I'm the cofounder and currently CEO of Nym Technologies, and our mission is to enable privacy for the entire planet.

We're creating a kind of next generation privacy infrastructure.

This is a mixnet, which was invented by David Chaum all the way back in 1979, and we're bringing it into production with some new designs and new theories.

We believe that our technology can indeed protect the patterns of communication for all internet traffic.

For those of you who don't know me, which I imagine is most of you, I got my PhD back at the University of Edinburgh.

I worked mostly at that point on machine learning, not cryptography or distributed systems, what we now call artificial intelligence. That's one of the things that helped me realise the power of surveillance and how dangerous it was, because with a very little amount of data, you can make vast predictions about people's future behaviour.

I did my PhD back in the early 2000s. Got my PhD, I think in 2011, or ‘12. I worked for the inventor of the actual web - web 1.0 - Tim Berners Lee, trying to fix data and add cryptography with the web crypto API to all browsers. We all have working crypto in our browsers now for JavaScript environments.

But I noticed pretty quickly that there was no real alternative to surveillance by Google or Microsoft or even Apple.

And therefore I got kind of excited when I saw Web3 happen, because web3 said: well, maybe there's an alternative business model for the internet that does not involve mass surveillance, which does involve a sort of sustainable economics, and which involves returning power to the people whose labour actually creates the value of the internet - ordinary people around the world.

So how can we make a new kind of internet where we are not simply the victims of surveillance?

I worked very heavily on cryptography with folks from Google and Apple and Microsoft, and I don't believe it's within their capabilities to fix these problems - because their business models, to some extent, involve the capture of your data, and surveilling you.

I’m not philosophically particularly adverse to advertising, I don't consider that the worst thing that can happen to someone.

But I do realise that this kind of mass data collection can be used for behaviour control, which is very politically useful, and this amount of data can also be given to governments to eliminate political dissent.

That really does worry me. My data was handed to the UK government due to my climate change activism when I was younger. And that resulted in, and still results in, severe harassment.

But what I get is very little compared to what friends of mine like Bassel Safadi, a Syrian activist received, who was leaking metadata via his communications and was captured and tortured to death in Syria.

So yeah, it's a really big deal and I'm glad we're all together here to try to solve it.

So there is something which most people don't realise anymore, which is that the internet is to a large extent broken.

The internet essentially reveals all of the data that you're communicating and who you're communicating to, by default.

What this means is that every time I turn on my computer - the entire planet, anyone who's at least interested - knows what I'm doing, if I'm talking to Instagram, or if I'm on YouTube, or if I'm on Zoom, or if I'm playing a video game, they kind of know what I'm doing.

If I'm sending an email or instant messaging, they can even determine, more or less, whom I'm communicating with.

Now, some of these problems are solved by encryption: so for example, on your browser there's a little lock button. If that lock button is on, and it should be on most of the time, that means the content between your computer and the web server on the other side of it is encrypted. And that's a good thing: that prevents someone from just reading your messages. But what is not encrypted, and most people not aware of this, is the website you are talking to - or the server you're talking to - or the other people talking to that server…

Who is talking to whom, that data is called metadata, data about data.

This leaks even at the network level. Even when you use encryption, which includes HTTPS, but also encrypted messaging apps like Signal or zero-knowledge base cryptocurrencies like Zcash or other private cryptocurrencies like Monero, this data is always leaked.

Now there are some techniques you can use to prevent that leakage. For example, you could use a VPN. I don't know how many people here use a VPN, but they're very common.

There's some misconceptions about VPNs. We'll go into those shortly. But they don't provide actual anonymity. The person that's running the VPN server knows exactly what you're doing.

They know where you're going and who you're talking to because your information is going to them. Centralised VPNs are to some extent dangerous. They may be useful for accessing Netflix overseas, but they just kind of move what we call the trust from your computer to someone else's computer.

And I don't know how much you trust them. Decentralised VPNs just do so in a decentralised way.

Now one step better than VPNs is Tor.

Tor is a kind of decentralised VPN.

And it's much more trustworthy than a VPN, but it has a problem, which is that it does not provide anonymity against adversaries that can monitor the whole network.

So a reasonably powerful adversary, for example, the NSA but also in terms of Bitcoin, private companies like cChainalysis and Elliptic or even other governments, China, Russia, so forth and so on. They can monitor all of the traffic coming into the connection and out of it, including in and out of the Tor network. That can be used to correlate and attack and deanonymise traffic going through Tor as well.

