy, and we wanna get the best on the entire internet at that on that. We wanna give not just blockchain's privacy but any app privacy. We wanna be able to give what's app privacy, telegram privacy. We wanna give privacy to any app and we work with any app. We don't just work with blockchain apps. So we're not demanding people like run their smart contract on our system, but we are saying that before you use any blockchain, you should probably use something like a VPN, and we're like a super VPN and because of that, when choosing a blockchain, we needed a blockchain that could do two things. We needed a blockchain that was extensible or maybe has called now modular, and you can think of NIM as a modular networking stack. And we needed a system that had what we call finality, that could basically make the decisions on consensus and make them fast. And when we looked at all different systems, all the theorem L2s don't really let you customize the validator code very much. So we ended up choosing Cosmos. Cosmos, we thought, and we still do think, is a great ecosystem. Delegate proof of stake is not perfect, but it's good enough, particularly because you know we're not. Our goal is not to make the world's most scalable blockchain. Our goal is deliver privacy. We want something fast and with finality and you know Cosmos has lots of great features like WebAssembly support and our systems all built into Rust and that's super useful and you know we thought it was just the best system and we still think it's the best system. We looked at optimism, we've looked at other L2s, we even ran as a L2 on Bitcoin, actually using liquid for about a year. But push comes to shove. We think Cosmos is the best ecosystem, at least for us, and we've run it more or less as Cosmos intended. It's an we're an app chain. Our blockchain is an app chain for the Mixnet, so people can access the internet freely and privately.
Host: 20:56
Makes sense. You know, I actually had interviewed a Cosh network. I don't know if you're aware of them. They do compute decentralized on Cosmos. So they are one of my favorite projects and they are on Cosmos, so they are pretty good stuff what they're doing there.
Harry Halpin: 21:14
Oh yeah, we like the Cosh guys and other people on the I just came back from Cosmoverse Istanbul and other people on the Cosmos ecosystem, like our friends who really said we should work together and make a NIM token Folks like Zachy Manayan and other people have been Aaron Buckman have been very supportive of NIM. The Interchain Foundation gave us a grant early on for some of the work and you know we're very similar to a Cosh and so far as that, even though what we do is very different, a Cosh is decentralized computing. We're both app chains. We're not. We're not trying to make a general purpose financial layer. There's already too many of those. We're trying to build a decentralized app and so we use Cosmos as an application chain. Coming back to the, I can tell you something funny. Still worried about proof of work. Do you know? Proof of work? The concept was invented because of MixNets. Host: 22:11
Oh, interesting. Yeah, I was in the notion that you know, the anonymous emailer service that was there was part of where Adam Beck came up with the sort of. I think he reused the. So the original idea might have definitely, you know, I think you're pointing to, because Adam Beck used the idea of proof of work in the emailer service, but it existed prior to him, I think founded by two guys, you know, and so you're completely correct the anonymous emailers were actually using early MixNet technology. Harry Halpin: 22:48
And it's because they were using early anonymous MixNet technology. In particular, they were using a system called MixMaster. Mixmaster was a MixNet invented by Lance Caltrell and Lynn Sassiman. Some people think Lynn Sassiman might have been that Kimmodo, because whoever was that Kimmodo knew something about anonymous real areas and proof of work.
Host: 23:11
And Lynn was. Yeah, lynn was Bram Cohen's roommate, so I have interviewed Bram and his stuff on Chia and so yeah, he was his roommate. So most likely.
Harry Halpin: 23:21
No, I mean I knew Lynn pretty well. I mean, unfortunately he died, but he was at a university called KU Luven when I knew him in Belgium, After, I think, he left Bram's roommate, and we still have professors in KU Luven working on MixNets. That's one of the things that Lynn worked on as a PhD student at KU Luven, a Catholic university in Belgium, and so they're still great researchers like Anis Bengurrat and Nim's chief scientist, Claudia Diaz, and a wonderful, actually Indian scientist, one of the best of his generation, Deb O'Doss, who are working on MixNets. So the cool thing is that at and back, the reason why he needed invent proof of work is because when you did an anonymous email or the bit that made it anonymous was actually a MixNet. It was a kind of primitive one, but it was a MixNet and you know, because you had to basically prevent people from using it to, like spam, attack everyone anonymously, so you didn't want a million Nigerian gold or bank number scams coming in your inbox. He couldn't stop. So, at and back basically used proof of work. I think he didn't invent it, but he was the first one to implement it and implemented it called hash cash on top of early anonymous emailers like MixMaster and so it's kind of interesting that Nakamoto must have probably used the MixNet to defend his email address. That's why no one knows who he is, because he used anonymous emailers and you know it was probably into the technology and we think that's kind of cool. And if people are interested in more, there's an interview with Bitcoin magazine about myself and out and back where we go deep into the history of I mean it's a bit geeky on proof of work and MixNets. But then you know again, when we say for blockchain, proof of stake is fine, but for MixNet we do prefer proof of work.
