We’re joined by Jameson Lopp. Jameson is the CTO of Casa, a company providing key storage solutions. Previously, he was an early engineer at BitGo. However, to most people, he is perhaps known for his Twitter presence and his excellent writing. Over the years, Jameson has written extensively about Bitcoin development, cryptocurrencies, and personal operational security. A self-proclaimed “Professional Cypherpunk,” aligns with the ideas of libertarianism and volunteerism.
Topics we discussed in this episode
- Jameson’s background and how he learned about Bitcoin
- His political view before and after discovering crypto
- What stands out for Bitcoin in 2018
- His writing on the decentralized nature of Bitcoin Core development
- His views on how Bitcoin compares to Ethereum on the topic of development control
- His thoughts on the Lightning Network and smart contracts on top of Bitcoin
- Jameson’s approach to personal operational security
- The tradeoffs of having air-tight personal OpSec
- Casa and its vault offering
- Microsoft Azure: Deploy enterprise-ready consortium blockchain networks that scale in just a few clicks. More at aka.ms/epicenter.
Sebastian Couture: so today our guest Jameson Lopp, if you are on Twitter and follow the crypto twittersphere you are probably familiar with this character, so we talked a lot about his early days in Bitcoin, how he got involved in Bitcoin his political views with regards to volunteerism or anarchy as others lie to call it. We also went in depth about his writing, as he is quite a prolific writer and writes about Bitcoin and also operational security, quite a bit about his operational security and the lengths to which he goes to protect himself and his privacy and his data in general. So it was a really great interview, we hope you enjoy it, and if this strikes a chord with you, if you think operational security is something that is important to you, please let us know on twitter what you think or things that you might implement or best practices that you implement in your own personal life to protect your operational security online. Here is our interview with Jameson Lopp.
Brian Fabian Crain: we are here today with Jameson Lopp, he is the CTO of a company called Casa. They provide a high end, high quality key storage solution. We are going to speak about that later. They also have a Bitcoin lightning node. He was previously at BitGo, he was an early engineer there which of course has been providing Bitcoin vault and storage custody solutions. and he is very well known for his writing, he is an excellent writer and I was actually just on a long plane ride and I read all of your blog posts and on MEdium. there is a lot of fantastic in depth Medium posts about Bitcoin, development, security aspects but also some things like operational security and some of the crazy things that Jameson goes through to make sure his operational security is top notch. So thanks so much for joining us today Jameson.
Jameson Lopp: Pleasure to be here, thanks for having me.
Brian: I’m curious to start off and often we ask that question but it’s always interesting to hear how people originally became involved in Bitcoin, learned about it and their way in.
Jameson: Yeah so I unfortunately do not remember the first time that I heard about Bitcoin, I’m sure that I heard about it several times and dismissed it several times as some new system that was going to get hacked and everyone was going to lose their money, but at some point, iit kept coming back, I kept reading about it on Slashdot and other tech sites, and I decided to look into it because it was not going away and once I read the white paper I realised that it was actually a fairly elegant computer science solution and that’s what really caught my interest and made me start wanting to dig into it more and really understand how it worked and I was just fortunate at the time I was working at this online marketing agency doing a lot of heavy lifting on the backend of the data analysis and whatnot and I talked to the guy who was sitting in the cubicle next to me and I was like hey do you know about this Bitcoin thing and he was like oh yeah man I’ve been writing bot to do automated arbitrage trading on exchanges between various cryptocurrencies and paid off my mortgage from doing that and I was like why didn’t you tell me about any of this. But thankfully he was able to answer a lot of my really basic questions about it and within a fairly short time prior I had surpassed even what he had knew about it and kept diving down the rabbit hole and eventually created my own fork of the Bitcoin software to get more analytics and data out of it and ultimately started a few meetup groups and after a few years there was enough capital in the space that I was bale to go full time so now I have been doing Bitcoin engineering for a good 4 years now.
Brian: And so what was it about Bitcoin that when you heard about it it was okay this something interesting and what were your political and philosophical views, pre Bitcoin.
Jameson: Yeah I mean I had never really thought about money and economics that much other than I took an Ecomm 101 class at University once I started looking into how it works and the idea that you can present money and do it in a way that nobody controlled it I realized this is a very powerful concept and it makes sense to me because I feel like money is this abstract idea that doesn’t belong to anyone it belongs to humanity at Large and it makes sense for something like this to be an open collaborative project and the idea that we can actually make it an open collaborative source project was very intriguing to me and really appealed to some of my anti-government sentiments and so from a political standpoint I had been all over the spectrum, was raised in a very conservative household and ended up going to a very liberal university and so throughout my voting career I voted conservative and then liberal and then after realizing that none of those parties were actually fulfilling their promises or seem to be improving my life in any way, started to go down the libertarian route and it was once I got into Bitcoin and then started reading the history of the cyberpunks and other crypto anarchist movements that spawned out of that, it really just pushed me even further down the libertarian thought process.
