Epicenter - Learn about Crypto, Blockchain, Ethereum, Bitcoin and Distributed Technologies - John Clippinger: Developing a Social Ecosystem of Trusted, Self-Healing Digital Institutions

Episode Date: September 14, 2015

It is often said that technology is neutral. It’s certainly commonplace for Bitcoin to be thought of as such. In reality, technologies may be characterized as an embodiment of human intention, and t...herefore, cannot be considered as culturally or politically neutral. At least, this is the opinion of our guest, . A research scientist at MIT Media Lab and CEO of ID3, a nonprofit which aims to deploy a new generation of trusted digital institutions, John has spent the last seven decades researching the interaction between technology and society. He joins us for a fascinating discussion on the flaws of our institutions in present day society and how we can leverage technology to build a new social ecosystem. Topics covered in this episode: The political values of Bitcoin The idea of cooperative currencies The current state of identity and data ownership, and how it may be improved The governance of decentralized technologies How cybernetics and self-organizing systems may be applied to governance The Open Mustard Seed framework Episode links: ID3 Institute for Data-Driven Design From Bitcoin to Burning Man and Beyond From Bitcoin to Burning Man and Beyond (Amazon) Open Mustard Seed Talk by John Clippinger: Bitcoin and Beyond This episode is hosted by Brian Fabian Crain and Sébastien Couture. Show notes and listening options: epicenter.tv/096

Transcript
Discussion (0)
Starting point is 00:00:00 This episode of Epcentor Bitcoin is brought to you by Shapeshift.io. With no account or signup required, it's the easiest way to buy and sell gems, counterparty, dogecoin, dash, and other leading cryptocurrencies. Go to shapeshift.io to instantly convert your altcoins and to discover the future of cryptocurrency exchanges. And by Voltoro.com, the gold to Bitcoin exchange. Trade gold to Bitcoin instantly and securely starting a digital. 1 milligram. Go to Voltore.com to deposit some Bitcoin and start trading today.
Starting point is 00:01:01 Hi, welcome to Epicenter Bitcoin. The show it talks about the technologies, projects, and startups driving decentralization and the global cryptocurrency revolution. My name is Sebastian Kutu. And my name is Brian Kravind Kroen. We're here today with John Klippinger. He has a, has a long history in technology, in law, and in thinking about these issues. So he is, currently is the co-founder of IDCubed. They wrote a book to publish a book that was called Bitcoin, Burning Man and Beyond. And when we saw the title, someone recommended it to me, it was like, obviously we need to have him on. And he's also a research scientist at MIT Media Lab. And previously, he was at the Harvard. He was the founder of the Harvard Law Lab. And he
Starting point is 00:01:44 was also at the Berkman Center there, which is, I think, quite well known for pioneering some of the thinking about internet law. So, John, thank you. Thanks so much for joining us today. Well, thank you for having me. I'm looking forward to the conversation. Yeah, absolutely. So you've been thinking about some of these issues that have come up with Bitcoin now for a long time. Many of you, of course, will not be familiar with your work.
Starting point is 00:02:11 But can you give a little bit of an introduction about what has brought you here to start thinking about, you know, blockchain and how that fits in? Well, for many years, I've been very active involved in, so the intersection of technology and policy and society. And going back to my college years when I was, helped set up the first sort of protest against the war in Vietnam, I was actively involved in the civil rights movement and worked with the Black Panther program. And I've also been interested in the whole area of development and developing countries. I was a research associate at Harvard many years ago and their program for information, resources, policy,
Starting point is 00:02:55 looking to developing countries. And I've always tried to track and see how technology shapes people and people-shaped technology and how the interaction between the two. And in that interest, I've also did my sort of graduate work and been fascinated by the notion of cybernetics and self-organizing systems. And this goes back into the way back
Starting point is 00:03:15 in the 80s and 70s, actually, how do you design systems that are self-organizing that are goal-directed, that correct themselves, that can learn. And there was a lot of activity then. A lot of it came out of MIT in those years. But the computational power wasn't there really to build systems that could do that. My PhD thesis was basically designing a model of language, language discourse generation, and cognition that was highly distributed and saying how can you control complex, linguistic,
Starting point is 00:03:51 and cognitive process in a high-ecentralized way. So the whole, what's happened in Bitcoin more recently in the arrival of the currency. Again, this is something I've been interested with people like Bernard Lutierre, who does work who is a designer of the euro, written extensively on currency design, and how currencies reflect policies, that currencies are not neutral. with respect to policies. And if you look at how you can affect policy, not just economic policy but social policy, the nature of the currency becomes very important. Hence, when Bitcoin came on the horizon, it was arrived, it was very fascinating to me. And actually, when we first formed
Starting point is 00:04:35 ID cubed, we had a meeting, a dinner at the state house in Massachusetts, where the revolution began and we had a big dinner celebrating the revolution in the financial services industry primarily where we handed out gold coins that said in data we trust and we tried to get bitcoin people to come to our event but that was really early and i think with bitcoin was selling at 12 dollars that but it was it was the idea that you could have a highly decentralized open source incentives mechanism that actually could launch a change and challenge a lot of current institutional structures which I currently believe now are failing. They're artifacts of another era.
Starting point is 00:05:20 And if you look at the institutions that we have designed, there are artifacts in the Enlightenment. When America had 3 million people, not 330 million people, when we didn't have the technology, it doesn't scale other wonderful principles there, but we really have to think fundamentally about institutional design and governance itself. So the idea that you can have the ledger and that you can sign for things without having a third party and you can run scripts and run contracts on top of that that execute and prove certain premises are really important.
