Vitalik Buterin & Roger Dingledine on Integrating Tor into Ethereum | Ethereum Privacy Stack

the downside is minus 100% % where you uh really uh depend on that level of uh reliability, right? And aside from uh privacy, the other reason uh is geographic decentralization, right? We do not need a network where even there there because of incentive pressures, nodes would naturally all concentrate in like London and New York, right? We need a network where it's not just possible but also economically viable to participate from uh you know places like Bonosirus um and uh Southeast Asia and uh like say yeah Zambia right and uh this is but but then uh improving privacy at the same time is uh is important right and I think we've started by uh identifying the u the places where privacy is the most important, right? And uh the really key one actually is privacy for denial of service resistance, right?

Because uh even though the network has a lot of nodes, there's a very small number of nodes that are responsible at each individual slot, right? And so there's one node that is the proposer and uh if you can identify the proposer ahead of time and you can do a a targeted denial of service attack that lets you uh basically kill the chain and uh the possibility of that happening is itself you know an incentive for people to just uh rely on centralized actors. So for a few years we've had this these proposals in the pipeline. They're called SSLE, single secret leader election that basically allow proposers to be chosen in a zero knowledge way, but where only the proposer is aware that they're propos the proposer right up until the very moment that they actually make their proposal. And actually, if we want you could even uh you know with the newer ZK stuff extends that and say you keep your privacy even after that point, right?

Um and so that we've been exploring that. Um and uh like in I think in all of these situations like you have to combine a kind of blockchain layer like data leak minimization with a network layer um equivalent and uh some form of uh onion routing and mix nets is going to be very important for that. Um and uh yeah the uh the core uh you know the core challenge here basically is uh making that uh both uh work and be fast enough and also be uh economically viable right so that people don't you know feel uh feel the pressure to give up on it and like do things like maximizing transaction fees. You said the word censorship resistance earlier and this is actually the phrase that that confused me at first because I found the the the cryptocurrency people years ago and I'm like I work on censorship resistance. We've got Russia trying to block tour.

We've got China trying to block tour. And they're like oh wow you got to come talk to us about censorship resistance. We do that too. And it it was months until I realized they meant application level censorship resistance as in you can't stop transactions from happening. And I meant network layer censorship resistance as in users can reach your network.

So China has said they don't like cryptocurrencies.

When is China going to block Ethereum at the network layer? When are they going to start rolling out great firewall style blacklists of the validators and uh and the proxies and so on and and what do we do then? Yeah, I mean know I think these are things that are really important to you know build build up uh defenses on and uh you know it's like we need Ethereum to be a globally access a globally accessible network where people from uh anywhere in the world are able to participate as users and also as uh you know nodes you know doing staking and participating in the network right and uh these are there's like a certain base level of this which is uh possible just by being a decentralized network and I think like 10 years ago that was enough and now that's not enough and uh really yeah building out the technology to go on top uh to actually yeah make it possible to have a node to have it not be immediately detectable as a node and uh at the same time like actually be fast enough is uh I mean it's a it's a super important problem. Um I mean one thing also I think proof of stake it uh often gets talked about either about helping the ETH supply go down or about not killing trees which are both good. Uh but uh it has this other benefit which is that it um allows u you like it basically means that to be one of the nodes that is holding up the network you don't need this uh huge server farm which is incredibly easy to detect and uh incredibly easy to uh censor and uh and instead like as a a proof ofstake note it's like basically you can run one off of a laptop right and uh you know in the future hardware requirements for that will uh go down even further right so that's uh something that we uh care about as well so no I think it's an ongo an ongoing journey and an ongoing process

so one of the things that we in the tour world have been working on are called pluggable transports and the idea is tour takes care of your privacy and the transport that you have inside tour browser transforms your tour traffic into some protocol that the sensor doesn't want to block and talks to some destination the sensor doesn't know to block and I was talking to Igor from PSSE and he's excited not only about adding onion services to everything and everything but also including some of the pluggable transports. There's one called Snowflake that transforms your traffic into WebRTC. And part of the goal there is to have like the wallets and the blockchain reading software not only be able to give you privacy at the network layer, but also eventually have some of these pluggable transports included so that if you're in China and you want to get to that website they blocked, then you can either it turns it on for you or you can click help, I'm being censored and then you can route your uh transaction read or write over something that still lets you reach the network.

Yeah. No, I think uh that would help a lot. I mean, I think from a yeah privacy perspective, from a censorship resistance perspective and uh even just from a resilience perspective, right? Like once you uh move away from just talking directly to a server, you know, often things break and having multiple ways to connect to a network helps things uh break less. So I think uh that's it's a major win for Ethereum and I yeah look forward to us integrating all of this.

Sounds good. We are officially out of time according to this, but according to the organizers, we've got six more minutes. So, we're going to keep talking for six more minutes. Um,

in in terms of like, so we've got some short-term plans of of ununifying, torifying a bunch of pieces of the ecosystem.