Sebastien: In that vein you did this interview with Crypto 101 which is a blog post, we will link to it in the show notes, where you said that you strived to bring crypto anarchy to the world. What does that mean exactly?
Jameson: So the word anarchy can definitely trigger a lot of people especially because governments use the word anarchy as a bad word and they try to make it seem like anarchy is equivalent to chaos and violence and destruction and whatnot but less triggering word would probably be voluntarism or just the idea of having voluntary interactions with people, so if we are approaching it from the standpoint of we want to build a society where everyone is interacting with each other voluntarily, rather than due to threats of force or coercion from this over arching entity such as a government then the way that we get there is that we look at all the different services that government are providing and we ask ourselves how can we privatise these services, how can you offer them in a way that is voluntary so that if I want to have my roads that I’m driving down, someone has to pay for them, well maybe people who are using them should be paying for them and right now it makes sense that a lot of these services that we’re paying for are done through taxes because it is just an easier way to pay for things than actually paying for what you’re using but as the technology continues to improve, then we should be able to automate a lot of these interactions and be able to have micro transactions as we’re going down the road or as we’re using a service that is out there where basically we need to decrease the cognitive load that is required to perform those interactions so the government is basically stepping in and managing a lot of stuff so that we don’t think about it so we just have a lot of money taken out of paychecks and then the government deals with all the coordination so if we can reproduce the coordinational software preferably software that is smart enough to understand you as a user and what you want then that’s when we actually start to conceive of replacing some of these coordination mechanisms that the government is doing with actual software mechanisms and this is a very long term view, I don’t think it’s going to happen even in the next few years but it seems to me that as we’re continuing to build software that is getting better at performing these actions then we are at least headed in the right direction.
Brian: How do you see that actually playing out, because I agree with you on a high level but if you look at something like roads, then okay you have this coordination problem and taxes kind of make sense but then maybe a lot of other things you could say okay actually you could easily replace it with a market based mechanisms but is the path you see here, do you think because of these increased technological possibilities, say if you take the US government they would increasingly move in a direction like that and say okay we privatize and we have these voluntary mechanisms instead of tax driven, or do you think what’s going to happen is that the fiat system is going to collapse and in its ashes you will have the rise of these more anarchist structures. What’s the path you see.
Jameson: I certainly don’t think it’s going to all happen at once, there will be these gradual evolutions ait certainly seems less likely that it is going to be a major collapse of US or Europe or whatever, but rather what I think is more interesting is the smaller countries or the more mismanaged countries and as they collapse, those could be test beds that are rife for adoption of technology like this so I know a lot of people talk about Venezuela and their hyper inflation and how Bitcoin could help people in that situation a lot more than those of those of us a lot more comfortable in first world countries, the same thing maybe true for any other types of services and technologies that can replace various government functions. How is it going to happen? I mean that is when you have to kind of wave your hands and say well if we believe in the free market than entrepreneurs are going to come in and find opportunities where a government is not doing a good job providing services and basically offers these new high tech versions for people and that is when an option would have to happen I think if place where the new methodology for coordinating stuff is superior to what is already in place with government so if that happens and it’s successful over a long term than perhaps the technologies will evolve to a point where they can provide even better services than first world countries.
Sebastien: not to get into any political discussions about the current state of the US government but what does the government shutdown, I believe it’s still going on, tell you about the possibility or impossibility of this to happen.
Jameson: well the US government shut down I think a couple days ago, but I think it’s only a temporary suspension, so they’re funding the government for another 3 weeks and then it might shut down again and I think it’s interesting to see at least in the US, we continue to polarize politics more and more I think at least in part due to the result of media and communications technology and that has, it seems to me, resulted in even greater level sof gridlock so that it’s even more difficult to get things done from a political sense and it seems like these nation states are floundering in what they can do, so that could provide more opportunity for these other types of technologies to step in, but I don’t know, I don’t even really participate in politics anymore, I don’t vote both for operational security reasons and I think it’s a waste of my time. I think that it’s a better use of my own resources to focus on the systems that I hope over the long term can replace a lot of the functions of a government.
Sebastien: well if it’s any lesson I think Belgium didn’t have a government for I think 2 years in early 2010 and I don’t think there were very many voluntary or anarchic style systems to emerge from that I was living right on the border of Belgium at that time and it seemed like that was going on there but maybe it was because I was too busy drinking beer and having Belgian fries. So since you became involved in Bitcoin did you have any periods of doubt when you were doubtful for the future and the project and if so, in what way.