Starting point is 00:05:56 When I was at the Berkman Center, started the law lab there, developed a self-executing contract that we took a Wilson City term sheet and made it into a program. and then we provide genetic algorithms to generate new variants of that and prove that you could actually have contracts that are computational approval. Well, that was in the 1980, 2011 and 12, I guess, no, even earlier in that, I think in 2009. So we were anticipating what happened with Ethereum and smart contracts, not to say we're the first, but I think we're on the edge of a very. very profound revolution and how to think about currencies and institutions. So the idea of decentralization distribution was part of it before and then I guess we have
Starting point is 00:06:53 a Bitcoin. Oh, yeah. And then it just sort of all came together, right? Exactly. I'm curious how you achieve that, how do you achieve those smart, well, I guess pre-curses to smart contracts without blockchain technology. How did you achieve the decentralization aspect of it? Well, we didn't, we didn't focus on the,
Starting point is 00:07:10 centralization aspect of it. I mean, our focus was, how do you look at a contract as a program, and how do you make it algorithmic? And so what we did is, I mean, we created a list program in which we represented, sort of the, we translated the terms of the contract into list programs, and so you could change, you could change the language of the contract, say the terms of participating preferred to do the, to common. And someone could see the effect on their, on their capital, table. They could say, oh my God, you know, I didn't realize that when you change this, that my shareholdings go from 30% to 3%. And then we tied it into actuarial tables. And that was another thing that we were able to show that the likelihood, given the, Kaufman did a lot of studies
Starting point is 00:07:59 on this, was that there were very low likelihood that people would be able to get an exit the way they thought they were. So it was an attempt to prove a concept and then be able to, and then be able to write a program that was provably complete. And using genetic algorithms, which is a nice feature to that. That was really cool. We could generate a lot of variance, set fitness conditions on the outcomes that we wanted and see how we could approximate a solution
Starting point is 00:08:27 that took into the different intentions. So you mentioned in the beginning that your focus was sort of, has been on the intersection between technology and policy. and when I was reading your white paper, one thing that struck out to me is that you seem to think of technology as something very political. And you often hear the phrase that technology is neutral. And especially in the Bitcoin space, many people say like Bitcoin is neutral. So how do you think about these issues? Well, I think that technology, I mean, and there's some people that have written some, I mean, Kevin Kelly's have written some great pieces on it.
Starting point is 00:09:09 Brian Arthur, who's an economist, has written a thing, sort of as technology as almost an evolutionary force, we co-evolve with technology. So I don't think we can disentangle ourselves in technology. Technology reflects human intention and human understanding. And it also reflects the relative power equations of different people with access to resources. I think when you look at the Bitcoin and the Bitcoin Protocol,
Starting point is 00:09:39 that inherent in that is a certain assumption about a public and private goods. And this is one of my concerns about the community and understanding that these problems have been thought about for quite some time. And so the difference between a public and a private good, a difference between a one-time game and a multiplayer game, you start to look at game theory and you start to think, well, what, What forms of cooperation were possible in order to get a stable outcome and for a preferred distribution of goods or services? And the Bitcoin reflects a particular private goods model, so like a one-time game. And it's very sort of in your Chicago school.
Starting point is 00:10:29 It reflects that mentality of impersonal exchange, one-time game. It's a one-time player. And it's an incentive that is sort of a zero-sum game. game. So, you know, each time you're doing the mining, you don't know the identity of the person, you assume that everyone's out for themselves. So there's a high trustless, I mean, you don't trust anyone, therefore you're going to, you need to create a mechanism that verifies the consensus, the proof of work that verifies the transaction. But there are other incentive mechanisms. And I should also say that the Bitcoin itself is sort of like, it's officially a demand-backed
Starting point is 00:11:17 asset. That's how the U.S. Commodities Exchange defines it. It's a very fascinating definition because it's not grounded any real asset, but it's still sort of a gold bug kind of thing. And I think that was the intention. And you're going to limit it. It's deflationary. And it can be highly speculative in that way.
Starting point is 00:11:40 So it reflects a particular view of a class of currencies. There are other kinds of currencies that are different. And this is true throughout history. You have like two different currency models. Typically, one's gold, the other is silver. And one is to create liquidity and create interaction. But Bitcoin, in sense, is built on the sense that you don't trust a central authority and that you have to limit the issues of the coin.
Starting point is 00:12:05 And so it is sort of rival. I have someone else, but it's not, the more you create, the more deflation or less value it has. But there's thinking about digital information-based currencies that are what I call cooperative currencies. And this is based upon a lot of people's work, but I would reference Brian Arthur, who is an economist. And I works at the SNAFE Institute and talks about the theory of increasing returns. And that's a different model. It says information, when you're in a digital world, you know, that the more the information is disseminated, the more people possess it, the more value it has. Whereas in a physical asset, say, is gold, it's rivalous.
Starting point is 00:12:48 So you're creating shortages. You're creating, you're speculating and shortages. And you make money off the difference between the supply and demand of the shortage. Where in the information-based currency, you actually want wide acceptance and the broader, that. information flows the greater value for everybody yes I saw that as well as well that was quite striking how you talked about that but don't you think that for money or a currency to be valuable it has to be scarce no I think it has to have inherent value I also think and actually I made the point in the white
Starting point is 00:13:23 paper and Bernard the he's a currency designer they will say you need both and the analogy I make I think is is in the paper is to, I mean, the, I don't know people are familiar with it, there's a program called Deadwood. It was sort of like the settling of the West. And when you didn't have rules and you didn't have, and it was sort of the survival of the fittest, you really, you want, you want to have a currency that is like Bitcoin, it's like gold. You can trust it in its hair itself. You don't, you don't depend upon a party to accept it. It's a one-time game. It's a zero. It's a zero. It's a zero. zero-sum game. I either have it or you have it. But it's what happens is you start to get repeat
Starting point is 00:14:09 games and then you start to get ideas of boundaries and the sheriffs and you have different ways of creating cooperative behavior. Then you want another currency that is cooperative and increases by the value of use. And this is where you're another, and this is, there's a lot of stuff in game theory on this is that when you have a repeat game and you know the person you're working with and you have a reputation metric, you can create higher outcomes for everybody doing that. And I also cited the work by Martin Novak. There are a lot of work that's being done in evolutionary biology that says this is also true of biological systems, that you have higher outcomes, the evolutionary stable solutions,
Starting point is 00:14:54 when you have some kind of collaboration through known reputational identity, you have to have the reputational. And then you do with a sheer, a sort of one-time game rivalous outcome. What would such a currency actually look like, though? Well, I mean, it's something where there are a number of people who are developing these things. And actually, Prima Vera is, we had talked about earlier, has some ideas of this. But we're looking at something like an artisan coin, where artisan currency, where, and I've been
Starting point is 00:15:31 referenced a Bitcoin and Burning Man, that when I was in Burning Man and talked to Larry Harvey, and he was talking about, well, he would like to have a gift kind of currency
Starting point is 00:15:41 that would be, in other words, people acknowledge what other people do and they get, they accrue some kind of credits for that. And those credits are, can be, they can be exercised
Starting point is 00:15:54 in the context of a, of a different kind of social contract. So if you go to Burning Man and you'd have all these people working together basically for a public good and you have you have mechanical engineers, you have carpenters, you have all these people coming together, builds up. And what we'd like, Larry was saying a lot of people would like to acknowledge it. So it's like a gift exchange currency that and that people were able to acknowledge what other people do. And you can actually develop so, social capital that way. Well, that's a...