Host: 25:19
So is that there are two different consensus? I'm sorry if I'm yeah, there's two different things going on, so there's the validators.
Harry Halpin: 25:26
They maintain a blockchain. This blockchain has the reputation of what's called the mix nodes and they're mixing up the package with mix nodes. But these mix nodes aren't validators. They're mixing. They're mixing packets up so they get rewarded by proof of mixing algorithm because the mixing is actually work. It's a proof of work algorithm. So our system has two parts. Basically, you can run either, if you're interested in working with them, you can run a mix node those you want to have high uptime and be able to ship a lot of packets back and forth or you might want to run a validator. A validator doesn't need, maybe, such a big internet cable, so to speak, because the blockchain, but it needs a big hard disk because it has to store all the block state. So these are just two different types of machines. We gave them two different properties. The consensus is only maintained by Cosmos, but the reputation of the mix nodes is determined by their work, their work of mixing up the traffic. Host: 26:29
I'm going to switch gears and I was researching about NIM and I'm not sure. Is the launching of ZK NIM, which is, I guess, your anonymous credential system, is that? Has that already been launched? Could you elaborate a bit more about that? What it is?
Harry Halpin: 26:53
Yeah. So one of the problems you have with the mix that is what we just described. How do you prevent spam? How do you prevent stable attacks? I've heard people from overloading network. So we basically say, hey, people should pay for it, just like people pay for VPN. In fact, we plan at the end of the year we're going to offer monthly subscription to mix that VPN combo, which we think will be very exciting and hopefully people will pay for it. That can people can use it just like they click a button, use it like a VPN. We think that will be cool. But the problem is, how do you prevent people from using it for free without de-anonymizing them, without invading their privacy, right? So we don't want to know who's using the system. So what we do is, rather than ask for you to pay for the system using, for example, nim token or using cash, we say, yeah, pay for the system, but when you pay for it, you get a receipt like a voucher and that voucher is in zero knowledge. So the voucher says I paid for the system I can send. I paid for a month, I paid for a week, I paid for this many gigabytes or terabytes, whatever you paid for, we don't really care. We just get a zero knowledge proof that you paid. But we don't need to know who you are and we just need to know how much you paid. That's great and that's really solves all problems for us. What we build is a general purpose credential system. This general purpose system is anonymous, based on what's called anonymous credential technology. Anonymous credential technology is a kind of very, I would say, simple privacy system. Or any list of values like maybe I'm from India, maybe I'm over 18, maybe, can be encoded into an array and those values can be selectively disclosed, which means basically disclosed in zero knowledge. You can make proofs that you possess a key or possess a signature or you possess an age, you possess an ID card or whatever, but you can prove that on zero knowledge or you could just show people that data as well. So it's a very flexible zero knowledge scheme. We say I called ZK NEMs because we like the word NEM and NEM means pseudo NEM, so we thought it's like a pseudo NEM. That's currently. We launched it in March out south by Southwest with Chelsea Manning on Testnet and it's still on Testnet. We were hoping it was going to get done much earlier but, to be honest, zero knowledge. Technology is pretty complicated and it requires a lot of audits and a lot of work. So we have not put it on mainnet quite yet. We're hoping to put it in mainnet by end of the year or, worst case, early next year. But we at NEM there's different kinds of teams and we're not really the kind of team that wants to build stuff fast and just whatever ship, because we deal with privacy. When you deal with privacy and you build stuff fast and you just try to ship, you make mistakes. People's life can be endangered. So I had, for example, I had a friend, basil Safadi. He was a Syrian activist and he got imprisoned and killed because he was using privacy tools that had, for example, like Skype, that had a back door. So we're quite careful, we try to be careful. We're not perfect. There's probably still mistakes in our code somewhere. But we really are trying to ship the technology to mainnet when we believe it's ready. But it's on testnet now and people can play with it and we've described the system pretty thoroughly.