Jameson: Sure we have been through a number of hype cycles and FUD cycles and Bitcoin is going to die for this reason or that reason, the greatest doubts were probably in the early days of the scaling debate when it seemed like we had a great opportunity in front of us to increase block sizes and allow more throughput on the network allow more use case and whatnot and there were times when I thought there were some pretty bog ground swells of support for that and we were looking at statistics like mining and stuff and it looked like it was sure to go through and then there were a number of surprises along the way that basically showed that statistics are not necessarily indicative of what is going to happen and there were also the whole censorship and moderation debate things got fairly nasty there, I had some posts of comments that got removed from Reddit and pissed me off and I went and became a moderator of the Bitcoin X subreddit because we were censorship free subreddit but after moderating that for 6 or 12 months I gave it up because it became clear to me that unmoderated forums are pretty terrible places and you don’t really get a whole lot of signal through the noise but I never lost enough hope that I wanted to stop working on the projects, this all happened basically after I had gone fulltime and was working at BitGo. Even within BitGo we had a number of arguments about where the direction of Bitcoin was going and what different people wanted to see out of it but ultimately even though there was a lot of frustration and periods of doubt I got to the point where I basically figured that so many people are spending so much time and resources arguing about what is going to happen with this system then it’s probably not an indication that it’s going to fail, it’s actually an indication that there are a lot of people that are dedicated to maintaining and improving the system and we just have slightly different beliefs about what the best way to go about that is and what the trade offs are that we are willing to make but ultimately out of that many year of debate, my conclusion was that Bitcoin can’t actually die unless we all agree that it’s dead, unless we all agree that we no longer want to work on it and try to improve it and so that is why I think the biggest threat to Bitcoin is apathy. It’s not 51% attacks or nation states and regulations or any of the other million reasons that you’ll find people who have written articles about why Bitcoin is going to die this time, really I think Bitcoin can only die if it becomes super boring and no one wants to work on it anymore.
Brian: And do you think it’s not let’s say 51% attack because you think if people still care about Bitcoin then they will hard fork to a different proof of work or something like that.
Jameson: Yeah i mean ultimately any technical failure or bug or anything that gets exploited at a technical level, if that causes the system to cease to be functional and operational then that means that we have to fall back to the foundation which is human consensus. So all of the stuff, the code, the protocol, the network, the hardware that is running nodes and miners and whatnot, all of that stuff is really just running machine consensus and machine consensus is just our best guess, our best representation at trying to turn human consensus into code but this is what the ultimate challenge is figuring out what the human consensus is for what Bitcoin should be and that started to get more philosophical and goes down the path that I went into great depth in my article entitled “Nobody understands Bitcoin’ where I was really just trying to describe this vague concept that is floating out there of what Bitcoin is and how developers and other people in the ecosystem who spend a lot of time talking about Bitcoin but kind poking at that they are trying to read the shape of what this actual concerns of Bitcoin is but nobody can actually completely grab it because it is dispersed amongst all of the people that are participating in the system.
Brian: Let’s say we think that 10 years ahead, where would you like to see Bitcoin and what would you like it to be, because I think you correctly pointed about there are different conceptions with the whitepaper it talks about electronic cash and this idea of digital gold has become more prevalent, maybe some people like the idea that it will be some sort of basis for trustless computing and those sort of things even though now they probably get built more on Ethereum modern networks mabe in the future Bitcoin could also be that or payment there are so many different things, so what is the thing you would most like Bitcoin to evolve into.
Jameson: I think that summed up a lot that in another article I wrote about Bitcoin being this trust anchor, so I am a technologist and ultimately I see Bitcoin and the blockchain that’s underneath it as a new type of database, we just happen to have a new set of rules and protocol around how that database gets replicated and how we pin new data to the database so from that standpoint I do think there is more to it than just money I think what we are trying to do is crete this global record of truth or at least this authoritative record that has no authority behind it so you can definitely expand more resources to building on top of it than just for money and finance, basically any data that you want to become part of this authoritative record you can put it in there and the question just becomes if you’re moving beyond the simple accounting ledger that the Bitcoin protocol supplies you have to basically create your own protocol, your own new consensus for whatever that extension is and so whether that is some sort of layer to network or a sidechain that is pegged to Bitcoin or extension blocks, there is potentially limitless ways to do this, it is really limited to our own creativity, imagination, technical engineering skills and our ability to convince other people to actually agree with us, to use whatever we build on top of it is. So from that standpoint I do think more complex systems, smart contract type stuff, better privacy is definitely possible by anchoring into Bitcoin and not having to change the Bitcoin protocol itself, so i want to see a lot of people experiment with this. What is the most recent one, I guess Vera Block is an interesting new one where they are anchoring a lot of stuff into the Bitcoin blockchain, they make use of the proof of work and it’s not quite clear to me how many different systems might get built on top of that but it is this blossoming of experimentation and a lot of them will fail but eventually any type of system that is being built on the internet and is meant to be some sort of global system with a tate that is backing, by state I mean data state, that is backing whatever you are doing interacting with that system, it could potentially benefit from using Bitcoin as an anchoring mechanism. So it’s really broad really general, even if we are looking at smart contracting systems like Ethereum or EOS or whatnot, I think a good example is RSK where they are taking that smart contracting language from Ethereum and they’ve created this sidechain that is pegged to Bitcoin so you can have the best of both world. Whether or not that ends up being highly adopted but that is the type of experimentation that I like to see and I just want to continue to see more systems get secure by these global consensus mechanisms because it’s going to make the m more robust against various types of attacks.
