Keynote: Public Goods + Ethereum Future | Vitalik Buterin, Ethereum Foundation | ETHTaipei 2025

[Music] Great. Uh so Ethereum depends uh is open- source software and Ethereum fully and completely depends on open-source software. Right? Uh so this is uh a list of uh open source projects uh that I uh found basically just by searching for it for about 10 seconds. And you can see just a huge number of different software libraries created by a huge number of different people.

Some of them in the Ethereum Foundation, some of them by companies in the Ethereum ecosystem, some of them by other nonprofits, some of them by independent volunteers. All of which are just a a critical part of making the Ethereum ecosystem actually work, right? So if Ethereum was not open source, then first of all, nobody would trust it. And second, it would be much harder to make progress because uh like if the code isn't open, that just makes it so much harder for anyone else to come in and uh and collaborate and build on top of what already exists, right? So um we can also think about this by by thinking about public goods right so you know in economics we've talked about the the phrase public goods for about the last 100 years or so it's one of the major topics uh in um of discussion about economics basically if you have a project that benefits a large number of people but where you can't uh like make some fine grain choice of who gets the benefit and who doesn't get the benefit, right?

So, if you think of something like cleaning the ocean, right? You can't clean the ocean in such a way that if someone pays you 25 cents, they get the clean ocean, but if someone doesn't pay you 25 cents, they get the dirty ocean cuz you know, there's one ocean. Um, so in a digital ecosystem, I think uh it's really important to like re appreciate that the average project is a public good, right? Um so software in Ethereum is a public good. Documentation, education, resources that help people understand what's happening in Ethereum and the other things that are being done in the ecosystem.

These are all public goods. Research um academic research, cryptographic research, economic research, all kinds of research are public goods. Protocol development. So if someone figures out an amazing EIP that solves all of the scalability problems, then that's a public good. Everyone in the Ethereum ecosystem benefits from it, right?

And so Ethereum crucially benefits from all kinds of different public goods. Now we can compare these two terms, right? So the usual definition of public good in economics is something that is non-excludable and non-rivalist, right? So something that provides a benefit to a large group of people. And so one person benefiting doesn't reduce the benefit that someone else gets.

And like you can't choose who benefits and who doesn't benefit. And like this is the way that people in economics tend to think about the concept. And it's a meaningful category because public goods are a category of important goods that's uh often difficult to um use regular market mechanisms to pay for. Right? If it was possible to like clean the ocean in such a way that you would only see the clean ocean if you paid 25 cents, it would be easier, right?

People would just do that and they would collect 25 cents from a lot of people. But this is not how the ocean works. And also this is uh not how the infinite garden of Ethereum works. Now in practice I think uh a lot of people also end up just using the word public good to mean like something which is good and something which touches the public or sometimes even something which is good and something which is created by a some kind of public entity right which could be a government could be a nonprofit. And so one of the challenges with the term public goods that I'll talk about is basically you know it does have some of these uh of challenges in its uh in its definition that uh can can make it difficult to really be rigorous about what kinds of things are we supporting what kinds of things are we not supporting.

Now we could also think about the term open source right so the words open source and free software um which are not entirely the same but they're very closely related they have about a 40year history and the open source institute has an official open source definition the free software foundation has a free software definition including the four fundamental freedoms and basically what these definitions boil down to is the source code being open and you being legally free to use, redistribute, modify the source code including commercially without discrimination. Right? So the nice thing about these definitions is like the definition of open source like it has teeth, right? Like you can't just go and say, "Oh, I built this tower and everyone likes this tower, therefore this tower is open source, right?" like for something to be open source like there are specific criteria that it has to actually fill.

So why do we care about both of these things right? Because uh Ethereum's future crucially depends on open source and it crucially depends on public goods. And it it depends on these things for two reasons, right? And I think of these reasons as being one direct value, which is basically that public goods are useful, right? So if the Ethereum blockchain did not have Ethereum clients, then the Ethereum blockchain would not be running, right?

And the Ethereum ecosystem would basically be what it was back in 2014 where everyone was excited and like thinking about what they would do on Ethereum if they could, but the blockchain would never start. And eventually people would probably give up and like everyone would pivot to AI or something. So this is direct value, right? Now there's also a decentralization value and this is also often really important, right? Basically public goods out compete proprietary and otherwise nonvalues aligned alternatives.