Host: 30:39
That is fantastic, that you know that's a fantastic thing. You said that privacy requires time and you know it's not like move the fast and break things or break democracy or break somebody's life in this case. So it's pretty, pretty powerful. So, on that topic, like I was wondering, you know, where do you see this future of privacy, this business of privacy, the demand for privacy and the products that you know you see in your roadmap coming out that will enable these sort of like, these new primitives for making privacy like a de facto on the internet?
Harry Halpin: 31:31
Oh, yeah, I mean, we think that privacy is big. Privacy might even be bigger than blockchain technology, right? So it's not. You know, first thing, that there has to be innovation and there's not going to be innovation from the big players, google and Microsoft. Maybe Apple's pro privacy, but most companies like Google and Microsoft and Facebook, it's not really in their best interest to build privacy enhancing technologies, but it is in the best interest of citizens. It's also, interestingly enough, in the best interest of most governments, because what people often forget is that governments like to spy on each other, right? So I'm sure, like, for example, india's spied on by the US. If India is going to side a trade deal with Russia or something, the US wants to know about it and stop it. You know, the UK probably spies on Thailand. I would assume China spies on Vietnam, so forth and so on. Everyone's spying on each other, and so, you know, when we started building NIM, the people that actually wanted to use it first were European governments, because they were kind of sick and tired of being spied on by the US. So there's a lot of interest in this technology and people say, oh, maybe it could be used by criminals. But that's what also people said about encryption. You know, in the 90s, when encryption was just barely working, people tried in the US tried to block it. The government tried to block its export. They would have prevented blockchain technology from taking off anywhere. And instead, you know, it was fought legally and we won. But people forget that without encryption, the internet would be fundamentally unsafe Without that little lock in your browser there would be. The lock in the browser basically means the connection between my computer and the server is encrypted using what's called TLS Transport Layer Security. And if, without that little lock in the browser, I couldn't buy things online, I couldn't use my credit card online, I couldn't talk to my bank online, I couldn't. You know, back in 2013 to 2014, very, maybe 10, 20% of the internet was encrypted. Now it's like 99% and we think it's going to be the same way with privacy. You know, because of encryption on the internet, you got companies like eBay, right. You got companies like PayPal, amazon, and then you know Alibaba, so forth and so on, and there's probably going to be a bunch of other new companies that are enabled by privacy. We don't know who they are yet, but we're pretty sure they're coming and that's one of the neat things, you know, are the VCs that backed us VCs like Binance Labs they wrote our first check Andres and Horowitz Polychain. They're also convinced that there's going to be a big wave of privacy and that's why many of them committed to this kind of $300 million VC ecosystem fund to build new what's called the NIM Innovation Fund, which we're launching shortly, to build new privacy enhanced apps. We think there's a huge ecosystem out there that can be built. Now, what are we working on now? What's our go-to-market plan? It's pretty simple. We think people will probably pay for this, just like they'll pay for a VPN. If I came up to you and said I have a VPN that's going to cost less, probably, than the VPN you currently pay for, that can't spy on you, that gives you better privacy, no-transcript and, to decentralize, better censorship resistance, so I want to access Netflix and the US. I get blocked in one IP address. I just switched to another. Would you be interested in that? Most people would. So we think that there's a really you know simple go-to-market plan and the VPN market's huge. Vpns are, like you know, one of every eight internet users have used. I'm sure many people in India and other countries all across Asia use VPNs, so we think that you know if we can get this working fast enough and stable enough. You know this use case could be one of the big major onboarding use cases for all of web three.