Brian: Are you a Bitcoin maximalist in this regard that you think that Bitcoin is the correct foundation for this, as opposed to having other chains, or proof of stake. Because right now for the most part maybe you can build some smart contract on Bitcoin but hardly anyone does it, 99% of activity is on Ethereum or other types of new chains, do you think those will migrate more towards building on Bitcoin.
Jameson: It’s going to require a number of things, I think there are people in the Bitcoin ecosystem who are interested in smart contracts and they simply don’t like the way Ethereum went about doing it. It’s like there is this big clash between the idea of execution versus verification and so a lot of the more conservative bitcoin developers don’t like having smart contracts that have to get executed by everyone on the network, they rather want to perform the same type of logic but where the actual execution happens privately and you’re just providing a proof of the execution that the rest of the world can verify and from that standpoint we are seeing stuff like syntax trees and the TapRoot and the simplicity smart contracting language which I would argue those are the things that some of these Bitcoin developers who are interested in smart contracts are trying to build their bitcoin version of more expressive smart contracts. Now how long is it going to take before that becomes a thing that is as easy for a newbie developer to use as like Solidity or Viper On Ethereum is once again up in the air. It seems like the like the space of advancements with the Bitcoin base protocol is a lot more measured and slower than a lot of other chains for a number of reasons but I generally call it conservatism or you could even think of almost like aerospace engineering level of thinking through all the edge cases and retesting stuff and not wanting to deploy anything unless everybody’s close to 100 percent confident about it.
Sebastien: But would you say that perhaps this conservatism and this time that it may take for these platforms to emerge and become stable might cause a situation where people build applications on Ethereum because it’s easy and you have a consultation of developers there and people already building on those systems and where it just becomes the switching costs just becomes way too high and where interactions between the two systems just don’t exist or are complicated where you in the end might not come to fruition. You know Bitcoin would become this system where no one can build complex applications.
Jameson: I don’t know about switching costs but it’s really more of network effect growth. I think the argument for for creating almost any alternative system to Bitcoin is that you have a lot more flexibility in what you can do and changes and evolve it. So you potentially have a better chance of exceeding growing faster than Bitcoin exceeding its network effects and becoming the dominant system. That’s seems to be basically true for for almost any crypto asset network out there, is that it’s usually because they’re some set of people or developers wants to do some things that were pretty clearly like not going to get accepted into the Bitcoin based protocol and they would feel like it would be easier for them to create their own new consensus around a shared set of objectives and roadmap. And that’s that’s why competition is great. I think one of the big push backs to the maximalism thought is I see a lot of people saying well you know you’re maximalism is pushing for a monoculture and I think that’s kind of a misunderstanding of at least what most Bitcoin maximalists think. I don’t think any of them are deluded to the point that they don’t think that other systems will exist. I think that it’s more about looking at the ways that network effects evolve and first mover advantages. The value of networks and how they are distributed where generally the vast majority of value between competing networks will go to one network and then the other networks will just be a lot smaller. But these tend to be more economic type of thoughts of how these types of systems tend to play out rather than a blind belief that Bitcoin was first and it must be the best and will never be superseded. There’s definitely plenty of potential for other systems to get greater adoption and surpass Bitcoin or somehow be order of magnitude more utilitarian than than Bitcoin is and therefore supersedes its network effects. I don’t think that anything is set in stone for sure, there’s going to be a lot of competition for the foreseeable future.
Sebastien: So you wrote this great post on Medium looking at Bitcoin in 2018 and so drawing the picture of what unfolded over the year and in that post were a lot of really in-depth statistics on everything from transaction volume to number of times Bitcoin has been declared dead over the years. So I really encourage people to look at that post. We’ll have it on our show notes. What stands out the most for you in 2018. What were the flagship things that we can look at standing out for this year for this past year.