Starting point is 00:16:31 And in order to do that, you're looking at something as a repeat game. It's not rivalous. And when you look at an artisan economy, you want to buy something not because... You don't want to create shortages. You don't want to create shortages. You want to go back and you want to have a repeat experience with someone who you value for what they create. It's not a standardized product. It's not a... What you're valuing is what is unique about it. And you want to be able to then bring other people to appreciate that you need this and it's not a you're not rivalous you want to work with them to be successful and their success is your success so it ties into empathy and shared theories of mind
Starting point is 00:17:09 that's another area a lot of work has been and done in evolutionary biology that that's a critical aspect when people have shared theories of mind then their coordination costs and their social value create is vastly increased but do you think that's compatible with the i guess current state of the world of the economy and how in large part we're moving away from artisan economies. Well, I mean, I think we're actually moving back to that. I think I think you're seeing a failure of the, I mean, the classic failures in Wall Street. I mean, is a case of point. I actually think we're going to move away from that. And I, and I'm not saying that you're not going to have things like a Bitcoin or a gold base or a limit. I think you're going to have a
Starting point is 00:17:57 combination of the two because you're always going to have cases where people cheat and fail one another. So you have to have to have something that is sort of trustless. But the benefits of having an identity and having a reputation and being able to distribute and create resources around that, I think far outweigh the artificial creation of shortages, which really don't exist. They're choke points that benefits some parties. So that's where your power laws come in. And that's why you have this huge disparity of wealth and that the genie index kind of things
Starting point is 00:18:32 accentuating because people think, oh, that's a natural consequence. You've got preferential attachment, therefore you get power laws. We can't do anything about that. I don't agree with that at all. Today's show is brought you by your friends at ShapeShift. Of course, Shapeshift is the fast and EWA trade all coins and then out support over 45 of the most popular altcoins, including counterparty, ripple, gems, and so many others that I've never even heard of. And they now support, you guessed, that ether, since the release of the frontier launch
Starting point is 00:19:01 shapeshift has added ether to the list of supported currencies. And so now you can buy ether and start building some sort contracts on Ethereum. Now there is a small caveat, and this is true for every exchange at the moment, is that it can take up for six hours for you to get your transaction confirmed. But just stay patient and you'll get some ether and you'll start building some really cool smart contracts every soon. And just as if ether wasn't enough, they also doing this cool thing with a gaming company called Spells of Genesis that's putting some of these in-game assets on on the blockchain and one of those they're selling in a very limited amount like 25 every week only on Shapeshift. So that's called the Wander and it's called Spells of Genesis.
Starting point is 00:19:43 So if you're a gamer, if you're interested in that, check it out because of course it's a very interesting way to make this sort of fake scarce asset actually scares by using the blockchain. So head over the ShapeShift. I'll give it a try. We'd like to thank Shapeshift for their support of Epicenter Bitcoin. I guess what becomes very central in what you're talking about, then, though, is the question of identity, you know, because if you have an anonymous party that can just join,
Starting point is 00:20:12 well, you can't. Exactly. Like this abuse can happen. Exactly right. So all societies, when they really succeed, I mean, people internalize, and this is one of the points I tried to make, rather than have a state-imposed it,
Starting point is 00:20:27 instead of norms. Basically, you want people to internalize and generalize certain codes of conduct that it's in their interest to be adhered to and hold. So when you have really strong reputational identity signals that you can be relied upon, then you eliminate enormous amounts of inefficiency in this society. You don't, and you actually create much more abundance.
Starting point is 00:20:48 So the identity component is very, very important in my mind. And I think that the same time is the fact that people, this is where the notion of self-sovereign identity becomes very important, is that I can, he enrolls controls. So if the government issues the identity credential or the bank does it, then basically they're the control point. And so we're talking about a new kind of third, a new kind of third party organization. It's not private sector, it's not public sector. It's more like an open source sector. And as a set of protocols, I can go through that I can, I can, I can, I'm going to, I can a, establish my identity, establish certain certificates that other people rely upon. They'll take that
Starting point is 00:21:26 risk. And I can build upon that. And I can control, I have control of their own data that I can create legitimate credentials to express who I am that other people can rely upon. And that flatens the authority structure that we currently have that's very static and is like tens of thousand years old. We're still running society on a patriarchal lineage system that goes back into our, our, deep pass. So if I can oversimplify that model, then rather than having, say, the government issue identity cards, and that becomes your sole means of proving your identity. You have multiple sources which can together, come together and form a consensus around
Starting point is 00:22:09 who you are. However, I mean, identity on the internet is something that is quite malleable. Well, you can make it non-malleable. I mean, I mean, you can have multiple identities. Everyone's going to have, and you... Right, that's the point I don't want it to make. So then does that... Does the fact that we can have multiple identities
Starting point is 00:22:32 not cause a problem for social order? Yeah, it does. That's why you have to anchor it in a root identity. In other words, you can't let someone game the system. I mean, and this is, you know, this is one of the problems where they had going way back to PayPal and things like that. That you can, you know, you can game the system, and you can get people to trust in a particular persona identity,
Starting point is 00:22:52 and then you can change and screw them. So you need to anchor it in some, I need to anchor my identities in a root identity. What we mean by root identity is something that's tied to me to my biological physical cell. Now, I don't need to have that identity be fully disclosed. I can have it be authenticated and anonymous. And I can, so I can have a persona out there.
Starting point is 00:23:18 that has certain verified attributes but doesn't reveal who I am on a PII basis. I can reserve that. How would you tie your physical identity to these other identities? What you have to do, and actually this is some work we're doing on, you need to get a collection of unique identifiers. It's almost saying that I as an individual have certain unique patterns that distinguish me from anyone else. So those can be biometric and behavior metric patterns.
Starting point is 00:23:53 And in combination, they create something that I don't know if you're familiar with like an eigenvector that uniquely identifies me. And I can, you know, and I can encrypt that and I can verify that and I can prevent and use that as a core credential. How do you prevent that input data isn't forged? Well, I mean, there are, there are, there, you. You're always going to everyone, it's an arms race in how these things happen, we all know. But what you can do is mitigate the vulnerability of having multiple dimensions of these things.