So Tor is about as good as you can get right now. But we know we should be able to do better.

So, I mean, there is an interesting question of ‘does anyone care?’

Now Fred Wilson from Union Square Ventures, who is a kind of famous venture capitalist, I think he blogs at AVC.com, even he notes that mass surveillance by governments and corporations will become normal and expected in the 2020s, which I think is true.

And people will turn to new products and services to protect themselves from surveillance.

“The biggest consumer technology successes of this decade will be in the area of privacy.”

We have seen some success already. DuckDuckGo, which is a kind of anti-Google search engine, is doing really well.

Protonmail, which is a privacy enhancing alternative to Gmail, adding features like private calendaring is doing very well.

My general hypothesis is that the privacy market could be as revolutionary as the advent of encryption.

Remember the lock in your browser, which means you're using TLS or SSL after the Crypto Wars for the web.

A lot of people said that crypto should be illegal, or, you know, governments were pushing for it to become illegal back in the 90s, before Bitcoin. But the fact of the matter is, without encryption in the browser, you couldn't have banking online. If you sent your traffic over an internet connection it could be intercepted by a random guy on the WiFi in the cafe if it was unencrypted, they would get your passwords and whatnot.

We used to do this at a hacker conference called DefCon. We put everyone's passwords to their emails and their various services on a screen. So it used to be very easy.

Now with encryption, there was a whole new generation of companies like Amazon, eBay, or in China, companies like AliBaba, so forth and so on, that were enabled by, essentially, privacy technology.

So we think, you know, COVID-19, I think has showed us how dependent we are on the internet for our daily lives.

And so the real question we have is: what kinds of new enterprises, endeavours, and projects will widespread privacy and anonymity lead to?

The question though is: does anyone care currently? So maybe in the future people will care. But will anyone care now?

So we have seen that a lot of people use Tor. I hope you guys use Tor. I use Tor every day. I run a Tor relay, we should all support Tor.

Tor isn’t perfect but it is the best we have currently for general-purpose internet traffic.

So you know, we have 20 gigabytes a day going through 6,500 nodes, at 2 million users.

The growth kind of went up and down, but somewhere between maybe 2 million and 6 million.

Signal is interesting. Signal used to be very small. I'm quite good friends with Moxie. And I remember back when Signal was thinking of shutting down in 2016.

Signal went from a few million users in 2016, and then Trump became elected and it skyrocketed up to 10s of millions.

And now we believe that Signal may have up to a billion users. That's a huge amount of users.

But the problem is that Signal and Tor and a lot of privacy projects are nonprofits. So when Trump cut Tor’s funding, Tor had to reduce their staff by a third.

Signal almost ran out of money, but the founder of WhatsApp, who's pro-privacy, came in and helped them out and is now CEO.

But the real question is: can we find a sustainable economic model for privacy?

I believe the answer is yes, we can have a sustainable economic model for privacy by taking advantage of the lessons we've learned from Bitcoin.

Bitcoin has given us a true wealth of economic knowledge that we didn't have before.

What we've done is we've taken this economic knowledge of what we gathered from Bitcoin and we've applied it to the domain of privacy, in particular, what's called a mixnet.

So I'm going to go into what a mixnet is and exactly how it works shortly.

But a mixnet is a multipurpose kind of anonymous traffic overlay, so it's on top of the existing internet.

And its function is to prevent traffic analysis by an adversary, an enemy let's say, that's capable of watching the entire network. We should be capable of resisting traffic analysis from even the most powerful of adversaries, including the NSA.

The problem is that historically, if you're running a volunteer network, how do you know if the network's actually working?

How do you get people to run these servers?

We have a Nym token, which incentivizes privacy, it uses a reputation token for the nodes and network.

And it makes the mixnet decentralised, sustainable and resilient.

And, of course, you're not just sending packets over the mixnet you're sending data.

So we also have a kind of zero-knowledge system called zk-nyms, which allow third-party apps to anonymise any kind of key value pairs, if you're familiar with JSON, so that users can privately reveal part or all their data at their discretion, as necessary for compliance, authentication or just making the app work.

So if I'm, for example, logging into a VPN, like we're working on, I would need to prove that I paid for the VPN but I might not need to reveal my identity.

If I was logging into Signal privately, I want to reveal my phone number, because that's what uses my identifier on Signal, or my username but without my phone number for Telegram, so forth, and so on.