Jameson: Well I think the biggest thing that also surprised a lot of people was the growth of the Lightning network and how quickly people were adopting it despite it still being fairly risky to do so. This is even even true with my own company and the lightning nodes that we’ve been shipping out there. There are still plenty of unresolved issues from security and usability standpoints where the Lightning network still has years worth of development ahead of it before it will become something that is capable of really being a mainstream payment network. But nevertheless the enthusiasm for that probably at least partially after years of stalemate with the scaling debate and people being excited about actually having something new to do a lot of people have just been plugging in and experimenting with it and as a result finding issues, breaking things, which is how it evolves is we we push the envelope. We we find problems and then we fix them. And this has certainly been my experience over the past year with learning more about Lightning network and having some close calls with losing money and blowing up my notes and stuff like that. It’s actually the basis for one of my newest talks that I’ve been going around which is basically entitled The Bitcoin Decade and Failing Forward. It’s once again looking at the history of this space. There have been innumerable failures over the years and actually I think Andreas Antonopoulos did a really good talk a few years ago with his Bubble Boy and the Sewer Rat talk where he talked about how these antifragile networks continue to evolve over time and the internet itself is a similar type of story and that’s why I think once again apathy is what is going to kill this thing is like as long as people are still interested in it they’re still putting their time and resources into using it and experimenting building and breaking, that’s how the technology continues to improve and that’s how we slowly but surely get to that next tier and then the next tier and the next tier of adoption.
Brian: So I remember we did podcasts 2015, I think maybe was beginning of 2016, and Lightning was four months away and it’s going to be huge. And now it’s taking much longer. I think last year there was a significant amount of activity but at the same time it seems all like the kind of activity you’re talking about, people were saying ‘this is cool I want to try it out, play with it and test it’ but it’s not really people using it yet, the mainstream wallets haven’t adopted it. So do you feel like this is just an inevitable thing and it’s going to take some time for it to happen. Or do you still see major risks and a big probability that maybe that network is actually never going to reach the point where it’s going to be mainstream capable.
Jameson: Well it currently seems to be the inevitable path because that’s what a lot of people are focusing on. With regard to the capability of going mainstream, I would say that there are still a lot of questions out there there are things that need to be built and improved upon. But I would say that one of the biggest open questions is mostly going to be around liquidity management. Not necessarily the technical side of the network but the financial side of it of how do we build tools that make it easy for people to manage the liquidity on the network and specifically manage the balance of the channels on the network. I think the first article that I wrote about Lightning network was around early 2015 and that was really one of the biggest problems that I was talking about back then as well is trying to model what the the economic issues are going to be with the network. And as we’ve had a lot more people actually experimenting and building out the real networks you know that now we’re actually getting data where we can better understand how this new network works, and so from a protocol standpoint that’s where things like the autopilot functionality that exists right now is not great. A human who is being careful about their channel management can do a lot better than what the autopilot is doing. But this is one of the things where we need more data in order to figure out what the best way of managing the channels is and that’s just like at a micro level. Then the next question becomes what are the macro economic issues. And I also talked about a few of those problems in my article but one of my conclusions was that in order for liquidity at a macro scale on the network to be more sustainable, I think it would be extremely important that we have exchanges that get tied into the Lightning network so that you can basically rebalance channels easily with out of band payments through exchanges. So lots of open questions for sure. There’s a lot of work ahead of us. I think that at least from a general standpoint that this type of layered protocol engineering does make sense. It’s the same way that the internet itself was scaled with various layers of technologies. So I certainly don’t believe in any of what people are putting out there saying that it results in inherent centralization and fractional reserve banking and all this other stuff but that’s not to say we’re doing something that has never really been done before. It’s going to come down to I think the level of dedication that people are going to put into trying to solve the problems.
Sebastien: You also wrote this post recently about who controls Bitcoin and in it you describe the history of Bitcoin core development and who has maintained the repos over the years and also describe the different layers of security in the different layers of decentralization all the way from when someone issued a pull request to a fork being adopted or something of that nature. It was a great post also one I want to mention. So why did you want to write about this why did you feel it was important for you to write about about this.