What does this look like in in a year or two once we've got

I guess what are the what are the what are the stumbling blocks between then? Do you

do you envision all the wallet operators being excited and they just do this or do you need to go one go to them one at a time and advocate to them? How how do we get from here to there?

Yeah. Uh so I think the good news is that with the Kohaku stuff that we've announced this week there's already a lot of momentum of uh wallets understanding that you know we are working on these kinds of things and uh that we are and are going to be churning out software packages that makes all of these things as uh plugandplay as possible. Um and so there's been already been interest from quite a few wallets in uh integrating these technologies. Um, another piece of the puzzle I think is uh going to be uh like basically verifying that wallets are actually doing it, right? It's uh the the difference between um you know claiming that you're decentralized as a marketing term and actually being decentralized.

And so you know like we've had a a project called L2B which like basically verifies like how trustless are individual Ethereum L2s and it publishes a bunch of information. and it kind of compresses it into you know like stage zero, stage one and stage two. Um and so there's um you know a project uh that's like still like starting early stage wallet beat uh to do a similar thing for wallets. Uh people have also suggest uh like I know talked about like doing zk beat for like zk protocols and uh like just expanding that kind of account um you know accountability layer for for the ecosystem. So that's uh important too.

Um so you know existence of technology existence of these like easy to use SDKs um like actually ver verifying that people are doing it and not just advertising that they're doing it but still collecting your data in a 100 different ways. Um like these things are all really uh important near term. Um and then if we can get to a place where like in you know one year at least wallets uh will users will be able to access wallets and we'll be able to access applications through wallets in a way that uh the users privacy is protected both on chain and in terms of their interaction with both the DAP and the RPC node. like I think that's already in a amazing goal and we'll be in a much better place and uh then you know in the strength of the Ethereum ecosystem is it's so parallelized right and so in uh five years hopefully we'll have much better user level privacy and we'll have a much more hardened and uh privacy friendly uh and resilient Ethereum network and we can have uh all of these things uh together um so um that that's exciting for What what excites you about the next 5 years of tour?

Yeah. Oh boy. Um part of the momentum of uh people around the world are starting to realize I mean we saw the the Snowden documents, we saw the NSA leaks, we see more and more companies losing their data. So there are a lot of

ordinary people out there who are who are still realizing, wait, privacy is important. And part of the part of the fun of working on all this is reaching so many different types of people who care about privacy for different reasons and helping them understand what tools are available and how they can be more safe on the internet. So uh again for me it comes down to the people. So finding all the different NOS's around the world who are trying to help their communities be safer. That's the part that that excites me and we have a lot more work to do in terms of like every time I read about a new law against uh against certain kinds of people in Uganda or like rightscon is happening in uh in Africa this coming uh couple of months and they have some laws that are making the trans LGBT LGBTQ community not not want to go to RightsCon because that country is a scary place to go to.

So, uh, while the world is realizing privacy is more important, also the governments are making it a scarier and scarier place. So, we'll see what happens when those two collide.

Yeah. Well, I think those of us who really care about privacy should remember that we have a lot more allies and a lot more people in the world who care the about the same causes than we realize. and we need to really figure out how to strengthen those uh relationships and uh work together to you know make a more private uh more free and more dignified internet happen.

Speaking of working together, one of the topics that I've been pondering so we talked uh on Wednesday about tour as a commons and like the capitalism based building block or the altruism based building block. So the tour network is made up of volunteer relay operators and as a project like Ethereum shows up and adds a lot more load to the tour network one of the things that's helpful to do is since it's a commons to give back at the relay capacity layer also and that could that could look like the Ethereum Foundation running a bunch of relays or there are a bunch of nonprofits around the world that already run relays and it could look like helping to support those nonprofits. maybe uh tell us about uh running a relay like what are what are the costs? How much compute do you need? How much bandwidth do you need?

Yeah, so uh running a relay is is just installing the tour program and editing a text file to configure it to be a relay and and having an IP address that's reachable from the rest of the world. So you can do this at your university, at your workplace, some people do it from home. In the modern world, people get uh virtual servers somewhere and run relays there. And part of the challenge with that is they all get the same virtual server at the same German company. And then we've got centralization versus diversity challenges.

Yeah. Maybe one of the things we could do is uh make it easy for stakers to also be relays at the same time.

Yes. So, yep. That would be awesome. So if if you're working on uh packaging easy staker software, come talk to us about how you could also put in a tour relay at the same time. And one of the other fun things about adding onion service support to all these things.

If you're connecting to like CNN.com over tour, you need to have an exit relay involved so you can get exit the tour network and go to that website. But if you're going to some wallet service over its onion address, you don't need an exit relay involved. So getting all the stakers running non-exit relays is uh is less of a request, less of an ask, easier for them to do. It's all about bandwidth and connectivity and not so much about uh whether it's scary or not to to be a tour relay.

So I think that would be a great easy onboarding way to help you help us help you.

Amazing. Thank you, Roger. Okay.