So often if an open public version of a thing or of or a solution to a particular problem doesn't exist then proprietary versions of that thing will exist instead right so you know if uh you don't have subways and buses then everyone will use cars and often cars are much less efficient right if uh you don't have open internet protocols then everyone will instead use proprietary internet protocols and propri Internet protocols come with vendor lockin. They come with a lot of centralized power being concentrated in a particular company. So these are the two reasons why these things are important. Um and because they are important, we need to fund them, right? We need to fund open source public goods.

If they don't get funded, they do not get produced. But also in order to fund something, you need to select for it, right? and you need to exclude things that are not it even if they try to claim they are right so I think like as someone who's really been trying to support the public good public goods funding and public goods governance and all of these topics in Ethereum for probably the last six or seven years one of the things that I often notice is there's a a lot of uh people um who basically say look I made this thing I made a public Good. And then like you look into the fine print. You look into what that thing is and you realize it's a commercial.

It has a company behind it. That company has a token. Uh the same group of VC firms invested in it. They have 46% of the token. The lockup schedule ends in 4 months and they're going to sell it.

Then the so the token price is going to go down. And of course before it starts going down they yell about it to the public so that retail buys it. And then if your application uses it then you know you have to make this small group of people richer and then often a small group of people controls it and they can change the rules to be what they want but they still call it a public good. So this is the thing that we do not want right and to me like the uh one of the term one of the benefits of uh the term open source that I really like is uh it it really captures well the things that we want right like generally public goods within Ethereum that we care about they are open source software or open-source writing or or open source academic research But also it's a clear definition that very specific tougher that are resistance to manipulation that are resistance to like everyone trying to push themselves into the category because it's better for their narrative. So open source is a concept that was invented and perfected over the last 20 or 40 years in the context of a modern uh digital society and economy.

It does have a couple of important weaknesses, right? I think uh one of the big ones is that there are risks of fragility and of abuse of centralized services even if those centralized services get built on top of open source code. Right? So if you imagine Facebook open source or let's say Twitter, Twitter open source to the entire Twitter algorithm all of its source code tomorrow. Would that fix everything?

No. Why? Because well the version of Twitter that is on twitter.com or I guess now it's on X.com like it's run by one particular company.

It's run by a particular group of people. That's the version. They have a database. They control who can send tweets and who cannot send who cannot send tweets and uh you know who has free speech and who loses their free speech because they insulted the wrong person. And uh you know they uh like it doesn't matter if there's open- source code sitting somewhere in the corner because there is this like one single thing this like logically it's a centralized thing that there is a structure of control over and that structure of control is completely commercial completely centralized.

So the AGPL the aero general public license this was a license created about 20 years ago. it was actually created in part because of these exact concerns, right? Basically, what the license um says, it doesn't quite solve this problem, but what it does do is it say, well, you you're required to open source the code, not just if you create a derivative work and you try to sell it, but also if you create a derivative work and you try to run it as a as software, as a service, right? And so it solves some problems. It expands the set of conditions under which you have to open source code if you build on top of it.

But it also is far from solving everything. Also open standards and uh interoperability. Right? So this is another type of lock in is if you have one application and then that one application has standards that other people cannot interoperate with and then those uh standards potentially like if a company wanted to be really evil they can keep changing to try to lock uh competitors out. So last year I uh wrote this post uh that was on uh making Ethereum alignment legible and I gave a set of criteria for what Ethereum alignment might look like.

And one of the goals that I had is to basically make this concept of being Ethereum aligned that a lot of people seems to care about something that is not just a social game where you are Ethereum aligned if you are friends with the right people but you are Ethereum aligned if you fit a particular clear set of criteria. And so one of them is open source, one of them is open standards. So using public standards, using the same standards as uh as everyone else, if you have to create your own new standards, do that in collaboration with other people instead of just going off and doing it by yourself. And also decentralization and security, right? So uh like I have two tests here.