Starting point is 00:24:31 And they can be updated in real time. The other thing that we're learning, and this may reference an earlier conversation, is the idea of cybernetic versus models of software models, is that we're constantly getting data updated all the time. so we can compute our metrics, we can compute our distributions on the fly. And we can have one-time tokens. We can have, and we have highly decentralized and distributed system. It's very difficult, actually, with these new kind of architectures to basically spoof them.
Starting point is 00:25:11 It's not so it can't be done, but the cost goes up dramatically, particularly given what we have now, which is totally archaic. So where do you see, let's project like 10 years in the future, what do you see the state of identity being then? Are we still looking at a system where we have a higher core? I don't think it's going to be higher coal. I think this revolution is imminent. I really, and I think Bitcoin is a huge, I mean,
Starting point is 00:25:43 is a huge vector for that change. and I think that, I mean, I was just talking to the CIO of the new CIO of the federal government, Tony Scott. And he basically buys into the whole idea of a decentralized blockchain-based services for identity and for government security. And they have something called security by design they're looking at. And that's a very decentralized system. So that brings us to the topic of governance then. There's been the rise of decentralized technologies, the recent years, especially due to the rise in Bitcoin. And in those circles, many have suggested that governance could be decentralized,
Starting point is 00:26:30 becoming more and more obvious with the recent release of Ethereum. And people have speculated that governance and institutions could be decentralized. What are the first things? I mean, we always hear these ideas, but with no real concrete. Right. What are the things that could be decentralized? So where do you think the first areas of decentralized governments are going to come from? What are the first things that governments are going to start to decentralize essentially?
Starting point is 00:27:04 Well, I think that one of the things that you can do in a decentralized manner of so much more effectively, and not only in terms of coordination costs and expense and everything else, is, and this is where the blockchain is so amazing. I mean, you look at remittances and payments and being able to verify transactions and identities and do that through smart contracts and be able to hold money or hold identities or hold some asset in escrow and release it when provided the appropriate credentials.
Starting point is 00:27:42 That is so pervasive. And I think when you're looking at the, and I've worked with a number of banks, and they totally have to rethink what they're doing. And what happens is you have a regulatory structure right now in a banking structure, it's a hub and spoke model, and there's this impossible preponderance of cost for financial regulation that makes the regulations.
Starting point is 00:28:07 By the time they're written, they're out of date, They're a huge cost. They only served in many cases to help the incumbents, but yet it increases burden. The fact that you can decentralize the payments and remittances systems in a distributed way and have oversight, then, basically, you can use smart contracts, you can use multi-party computation methods, the variety of ways in which you can do it. That's going to unlock so much value and have so much savings. It's just, it can't be resisted.
Starting point is 00:28:40 I mean, I look at this as very analogous to what happened to the newspaper industry, say, 15 years ago. They, you know, they'll, oh, no, will not affect us, we'll become portals, we'll adapt, no. They could not adapt. It's a very different model, so it really affects them. And so that's, you know, that's the thing that I think you're going to see. And I think another area that is interesting is the whole idea of location-based data. internet of things. And I mean, how are we going to scale internet of things without having some kind of
Starting point is 00:29:13 decentralized autonomous contracts into that? You know, I think that's an inevitability. And I have dealt with a number of, say, companies, car companies, they're capturing all this location data and things like that. And the question, they're a breach of the EU data director. But if you have API calls and enforce the... the regulations through API calls, then you don't have to have a high, you know, governance. It can be done very well.
Starting point is 00:29:45 And I think that these are the areas where you can see if some of the innovations coming. There are all sorts of companies that are working on various aspects of this. So you mentioned the newspaper industry. And they did have the monopoly of information, but they didn't have an imposed monopoly. that monopoly could and has been dethroned. The difference with governments is that the monopoly is imposed by power. Apart from the cost savings potentially, which I think are pretty obvious, don't you think governments would have, and institutions would have resistance to decentralizing their institutions
Starting point is 00:30:27 since they would essentially be losing control? I don't know if they're ready for that paradigm shift. Well, I think what you're seeing is a failure of traditional governance institutions. So they're genuinely looking, they're losing legitimacy. I mean, this is what this election is all about, whatever I would call it. And so I think, and what you're seeing in certain cities is, I think, and this is what's interesting, there's a lot of movement in the cities, and there's, I think, the maker movements in cities, the idea to get citizen participation.
Starting point is 00:30:59 We talked about, I'm talking about in several areas, what I call governance-free zones, but it's like enterprise-free zones, allow people to actually start to take over and run some of the local services and agencies. There are housing authorities, and there are whole lots of traditional government functions that are performed so poorly. I mean, Tim O'Reilly talks about this as government is a platform and that you should allow people, you know, allow some API access to different services, but actually in the delivery of the services, too. And you can start to, you create competing institutions that can compete with a traditional government way of dispersing benefits. Actually, in the UK, they're looking at this, developing a Gov coin for dispensing of benefits. And so you're creating a new kind of currency
Starting point is 00:31:54 around benefits that can only be used within a certain context. And that creates a, different kind of micro economy. And people will start experiment with that. And I think that's going to be much more effective, more responsive, more savings, more engaging participatory. And what will happen is the governments will say, okay, we're going to move up. We'll become more meta. We'll let these decentralized networks provide more of the function.
Starting point is 00:32:20 And we'll set certain ground rules. And actually, there's something called Instict, which National Strategy Retratured Identities, the cyberspace that came out of the bomb administration, they were saying, yeah, maybe around identity and personal data, the government doesn't try to fear out at all the details. They just set up the meta rules and provide safe harbor provisions, allow people experiment in it and try to establish those metals. I think that's what's going to happen.
Starting point is 00:32:52 Do you think that this change, so you mentioned that the government's doing some of this change, Do you think that's more the way it's going to happen? Or do you think that people will just sort of create these quasi-governmental organizations using blockchain that then will take on some of the functions, but without actually the permission or endorsement from proper governmental bodies? Well, here's what you're seeing already. You're seeing this already.