To understand intuitively how a mixnet works, we have Nym, Tor and VPNs compared in this diagram.

Let's start at the bottom.

So you can see the packets - those little orange dots floating by the screen. These packets, the size of the packets and their frequency, which is what you could also call the timing and the volume information, is revealed by the internet itself by the underlying Internet Protocol, TCP/IP transmission control, slash Internet protocol.

So you can see the VPN just sends all your traffic to another computer and then that computer sends it right back out to the internet.

And if I'm looking at that VPN, let's say the VPN is trustworthy, I just need to look at the data coming in to the VPN and out of the VPN and I can deanonymise packets just by looking at their size and timing patterns.

Now Tor is much better because there's not a centralised point of control.

The Tor entry node, which is the first node as you go into the network that then pops you to another Tor relay and then out through an exit node.

These guards, unless the entry and exit node collaborate, don't know what you're doing. And that's good. And that's why Tor is always better than a VPN.

However, you can see that the fundamental size and timing of the packets is still revealed.

So if an adversary like the NSA is watching all the incoming traffic of Tor and all the outcoming traffic of Tor, they can, indeed with some statistical luck, because the circuits in Tor change every 10 minutes, deanonymise at least in 10-minute bursts, the traffic patterns of Tor users.

You can see what a mixnet does is a bit different.

So with Tor, and with VPNs, the traffic goes into them in what's called a first in first out manner.

So the first packet that goes in is also the first packet that goes out of any given server, that's the same with Tor. Same with the VPN.

Now mixnets change this because they do, as we would say in English, what they say on the tin. Mixnets, not surprisingly, mix traffic.

So when packets go into a server in a mixnet, the mixnet server mixes those packets.

If you want to think of those packets like a deck of cards, each packet has a card, and let's say they're encrypted, so they're turned over - so you don't know if it's an eight of spades or a jack of hearts or whatever.

And then when enough packets go into the server, those packets are mixed. You could think about it as shuffling, like shuffling a deck of cards.

So the first packet that comes in is not necessarily - it could be but probably not - the first packet that comes out. Instead, maybe the second packet that comes in the mixnet comes out as the first packet and the first packet comes out maybe as the third packet.

Then we add fake packets. And all the packets are encrypted using what's called the Sphinx packet format. So they're the same size and their encryption is unlinkable.

This basically makes the packets anonymised.

The packets are mixed up, they all look the same, they come in and out a different order than they came in to the first nodes of the mix, from your computer.

The interesting thing about mixnets is that - we have a different design to the classic Chaumian mixnet, it’s called a stop and go or continuous time mixnet.

Essentially, these mixnets are scalable. The latency is a bit tunable by the user per app. So some apps will be faster or slower than others, they can make their own decisions about how much privacy they want, and they're generic.

So it's not just for blockchains, any application including web browsers, instant messengers, crypto, Signal or Telegram, they could all use a mixnet. You don't have to be in blockchain or cryptocurrency to use a mixnet.

23:29
Again, how it makes sense work is that to some extent, there's similar to Tor.

Traffic is routed through multiple nodes and this unlinks the origin and destination of the traffic, which is identified by what's called an IP address and an IP address looks like something like 255.0.147.33, essentially four numbers dotted together using IPv6 or something much larger.

The IP address can track you down more or less to your neighbourhood. Even down to your router, if you're very unlucky.

Like Tor, we basically delink the IP address of the user from the IP address that any service or website sees.

Now, one problem is I want to be anonymous. But let's say there's not enough people using the mixnet, or would using the mixnet reveal if I'm being anonymous or not, if I'm trying to hide my identity.

To some extent, we can hide that by adding what's called cover traffic, also called dummy traffic. And this prevents traffic analysis by essentially adding fake packets, fake information into the network.

No one else does this. So we can essentially add more privacy even if there's not as many real users sending packets through, because the users are always sending some packets through, even if they’re not actively using the network, at least if they're online.

If they’re not online, we obviously can't send cover traffic through.

Now the nice thing about cover traffic is that the more people that use the network, the less cover traffic you need. Because you have a larger set of people you can hide among, as someone will explain at some point.

Privacy is related to a large extent to how many people use a network. So if only three people use a network you’re private among a set of three people.

If 30,000 people are using a network, you’re private among a set of 30,000.