Jameson: I would say probably the majority of the long form blog posts that I write are fairly self-serving because I tend to write about things after I have received a question numerous times and I find myself repeating myself basically trying to explain a complex topic and so then a lot of times I’ll just say I’m going to write it once really really well. And then in the future I just seen that link whenever somebody asks the question. So this question of does Bitcoin Core as a group control the protocol itself of Bitcoin is something that has been coming up at least ever since the scaling debate started and we started seeing alternative Bitcoin implementations that were created specifically for the purpose of forking away from Bitcoin Core and their process and it’s very difficult to convince people of because of what a complex process it is, though if I had to sum it all up you know it basically comes down to the fact that Bitcoin Core can’t force anybody to run their software, but even behind the scenes there are so many other security considerations and processes in place to ensure the integrity of the code just to try to minimize the trust within Bitcoin Core itself as an organization that we want it to be as as verifiable as possible and as difficult as possible for anyone to inject bad code in there. Ultimately this doesn’t address governance issues of well what if I have an idea that will make Bitcoin so much greater. And the the Bitcoin Core development process rejects it. That is ultimately going to come down to the way that any open source software works which is, you have voice and exit as your two main options and if you can’t voice your opinion to the point that you can convince others to change the software repository that is being used by most people then you have to fork your own and try to build a new level of human consensus around that. But the main thing that I was trying to get at is that Bitcoin Core is just a name. The fact that it happens to use this specific GitHub repository is also not that important. Once again it comes down to this vague hand waving concept of, early we were talking about well what is Bitcoin what are the human consensus for Bitcoin. It’s something that’s out there you know in the Ether and we’re all trying to understand what it is so that we can turn it into code and it’s kind of the same thing for the main reference implementation for that code. This sort of focal point of development. There is no authority that that forces the focal point to be in one GitHub repository or forces it to be managed by certain people. That focal point has changed names over the years, it has changed platforms of where the repository is over the years and there’s nothing really preventing it from changing again if the human consensus occurs to change it. There’s plenty of reasons why that might happen. This is once again the voluntary interactions of this anarchic system can be very frustrating to people who like to have no hard and fast decisions made about things. And when we get into stalemate situations where the default in these systems is basically no or veto, if people don’t make a conscious effort then usually the default is no action. That’s when people get really frustrated and that’s when drama happens or people start forking off and trying to build new consensus. Ultimately I think that is the way that the governance of these systems is meant to work. It’s just a completely new model that people are not very familiar with and it can result in frustration and people getting upset.
Sebastien: I really like this notion of focal point that you use quite a bit in the article and a few things that I learned from this article – it kind of opened my mind to this this idea that these focal points exist in just about every form of organization in our society and the other also is that I realized that Bitcoin is a lot less centralized than I thought it was previously. It feels much more decentralized now that I understand the different layers and fail safes that are in place in order to protect the repository but also the network. Cmmitting to the GitHub repo ultimately doesn’t signal very much in terms of the direction of the network. So I encourage people to read the posts in detail. Compared to other GitHub repositories or other software repositories, open source software projects, does Bitcoin fall in the norm in terms of implementing all these fail safes and the signatures and the verifications and whatnot. Or is this really an outlier.
Jameson: I think that it’s an outlier. I don’t even have enough time to do that same level of research on all of the other repositories. But I have looked at it like some of the other the forks of Bitcoin Core and their processes and some of them at least do GPG signed Code commits but none of them seem to have that same level of automated infrastructure integrity checks setup. Really what you find with a lot of projects is that it’s one or two developers that pretty much control everything and that’s usually just due to the lack of size and interest in that particular project. Another particularly interesting thing that I find is ‘which node implementations have automatic update mechanisms built into them’. This is actually something that I ran into recently where I was trying to update one of our parody nodes and I downloaded the new binary for it and was checking the version from the RPC output and for some reason the version wasn’t changing and it took me half an hour to figure out that basically parody had this automatic update functionality and it was under the hood even though I was running a different binary it was it actually had some other binary on the back end that it was running in place. That’s just weird stuff where it makes sense for a lot of software to automatically update, that it decreases the cognitive load of the users of having to keep looking for updates on their own but it definitely changes the security model when you’re you’re trying to run this independent distributed network.
Brian: So I am glad you speak about this process of how Bitcoin is is updated and managed and I agree, I was impressed how much thought and levels of control and having automated tools to check all of the commits ever made and the decryptographic signatures etc. This is such a thorough thing has been built up over so many years and I recently heard this interview with some investment advisor, he was asked about what do you think of Bitcoin. And he was like ‘well you have so many cryptocurrencies and blockchain is interesting but the issue is it’s open source and somebody can take it and they can improve it, and why would the first version have been the best. Why wouldn’t somebody be able to go and say hey I’m changing something of a Bitcoin now it’s better and then if you invest in Bitcoin how would you ever be sure that it’s not going to be replaced’. And of course it could happen but I think also there’s so much infrastructure investment built and so much at such a level of quality and optimization and processes and automation and checks and assurances and you’re replicating that so hard.
Jameson: Not just for the repository though, what I think a lot of people don’t realize is the magnitude of the infrastructure across the entire ecosystem. And this is something that I ran into when I was at BitGo. We were running enterprise wallet API’s is that were used by exchanges and payment processors and and other various merchants and once all of these Bitcoin forks started happening and once the real explosion of tokens and stuff happened in 2017 it created a huge engineering workload for anyone who was working in this space because in order to add support for these things, even if if we’re talking about like forks that are very almost identical to Bitcoin or if we’re talking about like ERC20 tokens that are all very very similar, the ability to add support for new ones is a lot more than just a copy paste operation. You have this entire infrastructure stack that has to be replicated and then slightly modified and then have all of your new alerts and all of your other management systems running on that infrastructure stack and it’s a lot more difficult to get this entire distributed ecosystem with all of their own infrastructure to to basically spool up entirely new systems to support whatever your new Bitcoin 2.0 is. It is that power of network effects.