Starting point is 00:33:18 If you look at Uber and you look at Airbnb, you know, Uber is what, $45 billion, they're breaking all the rules. and and but they're they've achieved such scale and they actually probably benefit that people want and there's a whole set of services that that are going against the traditional government models and regulations they're forcing the government to change and and I think you're going to see that pervasive so the government's going to have to adapt to these changes and it's going to have to reframe its role. Not to say it's going to roll over, not to say there's not going to be a lot of resistance. But I think in the U.S., for example, I think there's a huge disaffection with the
Starting point is 00:34:04 role of the federal government. It's just it's not that these things shouldn't be done. I mean, this is the tension between the progressives and the Tea Party and the grassroots and the bitcoins and the libertarian view. I mean, I think in the extreme libertarian view, it's like there's, and that's sort of the Chicago school, there's like no oversight. Do not tread on me. You know, I don't want, you can't tell me what to do. And there's no sense of communitarian responsibility. I think that's extreme.
Starting point is 00:34:36 I do think they're going to come, you're going to reframe the problem. It's not going to be a big government solution. That just doesn't work. That doesn't scale. But you have to have governance. And you have, and you're going to, governance becomes a kind of technology. Just imagine if you're able to have governance. as a technology and that starts to innovate the way you have in Moore's law. That just changes everything.
Starting point is 00:34:58 And I think there are problems, fundamental problems that have been around forever. Like who guards the guards in this whole idea that can be solved. I really believe they can be solved through zero knowledge proof and new kind of technologies that will make that maybe not solve all the cases, but 80-20 and make a huge difference. So we're out of three. of being able to solve some very intractable problems about how people run, govern themselves. And I find that very exciting. One of the things you mentioned that I thought was a great point regarding that is that if you have a regular government body or some sort of institution, right over time, they start having employees and their rules. And of course, those people started having their own interests.
Starting point is 00:35:48 And all of a sudden, you know, when it was just supposed to provide. the service, now it has its own interest. And I think, you know, in Europe, for example, with the EU and the ECB and all those, one sees that extreme strongly. And I think in the U.S. is also a great example where the government has enormous power. And I think it's interesting to think that, oh, maybe with decentralized autonomous organizations and things like that, right, where there's sort of transparent rules. there really isn't something in the middle that started taking on its own interests.
Starting point is 00:36:26 Maybe we could get rid of some of that corruption. Absolutely. I think that's hugely important. I think what happens with bureaucracies and government is they become their own interest group. And then as their own interest group, they allocate resources and budgets in terms of perpetuating themselves rather than their constituencies. And I think we need to open that up and decentralize of that and actually make, I mean, it returns back to a notion that, well, maybe those functions can be distributed out broadly in the community. So the way you do it in an open source movement, there are people that have certain, this is where credentials and reputation becomes really important. So if I had a really strong set of credentials that were established, and I could then solve a, quote, government problem.
Starting point is 00:37:16 And collections of people could do that. So it doesn't have to be this hard divide between what's public and private sector. That's why I think that's a false dichotomy here. And not just public and private sector, but what you call in your white paper citizen sourcing, and it's what you're mentioning that, right? Yeah, I really think citizen sourcing is a real option here. And it's just not a do-good, feel-good. I mean, you can have a high-level professionalism in a citizen's competency.
Starting point is 00:37:47 And that is the way, I mean, look, the Athenian democracy, when it worked, it was a very short period of time, but actually republics when they worked, that's the way they worked. And the best part of the U.S. in its early phases, that's the way it worked. And even the early Roman Republic was that way. You had people take an active role in government, and it wasn't a, and they weren't there to lard their nest. I mean, what I see, this is another generation.
Starting point is 00:38:16 when I was I did work in Washington and helped set up an agency there and you went there with the intent of giving back. It was a public service thing. It wasn't a way of bootstrapping your career. Now it's just like Washington is owned by K Street.
Starting point is 00:38:32 I mean it's just a fact and the Supreme Court it said it's fine, you know, go ahead. So this is United ruling. So it sort of really is undermined a core democratic principle. You're listening to this show in parts thanks to all the support we get from our great sponsors.
Starting point is 00:38:52 We have some fantastic companies who work with and we're very proud of what they do and of their products and the companies we're happy to stand behind. Every week we reach a large audience of thousands and thousands of people deeply involved in the Bitcoin cryptocurrency space, people running companies, people having a lot of influence. And people who are just enthusiastic about technology, about progress and about the future. So if you're involved in a project or startup and think that reaching our audience could help you, email show at epicenterbitcoin.com, and let's talk. Now, with the idea of citizen sourcing, this has come to mind. So I mean, I agree that given, for instance, let's say you have a neighborhood and there are issues or problems in that neighborhood, who better to solve those problems than the people actually living in it than perhaps a local government or federal government or what have you coming in it and trying to solve those. Where does that citizen sourcing stop? Because it seems like there seems to me that you could have a risk there if you start getting into, you know, citizens ensuring security, citizens policing, that sort of thing where you could get into some sort of touchy situations there.
Starting point is 00:40:03 Oh, really good question. I think, I do think that, you know, you have to think of an architecture or governance and accountability and transparency. So you really need the architect institutions to perform different functions. And you need to hold them accountable. So when I'm, when I think a citizen source, it's not like crowdsourcing, just throwing everything out. No, I think there's a set of different kinds of institutional designs and roles for that that you need to establish. And it's an opt-in system. I mean, so, and people have different preferences as to how much and how little they want to be governed and this and that.
Starting point is 00:40:45 and their natural differences in personality and disposition. But I think there are ways, it's an empirical issue, is how to design certain kinds of efficacious institutions. And so if corruption's an issue, well, then how do you, if you're using certain forms of encryption and authentication and you can do chain of custody on data, and you have ledgers that you can audit, and you can rotate in and out people in positions of power in a non-gameable way,
Starting point is 00:41:22 and I think you can't, then you can create more effective kinds of institutions of oversight that themselves learn rather than try to arrest development and try to protect a special set of interest. And this can be an area of genuine innovation, experimentation. And, and, And I think there are certain empirical solutions that one can test. Have we seen this before in history? There are examples of this sort of citizen governance? We're moving into an era, you know, whether you believe in the singularity, literally, we are having a convergence of technologies and how we think about technologies is how we build our futures.
Starting point is 00:42:10 And that's never happened before. I mean, we can, you know, we can hack the human brain. We can hack our genome. We can hack everything. And that's only increasingly true. So that's when I, and the paper I made a big emphasis is, is that to understand the magnitude of those changes and put them in a historical context, the kind of things that we're doing, not see it in such isolation of a particular expertise. But we are, we are, things are up for grabs. Yeah, I think that's also one of my expectations and the hopes is that when you start having blockchains maybe integrated in all kinds of areas and all kinds of systems.