So I think that it's much better to have more users than not, because you have more people to hide among. Now, if you don't have that many people, you can add more cover traffic, that can make you more private.

Essentially, you can maintain the same level of privacy and reduce cover traffic if more people join the network. That's a really nice feature of our system. With blockchains, the more people that use a network, the higher the transaction fees get and the slower the network gets, it's a kind of form of congestion, because blockchains are essentially kind of like an auction for limited or finite block space.

But a mixnet is more like a classic web server. This is what we call horizontal scalability.

Unlike blockchains, mixnets can expand to allow for more traffic, and that makes them more like the traditional web.

The other point is timing obfuscation.

When packets come in, they aren't just shuffled, they're delayed, and they’re delayed by a process called a Poisson process.

It's kind of a random delay, let's say.

But this is a special kind of random delay. We don't know when any particular packet will go out of the network. Some packets may actually never go out, in which case you have to send them again. But we know the average rate of a packet going out of the network.

So that basically means that these packets reordering at each hop between each server prevents traffic deanonymisation.

Just to explain in a little bit more detail, you are the user, you're sending traffic into the network that can be either TCP/IP or UDP packets, you're running the Nym client on your machine, which currently is NymConnect if you want to download it from the website.

These packets are all transformed into Sphinx packets all made the same size, maybe some cover traffic is added on, they’re sent into the mixnet, the mixnet has some active mix nodes - it has some others that may be inactive at any given time period - we have currently I think 500 or so active mix nodes.

They mix the traffic up with three hops, each packet is sent through three. So they're all treated the same: if packets were sent through a variable number of hops then you could use that information to deanonymise packets, that's for example, what other peer to peer quote unquote “mixnets” do.

And as the packets come out, they're anonymized, they're all the same size and they go to the internet, apps, services, friends, whatever.


So I guess the other question is; what does the Nym token do?


The problem is we need all of these mix nodes.

Computers aren't for free, someone has to pay for them.

And so the idea of the Nym token is there should be some price for using the service that matches the demand.

And of course, in the beginning, like many things, there isn't much usage. So the packets that are mixed  may not be enough - and may not provide enough rewards to reward the mix nodes. In that case, essentially there's a sort of inflation schedule that decays over time. There's 1 billion tokens. They're ultimately deflationary. But there is an inflation pool that rewards them a bit like the Bitcoin mining rewards for those of you familiar with that system.

We're similar to some extent. But we have staking, so mix nodes and validators can stake NYM as part of the reputation and we have delegated staking, so you may not want to run a mix node. But you may want to delegate stake to that mix node.

By delegating that stake you show that you trust that mix node, you trust that it’s going to be up and they're gonna be mixing packets. You get some of those rewards from that mix node.

We have proof of mixing. So the mix nodes, their job is to mix packets. So we have a kind of novel proof of work scheme, that samples network traffic, rewards mix nodes for mixing traffic, it guarantees fairness by taking advantage of the anonymous Sphinx packet format.

Some of the edge cases of this are still under work.

But we'll go into those later with Claudia Diaz.

And then there’s value - so again, the NYM token is approximately valued, in an ideal world, to the usage of the system. So if more people use the system, it would, in theory, have some effect on the price.

And so the NYM token, if you're think of the VPN, will give you access to the VPN, for some period of time, and we'll go more into that later.

So again there's users, it's a complex system. There's mix nodes, which are mixing traffic, and there's validators. And then the mix nodes are rewarded proportionate to the amount of packets they mix. And the validators ensure that there's a blockchain running.

Why do we need a blockchain? We need a blockchain because users need to locate the mix nodes to find out their IP addresses and their cryptographic key material.

The blockchain also keeps track of the reputation in terms of NYM token, of every single mix node, and rewarding people is done via a smart contract, on what we call the Nyx blockchain, which we'll go into more later.

And there's service providers, which are just people who have provided Nym-enabled apps.

We don't imagine that most people will buy NYM token directly to access the network. But they can buy these zk-nym credentials, say I’m Signal, I could buy that zk-nym credential, embed that zk-nym inside Signal, and Signal would let me access the network.

For what appears to be free, but actually Signal is paying for it or Brave or whoever. That's how we imagine this working.

So just to go into that in a little bit more detail:

Zk-nyms are essentially what you would call an anonymous authentication credential, or what is sometimes called an  attribute based credential.

And they are used to access the mixnet.