Brian: And of course network effects bring us to an interesting question because the other big cryptocurrency or blockchain network that has a strong network effects is Ethereum. What is your stance on Ethereum, what do you think of it.
Jameson: Let’s see I’ve written a few articles about it. Ethereum in particular gave me a lot of grief as an Infrastructure Engineer especially during the CryptoKitties period or whatever you want to call it. During late 2017 the last big run up where a lot of crypto networks were seeing high adoption rates and and basically running into their own technical limitations of what they could process on the network. And as an Infrastructure Engineer at BitGo I was running quite a few different nodes. We were supporting Bitcoin and Bitcoin Cash and Bitcoin Gold and Ethereum and several ERC20 tokens and Ripple and probably a few other things I don’t even remember. And during that period when when a lot of adoption was happening I found that it was the Ethereum nodes and the Ripple nodes that were having the biggest problems from an infrastructure standpoint. The Bitcoin nodes never had any performance issues with them but of course there were plenty of issues on the network at large just due to lack throughput capabilities and resulting downstream usability problems for people that were trying to make transactions on these networks. But main problems with with Ripple and Ethereum was that they were really really disk IO intensive compared to Bitcoin and its derivatives. And if I had to speculate then I imagine at least for Ethereum that is because of all of the state changes when you’re executing all of these smart contracts it’s having to go look up a lot of data and do disk reads. And from what I’ve seen the Geth and probably also Parody developers have made some pretty good progress since that time of reducing the disk IO requirements. But this is one of those things where these these networks they have to get stress tested in order for you to find the limits of what they’re capable of doing and then you find the bottlenecks you try to fix the bottlenecks as much as possible and then you continue forward until the system gets adopted to the level enough that you find new bottlenecks. And that’s the way that pretty much all of these things are going to have to continue to evolve. And I think that what a lot of people are arguing about when they talk about like long term adoption and technical capabilities is that they’re trying to argue about foreseeing bottlenecks far in the future which I don’t think that’s really possible, bottlenecks are often surprises. And it’s generally hard to predict them unless you’re doing a lot of diligence of basically creating your own networks and running a lot of stress tests on them which as far as I can tell there aren’t many people that are doing that these days.
Brian: Maybe one more question on on the Ethereum versus Bitcoin side where I think we have a big difference. So we spoke a little bit about the processes around Bitcoin and those processes revolve a lot about Bitcoin Core and that is very sophisticated in making sure changes are safe. And of course Bitcoin Core is this reference implementation and all of the miners basically run Bitcoin Core or basically that software. Now with Ethereum we have a specification and then we have multiple clients. So Parody and Geth I think the most popular ones. And then I think those are much less decentralized, Parody I think is basically about Parody the company. And I’m sure there’s some external contributors but probably not too many. And then Geth is mostly this Ethereum foundation. And again no probably external contributors. But then you have some process where they have to coordinate and make sure that the changes they make actually align and don’t end up splitting the network. So what do you think are the pros and cons of That approach versus bitcoins.
Jameson: There have been some very interesting debates around specifications and what is the specification for for Bitcoin. And people generally say the specification is the code and the reference implementation. I don’t fully agree with that either. I think that it gets you most of the way there but then with Ethereum actually having a written down specification that can certainly help and I know that there was probably a few cases where one Ethereum node implementation had a bug and when they went and they looked at that implementation versus the other implementations it was pretty clear that that implementation was not following the specification, but I think ultimately the question is what is the specification for any of these things. And I have to fall back to my hand wavy think of what is Bitcoin or what is Ethereum and what is any public permission less protocol while you can definitely write down the rules of what is in the code, it does a pretty good job of allowing you to understand the machine specification. I still believe that it’s not really possible to write down the human consensus for a specification. Ultimately you can write down whatever you want and you can go about you know trying to find human consensus a number of different ways. But there’s no guarantee that you’re gonna get that right. And unforeseen things can happen. I guess a good example at least with Ethereum, they had the DAO fork and I don’t recall but I don’t believe that reentrancy or what not was a hard part of the specification there. It really became more of a philosophical question around specification of the code versus actual intent of the code. And once we get away from this cold hearted machine specification we start talking about human intent or what it is that we really want. That’s when I think we get more towards this vague handwaving notion that the actual consensus for what any of these public permission lists networks is is just out there and it’s it’s hard to actually formalize. So TLDR, it can certainly help in a few situations but I don’t think that you can fully formalize any of these things because it’s what’s often here except it’s distributed amongst thousands if not millions of people.