Starting point is 00:42:54 And initially, of course, often it will be because there's a commercial benefit to it, right? It's cheaper. It makes something possible. That wasn't possible before. But then that over time, actually, the effect, and that might even be a secondary effect often is that actually reduces corruption. It reduces some of these. flawed incentives that we have today. Oh, yeah. I mean, I don't, I don't look at it as a secondary intent. I look at as a design point. I mean, that's what interests me.
Starting point is 00:43:25 I think it's a design point of blockchains, but when people will adopt it in a commercial setting, you know, they might not do it for that reason. That might be a side effect. But I do think that the need to have things that are, I mean, trust is important in commercial settings, transparency and predictability, and how do you manage risk? I mean, those actually all create value. And I think the ability to create systems to perform that way, I mean, if you have a highly trusted, highly learning system that is rapidly evolves and changes and is able to maintain high levels of sort of value creation versus one that is over. maybe not governed but hacked and porous, it's going to outcompete that.
Starting point is 00:44:20 I just, I just think we're at a point of different kinds of competing institutional economic designs. Today's magic word is mustard. M-U-S-T-A-R-D. For the let's-talk Bitcoin.com to sign in, enter the magic word, and claim your part of the listener award. So this ties into a lot of your work regarding cybernetics and AI. I've often had this sort of idea that, you know, smart contracts, although they are very primitive individually,
Starting point is 00:45:04 together when you have multiple smart contracts, it starts to resemble a living, multiple smart contracts that are interacting with each other and doing all kinds of things. It starts to look like a living organism. I agree. Is that how you see it? Oh, I definitely see that. I mean, I think we're, I mean, and by a living organism, and that's where cybernetics is so interesting.
Starting point is 00:45:26 And people like people are not familiar with cybernetics. But the whole idea is gold, I mean, it was a theological system. It had a goal that had feedbacks and tried to maintain certain equilibria. It could be dynamic equilibrium. There could be multiple systems working together. But it's like, well, how do you? How do you approximate living things through cybernetic mechanisms? And I've always been interested in the work that's done by the Sine Faye Institute,
Starting point is 00:45:53 in which they've really started working on this 25 years ago. And I think what you find is the notion of cybernetics is sort of mutated in the notion of complex adaptive systems and which John Holland's work. And that's gotten very sophisticated and applied to lots of different fields. I think there's a rich body of knowledge looking at things this way and a lot of it does come out of the Santa Fe Institute and looking at social systems this way.
Starting point is 00:46:23 These guys work of Sam Bowles, Herb Gintas. There's just a really fascinating and very rigorous ways of understanding social organization and how people adapt and change and don't have changed and what are the factors that affect them. So we're getting a very sophisticated understanding understanding of sort of adaptation and evolution and how different kinds of organisms adapt. And I'd recommend the work of Martin Novak, who runs the evolutionary dynamics center at Harvard
Starting point is 00:46:55 and his group as a whole. It's very rigorous work and it can be applied to designing social institutions. I mean, we're all part of the same evolutionary process. Looking at it from that angle, of the study of self-organizing organisms, what do you think of the design of Bitcoin or the way it functions? Well, I think that was, you know, I mean, it was really brilliant.
Starting point is 00:47:22 I mean, you just, I think that, I mean, when you consider all the things that were pulled together and there was a, there's a, it seemed to be a single person's effort and to get the incentive mechanisms right, and then be able to launch something like that. I think it was just extraordinary. And it's sort of epical. I mean, and I think it, I'm not sure how you could have done it.
Starting point is 00:47:50 So I may disagree with certain things, I say, but that's part of the evolutionary paradigm of it. And, you know, there is a lot of debate about, you know, private blockchains, public blockchains, what kind of currency design, what kind of incentive mechanisms or consensus mechanisms you're going to use. I mean, it's broadened that discussion and whether proof of work is really what's required, particularly an ecological way.
Starting point is 00:48:14 And if you have identities and you're going to have whitelists, then do you really need such an onerous consensus mechanism, that kind of thing. But the fact that something like that took off, I mean, it's, but it's the same, it's the same basic phenomenon that you have, you know, in any kind of peer-to-peer distributed system. And so I, you know, I lot. I mean, it was just, it was great. It changed.
Starting point is 00:48:44 It created possibilities where there were none, say four or five years ago. I don't know if you've seen, here I was your thoughts on this. I don't know if you've seen this really fascinating talk by Mike Kern that he gave a couple of years ago at Google, exploring the idea of autonomous vehicles and how you could have autonomous vehicles that were basically a public good and responding. to markets and going where there's demand and then perhaps like shutting off for a while when there's no demand or moving to another city and the idea that those vehicles could essentially run themselves and then the question is, you know, who owns the vehicle?
Starting point is 00:49:24 Exactly. If you have a market of these, it could be vehicles, it could be drones, it could be, you know, whatever, does that constitute an artificial intelligence to you? Is that just devices responding to programs that were made by humans? I mean, yeah. I mean, and I'm very, and I just give a keynote on the Internet of Things. A group of Massachusetts was focusing on that. And I think that the Internet of Things, what is, you're having more autonomous things.
Starting point is 00:49:56 People think, oh, you've got to have people in the loop all the time. You can't scale it and have people on this. So therefore you have to allocate certain agency to the device. That's inevitable. And then the question is, to what extent, is that device acting on your behalf? And if it is not, then how do you revoke that set of privileges? And when you start to think about the Internet of things, I mean, these things are pretty smart things.
Starting point is 00:50:21 These are arm chips. I mean, that's sort of the basic, you know, the 64-bit arm chips. You're having things that, and then they can tie into a cloud. So if you look at computing, I just was looking at some of these new drones. and you can have autonomous drones. And now from an AI point of view, which blew the top of my head off of, was the problem with AI in my generation was it was very rule-based and very brittle. And so you didn't have a lot of data coming in,
Starting point is 00:50:51 so you didn't have the ability to do dynamic modeling and verification and do real-time computation. So there was everything that was tried to capture a brittle set of rules. Now that's not the case. And so you can have drones flying in the wild and they can know what a window is. And there are also things that they've never seen before that they can make reasonable judgments about. And then you can tie into the cloud. So the drone is as smart as the cloud.