So obviously, if the mixnet was just free and open to use, people could spam it, tonnes of network traffic would come through. That's why Adam Back invented proof of work, in order to prevent spam through mix nets, like including anonymous email remailers, which were a version of mixnets.

We have a similar system, it's an anonymous credential system. It's a bit like privacy pass on Tor, but it's decentralised.

And it lets you have a lot of utility, you can basically say: I get a zero-knowledge proof, that's the zk, that I have paid, for example, for using the mixed net, just like I paid for using a month of a VPN.

When I pay for something, the danger is that that payment deanonymises me, the very fact that I'm paying for something might reveal that I'm using it or when I'm using it. We don't want that.

So the zk nym credential system is essentially zero knowledge.

So I can have a zero-knowledge proof that I paid for something without revealing who I am, or even which user I am. I'm just revealing the fact that I paid for it.

These credentials are cryptographically unlinkable via randomization, and they work not just on Nym, but with any kind of currency including Fiat.

They enable selective disclosure so you can produce not just proof that you paid for something, we can produce other things with zero knowledge including KYC AML information or unlinkable pseudonyms for accounts, so forth, and so on.

It is decentralised. Unlike other anonymous credential systems, we are decentralised and no centralised third party is required to issue these credentials. It just uses what's called a threshold signature, issued by the majority of the validators who are maintaining this blockchain anyway.

This is kind of an overarching flow diagram. The way you want to think about it is that there's a blockchain, that's the Nyx blockchain in the middle. When you're a user, I get some tokens somehow, or maybe a credential for some token. These create a credential that shows I have this much bandwidth to use, I show it to the first node in the Nym mixnet, that's called a gateway.

The gateway checks to make sure it's a valid credential, but it doesn't know who or what payment that's connected to, because again, that's done in zero knowledge. Then once that's done, the credential lets you access the network, you can then get rewards.

You access the network, your traffic goes in the network, each packet is mixed, comes out through another gateway, and then goes to a service provider. And at the end of every kind of time period, the mix nodes and the gateways are rewarded based on how many people have used the network and any inflation pool.

So this is a kind of exciting system.

We have lots of people that have backed us, and all sorts of funding and whatnot and that includes Binance Labs, Andreessen Horowitz, Digital Currency Group, Polychain, lots of interesting other firms from Europe like Kryptonite, Tioga, 1KX, Swiss Staking.

We have exchanges that everything is on such as Kraken and so forth that were also interested in us.

And lots of validators that are running nodes. I won't go into each of them. But they're all really wonderful for doing so. Including some of the people that back us, like Edenblock also run their own nodes, a lot of Asian funds, GC Ventures, Fenbushi, Hashkey. And some national ones - we are based in Switzerland. So Swiss Staking, for example. And yeah, we have lots of different partners. We're a pretty pretty large ecosystem, I would say next.

We don't know approximately how much the [privacy enhanced] ecosystem is worth. But we can say it's pretty undervalued right now. The General Data Protection market is about 70 billion and it is growing pretty rapidly. It mostly goes to lawyer fees. So what if technical solutions can make those lawyer fees go down?

The mixnet is not exactly a VPN, but it's pretty close. We know that the VPN market is at least $25 billion.

One in eight people on the entire planet uses the VPN and about 30% of VPN users say they're interested in privacy.

So eventually we hope that Nym can be built into the core protocols and routers.

And of course, cryptocurrency: a lot of people are very interested in privacy and cryptocurrency, being at least equivalent to privacy in the traditional fiat system.

Your cryptocurrency transactions are completely linkable to who you are. So I can look at your ETH account and see how many NF Ts you have and see what your ETH balance is. See what your Bitcoin balance is, I can determine that pretty quickly. And that could be dangerous.

So you know, I can just get someone's bank account balance by looking something up on the internet. So we think that there's a huge possibility and a lot of projects very interested in integrating against Nym in the cryptocurrency market next.


So we hope all sorts of people jump on board. We need of course app developers, we have a grants programme, we're getting very interesting people to build on top of them. There's all sorts of people building things like file storage systems and whatnot on top of Nym now.

So I want to share a file privately, something Chelsea Manning, for example, could describe.

We need people to run nodes, mix nodes and gateways, we're gonna need a lot of gateways in particular.

And validators is going to open up pretty soon. So there's lots of different kinds of machines you can run.

And we need people just to use the network, people who actually want privacy.