Host: 35:48
Yeah, vpn usage is huge. So definitely that's a part to adoption. I had one second. Let me see what questions I had. So you know. So I have interviewed some hardware privacy companies like Piorism, who deal with free software movement, who are part of free software and they have free, not the open source, but the more, like you know, richard Stallman level, sort of freedom from trackers and stuff. Any thoughts on the Mixnet Sort of also being a weapon against or a deterrent against, like any hardware backdoors that are that currently, you know, manufacturers perhaps are all, like you know, are asked to build and that sort of a thing, cause you know, as you mentioned, governments are always from all other countries are trying to build those backdoors into the hardware itself. Will the Mixnet kind of slow that kind of or deter it in some way? Yeah, so I mean we're big fans of free software and free hardware. Harry Halpin: 37:09
I, like Piorism, friends with Richard Stallman. Richard Stallman supports them, to my knowledge, and you know we think fundamentally having free, free software and privacy, people need to be able to have control of their own hardware and control their own software and, ideally, control the whole stack, because privacy isn't just about zero knowledge proof. So they're pretty cool and we use them. Privacy is what you would consider an holistic property of an entire system. This means that you know you need to be private at all layers, just like you need to be free at all layers. Like if, let's say, I'm running zero knowledge proofs on a blockchain but my network isn't private, that's going to be a mess, right? People can just de-anonymize me using the network. I actually talked to a fellow now working in consensus but previously working at the NSA, who said that he said, well, we don't care what you use on chain for privacy, because we can just track your computer and your IP address, so the Mixnet can help there. But the Mixnet isn't magic as well, right? So if you have a phone just a back door in it, or your hardware is a back door, but at the current moment the Mixnet doesn't help, we need you to run something on your phone like Grapheme or work, you know the latest Apple phone or whatever. That being said, we are very interested in free hardware and one of the things we're doing is we hired Chelsea Manning, so she's a famous whistleblower and Chelsea is working on secure and speed-optimized hardware. So the vision is that we would be able to provide at least the Mixnodes, this kind of optimized, fast hardware with safe key storage, and that this would just as people would buy GPUs do better Bitcoin mining people would buy this hardware to get more NIMM rewards for mixing more packets and be able to mix them more and more safe and for a lot of the big companies and big governments who are interested in using them. So, for example, I know it sounds crazy, but one of our advisors is a Google employee. He was also Ben Laurie. He was also an advisor to the WikiLeaks a really great guy and Ben Laurie basically said look, if Google is going to use a Mixnet, it will need to have secure hardware, and that's also the same Moxie who invented the signals are pretty good for it, and that's also Moxie's general opinion that you need to run things in secure hardware. But the problem is, secure hardware just takes time to build, it's expensive. So Chelsea Manning is doing the research now and while we don't have secure hardware now, we do think we'll have in the future, I would say in a year or two. That should help prevent those backdoors and these problems that you're concerned about.
Host: 40:02
Interesting. Yeah, I think we are almost on time. I just have two more questions. I know it's going, of course. I love talking Awesome, awesome. So, yeah, I wanted to make sure that I'm not going to.
Harry Halpin: 40:21
No, no, the Mixnet is near and near to my heart, so just feel free to ask away. Host: 40:25
Okay, so when there is mixing going on, is there any? So in Tor, the Tor users, the problem, one problem is the speed, the latency, when it makes multiple hops. How do you account for that in NIM?
Harry Halpin: 40:54
Yes, the thing is that with privacy complicated, so if you're going to use a zero knowledge proof, you're going to have to prove something as your knowledge or verify something as your knowledge. That's going to take some time. Likewise, with Tor or NIMB, you're going to send it your packet. It's through three computer servers and that's just going to take a little bit more time than sending it directly to another server. But that, we think, is ultimately people may overestimate the problem. So with NIMB, what we noticed is we could get the delays down to seconds. It's not minutes of delays. We can get the delays to be pretty small. And the great thing with being anonymous on NIMB is the more people that use the network, the more. If you want to be anonymous, the more people that use it. You have like a bigger crowd. You're anonymous and so we can speed the network up. More people use it. So, like most things, the more like a blockchain or people that use the network, like I try, more people using a blockchain slows down. There's more gas fees. It's like there's an auction for limited blocks, chain spaces, so the way to think about it. But with NIMB it scales more like a web server. The more people that use it. You just add more computers, you can maintain the same level of privacy, you can speed it up, and so we've been able to really get the speed to be pretty fast with a reasonable amount of mixing. And so the question is well, tor does not mix, so a Tor, when a packet comes into Tor, it's like a decentralized VPN. People that don't know. With Tor, the packet is coming first in, first out, so they should be faster than them. But then one thing we notice with Tor is because Tor doesn't reward the nodes. So I, for example, I love Tor, I use it every day. I run a Tor relay, but if my Tor relay goes down, my Tor server goes down, I don't even notice. Why do I not notice? Because I'm not being paid to run it. So, because I'm not being paid to run it, if it goes down, it's like, well, no skin off my back, so to speak. I mean, I turn it back on as soon as I notice, but I don't get instantly notified. Maybe I should set that up, but I don't. That being said, when we were running our test nets, when we had no incentives, people's online availability was about 60%. They were online about 60% of the time. The other time they were off or not performing very well. But when you turn on incentives and people got rewards in NIM token for running servers, their uptime went up to like 95% 98% because they didn't want to lose those rewards. And then what we've noticed is that this can use the same fundamental insight as Bitcoin but create a kind of mutually beneficial cycle where, because you get more rewards, the better your machine is and the more uptime you have. This really creates a really wonderful system where people are pushing more and more traffic. They want to push more and try to system so they'll invest in higher quality systems. They get paid so they can buy higher quality networking gear. They can even buy secure hardware or optimize hardware, and so this creates a mutually beneficial cycle where the more people that pay and get paid, the better, and so more people, for example, buy the VPN, the nodes get more rewards and they mix more traffic. So we think this can really help keep the network stable, available and as fast as possible.