Starting point is 00:51:21 And these things can work in formation. And of course, when you weaponize them, you're creating quite a foe. And we have not thought that through. And one of the areas we've really not thought through, and this is where people say, oh, personal data, privacy, who gives a damn. These are who you are. Data is a new asset class.
Starting point is 00:51:44 And it's got to be signed and verified. I mean, the military is running into this issue, and that they don't know whether they have control over a drone or a satellite. Because they can't sign for the data that goes in it. They can't verify their providence of it. And that's happening in spades throughout the net. If you look at your home, it should be your castle.
Starting point is 00:52:10 It should be the heart of privacy. Well, you've got a Samsung TV in there that's watching everything you're doing and doing facial recognition, voice recognition, and feeding it back and selling it. It's a surveillance device. Same thing with Nest. And now, there's no design point that gives you control of that data and what you want them to see. You can't intervene and control them. And that's just going to escalate.
Starting point is 00:52:35 This is not a good situation. It's time for our word from our sponsors, volatoro.com, the goal to Bitcoin Exchange. Now, we all know there's been no shortage of Bitcoin exchange scams and hacks in these recent years. And that's why when Philip and Josh, the two brothers behind Voltau,
Starting point is 00:52:55 decided to start that exchange in the wake of the Mount Gauks disaster, they wanted to do things differently. So they're really pushing the bar forward and innovating in terms of security. transparency and auditability to ensure that customer funds are safe, secure, auditable, and so there's no, there's none of this Mount Gox stuff going on. It's all on the up and up.
Starting point is 00:53:19 So if you've been listening to the podcast, of course, you know Voltaur and perhaps you like Voltour and you like what they do. Well, something new is happening, something really exciting is happening. And that's Voltaur is doing an equity crowdfunding campaign through Bank to the future of Sam Dixon, of course, you know as well by this point. So if you're interested, you now have the chance of actually owning some equity in a startup, which is sort of a new thing, equity crowdfunding, and not just a startup, but a great Bitcoin startup, a great startup in the space.
Starting point is 00:53:49 And you can even invest with Bitcoin. So if you're interested, check it out. That's on Bank to the Future. So bNK2TheFuture.com. And of course, we'll put a link in the show notes. And we would like to thank Vultura for their support of website. of Bitcoin and hope they're going to have a fantastic crowdfunding campaign. So you've worked on a specific software project too called Open Mustard Seed.
Starting point is 00:54:15 Right. What is Open Mustard Seed? Open Mustard Seed. It was an open source project that we started about four years ago. And it was in an effort to, and actually there's a video on it that pretty describes it. a set of notions to give people control over the personal data and be able to have the aggregation of a point of personal data and create trust networks whereby people could have a principled way of sharing and releasing that data. And the key element here is that in a classic internet notion, the control should be at the edge of the node.
Starting point is 00:54:52 And the edge of the node now is now an individual in data and computation. and at that point that you should any other you as an individual should have control of that I also as it evolved we had a developed we call windhubber principles around personal control of data and sovereignty of individual identity in other words you should have not only control your data but you should have a control of your identities and by that I mean that in the classic notion he enrolls controls, which means that the government or the banks or some of authority enrolls you, and they basically control your identity and tell you who you are. This is not something you control.
Starting point is 00:55:40 In our model, you have an open source, you have a protocol whereby you control it and you assert it and other people to accept that. And on the basis of that, then you can decide in a peer-to-peer way how you want to share your data. Now, we worked on post-Mustards Open Mustards Sea. We worked on a peer-to-peer terms of service agreement where everyone in the network is in parity with everyone else. And that other party, that other entity could be a company. But I have the right to extend and revoke my data at any point. And so I can use access tokens.
Starting point is 00:56:15 I can do a way of one-time tokens for the release of the data. I can verify who's gotten it and have it signed for. but the aggregation point is the individual. And that gives you a 360 degree view of, and therefore, if I'm going to verify that, it becomes much more valuable to all parties' control. That sounds, of course, fantastic. And we've talked often about, you know,
Starting point is 00:56:38 topics and ideas like that on the show. But at the same time, it seems like where the world is going is not that direction, right? You have larger and larger institutions gathering more and more data. And, of course, very often their commercial interest is very much tied with the ownership of that data. So do you think this really has a chance? Yeah. I mean, no, I've been pushing this rock up a hill for 10, 15 years.
Starting point is 00:57:04 And I can, so I've seen a whole change in attitude over the last five or six years. And I'm also on the World Economic Forum. We came out with a notion of personal data as a new asset class that people should have control over it. That was adopted there. There's also, there is the consumer data bill of rights in the U.S. government adopts that position. So does the EU adopt that position. And now even someone like GMSA, which is the trade association from mobile phone operators, believe that people should have control of their own data.
Starting point is 00:57:38 The fact is they want to be the people that host it and monetize it. But there's a recognition that, in a classic Internet notion, that if you can push it to the edge, you can create more value in that because you're able to have a natural aggregation point. And what you're finding at companies like TAR, you know, all these data breaches that have happened is create a liability for companies in holding the day. They really don't want to have the PI, the personal identifying data.
Starting point is 00:58:08 They would like to have access to rich, verifiable data that can build a relationship with a customer. So if you give them API access, in a protected way, in a better way, and they offload the liability. They're very, very happy with that. So I believe that there are strong incentives for major companies, and I've talked to them.
Starting point is 00:58:28 I mean, I've talked to a lot of banks agree with. A number of banks bring with me on this one. I can't say which ones, but huge banks. And they welcome this. Also, major brands welcome this. So I think this actually is where it's going to go. Yes, you have the Google's, and you have the Facebooks that monetize your data.
Starting point is 00:58:49 And this is where the big battle begins, because you can see where Google, and they're collecting all the data about us. And I have no control over that. They're going to try to be an identity provider, and they're going to try to monetize it through advertising. I think advertising is a dead model, too. And there are a lot of people in the advertising to acknowledge that.
Starting point is 00:59:07 Because once you know a lot about people, and the impedance matches that actually, if I get enough data, I can get people what they want. that's a different model than advertising where you're sort of scattershot and you're making money on just being able to spam people. And you get all these false hits. Now, I think that's going to change dramatically. So I think you're going to have a different business model. And so when you get a different business model, a different value proposition and different economics, then you're going to have a shift.