You can currently test the network using the Nym Connect app, which I'm sure we’l explain to people how to download and use.

So our current plan, just for the last few years, we've launched our original system at the Chaos Computer Congress way back in 2018. we have a white paper, you should read it for all the technical details, I skimmed over a lot of stuff. We launched our mainnet, we got lots of funding from Binance, a16z, and Polychain.

We launched everything on Coinlist in 2022 after our mainnet launch, we've had integrations with Electrum, Keybase, Telegram. We have lots of apps, mix nodes - we started the Shipyard academy, this is the second round of it. We did another round last year. And now we're interested in scaling Nym, including launching something that more or less looks like a paid VPN. We've opened offices in different countries, Paris is the latest. We're in Neuchâtel, Switzerland, for those of you who are Swiss.

We're trying to have larger communities in different parts of the world, China. Asia in general is a big focus.

But also a lot of countries in South America, Africa, the Middle East.

And if you want to develop, we have a great Rust and WebAssembly SDK.

And we hope to really launch this VPN by end of the year so people can really use the Nym network to access anything they want.

After 2024, we're going to focus on scaling, making the system easier use, looking at places that are under censorship like Iran, to try to make sure the mixnet is really, really secure, we could still use a few more audits on it, and make sure the validators get rewarded and eventually go to millions of users and integrate with browsers, operating systems, apps.

Again, I used to work on this stuff under Tim Berners Lee so I’m familiar with the whole ecosystem, including standardisation of Internet Engineering Task Force, and use secure key management and hardware optimization as much as possible.

So this means just as people would buy, for example, ASICs to do Bitcoin mining, we can eventually imagine people buying special hardware to mix more packets so that we could speed the network up 100 times. So that's a really big deal. And that's what Chelsea Manning is working on right now.

So we have an ambitious roadmap, and I am open to answer any question.


################# Q&A ###############

Our data is stored by other applications, no matter how much it is said that it is not. Does Nym store our data? Because our data will first go to the Nym VPN to be anonymous. Will Nym then store this data?


Harry Halpin: It's important to remember that with Nym, your data is encrypted on your computer. That's why you have to install this NymConnect device on your machine. This program encrypts the data locally, then once it's encrypted, and put in what's called the Sphinx packet formats, where it all looks the same, then it’s sent into the mix net.

The nice thing about that, is that given the encryption happens locally, even though the data is technically stored by other servers in the system, those servers have no access to the data and everyone's data looks the same.

So as long as multiple people are storing packets on the same server… But if the system has more than one user, which currently it does even today, then the servers, they store data, but given that they store it in encrypted form where all the packets look the same, they don't get any knowledge from storing your packets.

That being said, when you send packets to a system through them, the server at the other side does decrypt those packets. So you are trusting the server at the end.

So let's say I am using Nym to access a Google service, Nym would prevent someone who's looking at my computer from knowing that I'm accessing Google. But at the end of the day, I'm sending information to Google, so Google would still have that information. It's just Nym wouldn’t know. Actually, the last server would know that some information is going to Google but they wouldn't know exactly what.

Intuitively, you want to think of it as a pipe. So you're sending - or if you are familiar with academia, it's called an anonymous channel. You're sending data into a pipe, a data pipe makes it anonymous. But then when the data comes out of the pipe it's not anonymous anymore, but the pieces of the pipe don't have any access to what you're doing. And they prevent people that are looking at you from knowing what you're doing, and that's really the value that we give.


So Nym is a truly great revolution. But how will you make big states and organisations accept it? I guess the question here, the most important part, is states.


Harry Halpin: The important thing about nation states is that they want to spy on their own citizens, almost across the board universally, but they're afraid of other nation states, spying on them.

So originally Nym comes from a European Commission research project called Panoramix.

Where the European governments were upset, in particular Merkel's German government at the time, that they were being spied on by the US. They said, can you produce a VPN that’s spy proof that can defeat the NSA? We said yes - I was actually working for the French government at the time.

And we did produce this system. I don't think they're using it right now. In fact, we had more interest in terms of funding from Binance and others. But, the fact of the matter is that, you know, we currently still have interest from governments using this system to defend themselves from other governments. So smaller governments, for example, like Switzerland, who are very pro-privacy would like to defend themselves from larger governments.

So you know, I do think governments at some point will have to make a choice. Will they let their citizens have privacy?

In that case they won't be spied on by other governments.