Host: 44:54
Sounds good. Sounds good. Historically, countries like Southeast Asia, like I'll talk about India, the culture here never grew up with, say, sci-fi things like Neuromancer or all the other dystopian sci-fi genres which inspired cyber punks in the US and other countries, and so initially the privacy wasn't like a big issue to the people out here and that psychology is changing. But wanted to ask you, what would you say to developers in say, especially in India, that why should they develop for a platform that is privacy-protecting and is trying to enable make the world more privacy-aware and be de facto privacy in the world? What would you say to them?
Harry Halpin: 46:07
Well, I would say that maybe privacy isn't the right word, but what these technologies do inherently enhance is freedom. I can't access the internet when I want to access the internet If I'm afraid when I access the internet because someone's watching or maybe I'll get in trouble. I don't have real freedom, and we want people to have freedom to access the internet, freedom to gain collective knowledge, to participate in new capabilities and all of the wonderful things the internet enables. And ultimately, there is a long and really wonderful history, particularly in Southeast Asia India fighting against the colonialism of the British, vietnam fighting against the colonialism of the United States and the French, and, of course, other countries in Southeast Thailand fighting against other forms of colonialism. And so we think that, essentially, the struggle for freedom is super important. The struggle for freedom is maybe the basic struggle we're all engaged in, and we're very lucky as technologists that we can help the next generation of people get freedom, just as our parents or grandparents fought for our freedom. And so we consider this the technology we're building at NIMM of a larger anti-colonial struggle, that we consider that these technologies in places like India against the British, against the French and Americans, in Vietnam and so forth and so on, and that this is a long and proud anti-colonial struggle and we technologists can participate in this, just as our grandparents fought for our freedom. We can participate in this struggle by building technologies that enhance the freedom of future generations. Technology is now part of this larger political terrain we all live in, and so I would say that I wouldn't worry too much about what you call privacy technology or something else, but we do think that technologists have responsibility to build technologies that increase human freedom, and mixed nets, we think, are very much part of that, and we hope people in Southeast Asia recognize that. We have a pretty active Vietnamese community, a reasonably active Indian one where we're just taking off in Thailand, of course, lots of people in China and other countries, and we, you know we at NIMM we're not particularly interested in what's going on in Europe or what's going on even in the United States. You know these countries want to ban crypto. Fine, it's their problem, but we think the people, the vast majority of the world, lives in places like you know, southeast Asia and the wider global south, and this is the future of the planet. So we want to be part of freedom in the fullest sense of the word, and privacy is part of that.
Host: 49:16
So that where can we you know folks can find more about NIMM reach out to you or any of the folks?
Harry Halpin: 49:26
Yeah, and yeah, we have a blog website white paper, like I was saying. Yeah, if you want to learn more about NIMM, it's pretty easy. We have a website NIMMtechnet N-Y-M-T-E-C-H. Dot net, a telegram channel N-Y-M-C-H-A-N NIMMCHAN don't fall for all the fake telegram channels, just make sure you're in the real one and we have a Twitter account at NIMM project, n-y-m-p-r-o-j-e-c-t and tons. You know we have a privacy enhanced training course called Shipyard that people can participate in to learn more about privacy. We have grant programs. We have ambassador programs. People can earn NIMM token. We have all sorts of things going on and all those information is on the website and we kind of keep announcements on Twitter going pretty heavily. So we'd love for people to get involved. We do have communities, like I said early, in Vietnamese Chinese. We have an Indian community and we'd love to see people jump in, and we just started a Thai community as well. We'll be in Southeast Asia also, probably for DevCon. So, well you know, we hope to be more in India as soon as possible. I haven't. I used to speak at free software conferences there, but it's been a while since, like 2017 in Kerala. So I want to come back.
Host: 50:49
Awesome, Glad to hear, and you know I would love to have you again on the show sometime in the future and I'm looking forward to this project and really looking forward to you know, have this full success in the future. So thank you so much, Thanks for your time.
Harry Halpin: 51:08
Thanks so much for your time. You know thank you and take care.