One of them is the walk away test, right? So the walk away test basically says if your company and your servers and all of its infrastructure just instantly disappear, will users still be able to get their assets out, get their data out, continue using the application, right? And this actually matters a lot, right? So imagine for example, you create a wallet and then for your wallet you use say zk email uh to do a recovery for the account. But then the way that you use the Zika email like the the actual recovery flow is uh depends on some like an interface that is part of the wallet and then there's like no it's not done in a standard SKIM compliance way and like there are and so if your wallet becomes unavailable then like the user can't use someone else's interface and like you haven't even provided any alternative interface and so the user actually doesn't have the resources that they need to actually be able to recover their their account because in theory it's controlled by your email.

In practice, it's kind of bottlenecked by this piece of interface code that you that you wrote and that you didn't really put anywhere else. Um, so the insider attack test basically says if your team itself tries to attack the system, they uh like basically they should not be able to steal your data and your assets, right? And so we saw for example, you know, with the buy bit hack the limits of uh the um resistance of the safe wallet to the insider attack test, right? Which is basically that the smart contract wallet passes the test, the interface does not, right? And so now there's a lot of work being done on actually putting UIs on chain, doing onchain versioning and like things that really increase that kind of security.

So another area where open source is not enough is uh open source has to get funded right and today in the Ethereum ecosystem open source gets funded by the Ethereum foundation it creates gets funded by a few other organizations but there is this open question right how do we create a self- sustaining flywheel right how do we create a self-sustaining ecosystem where you actually have projects succeeding because of open source and then some of the value of that success actually flowing back into the next generation of open source so that the cycle can continue to happen. Right? How do we make that fly work without depending on a centralized organization with a limited budget to keep using that limited budget in an unlimited way? And also like what about other aspects of uh doing the right thing, right? So, you know, you can uh open source uh soft software that can help people on the internet identify what's true and what's false or you can open source software that makes it easy for people to like spam the internet and social media with even more misinformation, right?

Like open sourcing a thing is not automatically good, especially if the underlying thing itself is not automatically good. And so like also where like where in this whole uh thing um like categorization do we actually think about these kind of aspects of doing the right thing that aren't just about openness, right? And this is one of those places where no the language of public goods really does apply, right? Um so you know if uh you put uh you know like substances in the water then maybe the substance that you're putting in people's water is florine. And then there's like one kind of debate about whether that's a public good or a public bad.

Um or that sub the substance that you're putting in people in people's water is like chemical pollution. And in that case you're obviously making a public bad right whether something is a public good or a public bad depends on whether the thing is good or if it's bad. So with open source like the the difference between good things and bad things doesn't get included in itself but like it has to exist somewhere right and so in this uh Ethereum alignment definition I have uh I I added these two other criteria which is being positive sum and being positive sum toward Ethereum uh toward Ethereum users and also positive sum toward the broader world right so the Ethereum ecosystem should not just exist for for its own benefit, it should also exist to provide broader forms of value to the wider world, right? And so these are things uh that I think we all care about, right? But then I think the interesting question is how do we create more mechanisms to ensure that our ecosystem supports the right kind of open-source public goods in a long-term sustainable way, right?

How do we do this in such a way that these things do not just depend on a small number of organizations that have one-time pools of money? How do we actually create these kind of robust long-term institutions that can keep preser continue funding these kind these kinds of things for the next years and for the next decades and then and continue um finding new sources of funding and so that these things can continue happening right like I think Ethereum should be an ecosystem where we work on resilient and long-term solutions to make sure that this shared infrastructure is being funded right like Ethereum I think basically since the beginning it's been an ecosystem where we really care about things like decentralized governance right we care about DAO we've uh cared about like things like quadratic funding we've cared about how to like govern things like oracles for defy protocols you know we've bu a lot of people have been enthusiastic about building like various forms of digital companies digital nations and I think for all of that those ideas and that enthusiasm to get somewhere. I think like basically we should start with Ethereum itself, right? We should start using these ideas to really try to improve the state of funding and uh maintaining all of this open um software and public infrastructure that the Ethereum ecosystem itself maintains on and basically make the Ethereum ecosystem itself an example of uh this entire like social and economic process working in a sustainable way for the long term and uh working Well, so thank you. Thank you, Vitali.

[Music]