Or will they want to be surveilled? And the fact of the matter is, you know, if I can surveil my citizens, other governments can surveil my citizens, and that's quite dangerous.

So we haven't had any pushback from governments yet. Although we have a lot of lawyers in case we do. So we're very well prepared.

How do apps know who is sending info if the data is anonymised?

The data, they won't do anything. You choose the app, you can see this with NymConnect when you use NymConnect. And then that app talked to by the Nym network.

So while it's in transit, it's anonymous, but you kind of bundle up the data For the particular app you want to talk to, so when it gets that last node in the network, which is called the gateway, that gateway sends that final packet to the app, and the app de-bundles, and they get the exact same data that you send on your computer.

So as long as the app supports Nym natively, they get the exact same data they would have sent without using Nym, so they know everything that you're going to send. But in between the data is kind of wrapped up.

It's like I'm sending a mail package. So you know, I wrap let's say a birthday gift locally using a box, I send it to the post office, the mailman doesn't know what the birthday gift is. And using the power of cryptography, you can’t open it, you just ships, it ships. And when it gets to the friend's house, my friend opens it and it's their birthday gift.

And that's kind of the same way the Nym network works.

Now, that being said, that does depend on you, the user having the Nym software installed, and the app, having the Nym software installed. That is not going to be the case for all apps. For apps that do not have the Nym software installed. The last hop of the network will have to kind of open that packet, open that encrypted packet, open the encrypted package, and they will get to see the data and send it out. But you know, assuming that data is encrypted, they shouldn't see that much.

An important distinction you might want to make between apps that have native Nim integrations, where the app actually speaks in the protocol. And apps that you can access via Nym but they don't speak the Nym protocol.

So for example, let's say Zcash integrates against Nym directly. Then the ZCash nodes wouldn't really know too much about what's going on. Or let's say Signal integrating them directly that Signal itself would not know too much about what's going on until you get the last hop.

But you can also access apps that aren't integrated with Nym. But in that case, the last hop of the Nym network does get to see which app you're going to because they have to open that packet up and send it to the right server. It’s a subtle point.


How easy is it to monitor someone online for a regular internet user?


Harry Halpin: That's pretty easy. I mean, it depends who you are, right? So if you’re using, let’s say Wi Fi in an internet cafe, I can turn on a program called Wireshark. And then I can watch all the packets and where they're going. If you don't have them encrypted, I can look at your password and get into what you're doing. And we actually did that live. This T-shirt is from Lugano, Plan B. It's a Swiss city. We'll be there again next week to demonstrate our software and talk to folks. We did a live demo last year in Lugano, where we were in a room full of Bitcoin people. And we showed them that we could track their Bitcoin transactions, just sitting in the same room with them. And if they shot something across the network, we could see it.

So it's pretty easy, you don't have to be the NSA to sort some of the stuff out.

Now the NSA or governments or Chainalysis, or Google, they can watch much more than we can in much more detail because they can watch all the servers on the internet, which I can't do using a simple programme like Wireshark from my home computer.

But nevertheless we can expect that the future will be a bit cyberpunk. So that these capabilities which were once in the hands of the NSA and only the NSA, maybe you trust the American government, but you don't trust others, will go into the hands of other governments.

So Russia and China are now having probably greater greater cyber cyber surveillance capabilities. Imagine - the Bitcoin network, it's private actors like Chainalysis.

We could imagine that these kinds of capabilities will eventually get in the hands of almost ordinary people to do mass internet gathering data gathering.

There are whole companies, such as what was it called Team Cymru, who, weirdly enough were on the board of Tor, whose entire business model was  we capture lots of data, we sell it to people.

And they claimed that they could do these kinds of mass data gathering attacks to deanonymise even Tor, then sell that data to folks like, I think they sold to the US Navy or CIA or something, I'll have to check.

But we can imagine that these kinds of companies and kinds of surveillance capabilities will become more powerful over time. We haven't had something like the Snowden leaks recently. So we don't really know.


It's fair to say that we're just in time with Nym, right?

Harry Halpin: Yeah, well, I wish we had done the software 10 years ago, we kind of knew how to build it in the 80s. Just no one had the funding or the time. I think the wonderful thing about cryptocurrency is that a lot of talented programmers who could be working at Google or other places are now working on privacy. And we think this is a much better and more ethical use of their time and that we do believe that the software is probably the last hope we have against mass surveillance.