Starting point is 00:59:40 So in this model, maybe I didn't understand really how it works, but how do you prevent someone, who has access to your data, to whom you've given access to your data from copying it and keeping it for itself? Well, I mean, there are technology where you can give them, they're various, there are about three different ways you can do that. And I think that there's one model, I mean, this is something we do, the MIT will say you have, you ask questions and you don't get the data, you're just asked questions of the data.
Starting point is 01:00:14 So you really don't possess the data. There are ways in which you can actually have a temporary hold on the day and then bring it back. And there are other ways in which you can actually then do it through, and you can actually do it through in a contract. So you're saying the part of your agreement, in terms of service agreement, is any party can retract the agreement and data. And we assign for that on the blockchain. So you can have that verified.
Starting point is 01:00:39 Now, if a party breaches that, then they can be expelled from the network and they can be subject to certain kind of remedies. So I think there's a combination of these things. Yes, you can always get some bad actor that's going to hack some kind of weakness in the system. But the fact is I think that you can increase the cost of that in such a way that you can minimize, you make it an edge condition rather than an essential condition. That's an interesting thought to have a terms of service on your data because there's terms of services to use Google, to use iTunes and essentially we are subject to their terms.
Starting point is 01:01:20 But I agree that it would be great if we could have our own terms of service. That's exactly what we have. You can use my data like this. You can do this with it. Right. I've been working in that for many, many years. But we've actually now, it's a peer to peer terms of service agreement. And so, you know, and it's enforced through a smart contract.
Starting point is 01:01:40 And so if a person's a bad actor, you know who, this is it. You know who they are. They can be a huge reputational hit, and you can do what the Greeks did. You can throw them out of the city, you know? That was the, when you expelled someone from the city when they're a bad citizen outside the wall, that had a huge cost.
Starting point is 01:02:01 And you can do that through contract law. This is the interesting governance mechanism, particularly in a common law jurisdictions. You can do that. You know, that sounds extremely exciting. And yeah, I mean, I do think if you start having these like blockchain-based, maybe systems where it's like reputation on both sides and maybe things like that become possible. So you mentioned before that you were also working on a startup.
Starting point is 01:02:27 Can you share about what the goal is with that and what do you want to do? Yeah, I think what we recognize is that it's very important to establish an open source component to this new ecosystem. And, you know, one of the things that I think you asked me earlier is the success of Bitcoin. The fact that it was all open source was a huge, huge issue, a huge plus. So you're really trying to create a resource that is open to other people they can build value on. So what we are about an intrinsic is actually being able to support, and it builds on an open source. project right now called Fluxstream that has
Starting point is 01:03:19 it's it was done out of Cargay-Mell and that gives people control over their sort of like quantified self stuff the point is to be able to give have someone set up their own container of their own data
Starting point is 01:03:35 and be able to assert certain credentials about themselves put it on the blockchain that they control and then being able to assert them I think you create the whole different kind of dynamics that people have different badges, different ways in which they can assert and authenticate certain attributes to themselves that will enable all sorts of possibilities. Now, one of the areas that I think that I mentioned is most critical is this what's happening
Starting point is 01:04:07 in this, what I call sort of the gig and guild economy that's coming out. Well, you look at the maker movement, you look at Uber. you look at Instacart, you look at all these efforts to say, well, here's an independent contractor. How do we create, how do you, when they're working for a company, you're creating a certain assets, assets, and your skills and your accomplishments. Why shouldn't you as an individual be able to benefit from that? Why can't you, in turn, work through your own network to create learning value, increase your craft, but also mitigate risk and enter into agreements with the Uber's or the Instacar,
Starting point is 01:04:53 all these other kinds of service, or be a maker. I mean, the maker movement is, I think, inevitably going to be there. And that, to me, it's like an artisan movement. But it's not, a guild is not in the sense of the old sense of a guild. A guild, I think is, yes, it's a way in which people can come together and you can use smart contracts to reduce the enormous friction, legal and economic friction, and organizational friction, and how to provide services. So one of the things we're very interested in and comes out of the background at the Law Lab is how do you create, you know, solve, there's so many legal contracts that are just very standard agreements that once people agree to them and I have the right credential, I should be able to push a button.
Starting point is 01:05:40 and get a loan or push a button and get a new passport. Push a button and create a credential around credit that be independently verified. So we're looking at this in this new networked economy and also in developing countries where there's a huge opportunity and need for that. So that's sort of the intrinsic value proposition. And it's sort of based upon these principles and how to create basically an API network of a value that will bring a new economy and be allow you to monetize us in a very different ways than having the control of a Google or Facebook.
Starting point is 01:06:19 Like massive participation for lots of different people to be sort of distributedly innovative around how to create credentials, how to create value, and not how to be captured by a few small big companies. Cool. Well, John, thanks so much for coming on today and joining us. It's really exciting the things you're working on. Well, thank you very much. It's exciting to be all part of this together.
Starting point is 01:06:45 So if people want to learn more about your work, so we'll, of course, link to ID cubed and to the Bitcoin and Burning Man book, is there some other place you want to send people to? Yeah, well, I'll send them to ID Cube, and eventually we're going to be sending them to intrinsic as well. So that, and we'll be able to do that and be more official. But the white paper I'm going to be putting out on the IDQ website. Okay, fantastic.
Starting point is 01:07:14 Yeah, because you wrote a nice white paper that unfortunately isn't draft stage, so we can't put a link up yet. But yeah, if you let us know, then, you know, we can let people know. I will definitely do that. And I'm always welcome for feedback. I mean, this is a living conversation that people have. So I'm always, I look forward to that. Yeah, no, thanks.
Starting point is 01:07:32 So one last thing before we wrap up. So we started this T-shirt contest, basically bribing people so if you leave us an iTunes review we're going to do a drawing I think we said once a week but let's do it once a month and we give out four or something like that and then we send it to you and of course you can say it was the most enlightening hour you ever spent or you can say it was miserable and I want my money back so either is fine and so yeah and of course if you do that you need to send this an email because we have no way of identifying
Starting point is 01:08:07 people on iTunes. So you need to send email to show at EpisodeUpentibbukon.com. Just let us know which one it is. So yeah, so that's it. So thanks so much for joining us. Thank you very much. It was great. Enjoyed it. Absolutely. So yeah, so we put out new episodes of WebSend and Bitcoin every Monday. Of course, you can get it on iTunes or your favorite podcast app. You can also get the video on YouTube at YouTube.com slash Epiccent and Bitcoin. And that's it. So thanks so much. And we look forward to being back. next week.

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.