Sebastien: You also wrote another blog post describing your operational security protocol or process or whatever you want to call it. And this was something that really struck a chord with me because it’s something that I’ve been really trying to get a handle on as well as my own personal life. But the level at which you seem to have gone to protect yourself your data and presumably your family is at a level that I never would have imagined someone could go to. You’re really trying to protect themselves while remaining a public figure. Now without spending a lot of time on why you decided to do this, which people can read about that. That is probably because you were swatted in 2017 I believe, why did you feel that you needed to go to these lengths to to protect yourself.
Jameson: Well the biggest issue which I think I talk about near the beginning of my very long post of what I did. The biggest issue is that you don’t know what might become an issue. In the internet age now we have the ability to easily reach millions if not hundreds of millions if not billions of people with a single tweet for example. And there are a number of examples out there where people have unintentionally said something on social media that triggered a horde of people as a result. And within that horde of people there might be one or two imbalanced people or people who have mental issues or they don’t know where the line is and they’re willing to go to an extreme length to try to harm you in some way or at least to to make you afraid. And so I think that’s kind of what happened to me as I went from having a thousand followers on Twitter and most people not really caring what I said having you know close to two thousand followers. And now if I say something that offends someone or that might be against someone’s financial interests because they hold a certain crypto asset then they might feel compelled to try to do something to start to hurt me or to make me afraid or in the case of the swatting they were trying to extort me though they didn’t do a very good job at the extortion. So it’s from my perspective trying to look at where I am now and then think well look I should probably be conservative and assume that it might get an order of magnitude worse. So I should try to improve my security and privacy to the point that someone who might expand an order of magnitude or more resources trying to find me or hurt me or whatever. Because you can’t put that protection in place retroactively. Or at least if you do you have to do what I did and basically burn your old life and start all over and that’s very difficult for most people to do. So it’s a lot easier to have the the privacy and security upfront like way more than you think you need. In case there is an attacker because you know if an attacker succeeds then the consequences are probably going to be no more devastating than whatever resources you put into the defenses upfront. If you’re trying to be proactive about it.
Sebastien: What do you think is the trade offs of having such rigid operational security. Because I mean I’ve implemented a few things in my life. One Of the things is I’m working to get off Google completely and off most social networks. And the tradeoffs are that you know once in awhile I need to do a little bit of more searching in order to find the closest restaurant that I’m looking for or something like that. But in your case this used to be a lot more than it’s used to be a lot more of a burden or at least I would assume. How have you found it impacted your life.
Jameson: I mean the tradeoffs mostly occur when it comes to like physical real world interactions of stuff. So on the extreme case, noone in my like physical proximity or no one that I interact with physically where I am now actually knows who I am. They don’t know my real name. They don’t know what I do. They just know that I’m a programmer. I’m a boring old programmer. We don’t have to talk too much about what I’m actually doing because you don’t want to hear it. And so that can affect you know your like real world social life basically is that I consider most of my real friends that I share interests with are on the internet or you know they’re remote, I no longer have friends with shared interests who are in my physical location. I do have friends that I’ve made that you know we can do things together and have fun activities and play games and and entertain ourselves and whatnot but it’s not in the you know crypto or privacy sphere of shared interests so it is kind of like living a double life almost. And sometimes that feels James Bond spy type stuff and other times it’s just plain annoying, having to drive around, if I want to pick up my my mail I have to drive fairly far to go to a private mailbox if I want to do anything that like requires a membership where they’re going to I.D. me or whatever, then I’m probably going to have to drive pretty far because I don’t want my name in any databases that are tied to location, so it can definitely be inconvenient in quite a few different ways. But on the other hand thankfully there are a lot of services out there these days that allow you to sign up pseudo anonymously. So that has been helpful for a few things but for things that don’t that is where it’s become a lot more expensive and you know hiding my real identity will tend to involve lawyers who charge me a lot of money to basically act as a proxy on my behalf.
Brian: That’s pretty mind blowing. So your neighbors and stuff like that don’t know you name. I mean that seems to be tricky especially if you’re a pretty big public profile. I mean the chance that somebody listens to this podcast or sees you on Twitter or something and says isn’t that the guy that I have this other name for. That seems like a high risk no.
Jameson: Well I guess I’m not actually a celebrity. You know I’ve only ever been recognized out in public one time and I think that was mostly due to the beard I had at the time. But other than that most of the time when I’m out and about I keep it pretty low key and I just look like another guy. So if I ever got to like real celebrity status level then hopefully that would mean Bitcoin has done so well that I can buy my private island or something.
Sebastien: I had this conversation with someone over the weekend where we’re talking about privacy, not so much personal OpSec but more on the privacy side, and at the beginning your blog posts you say something to the effect of, well most people would look at this and say well I have nothing to hide or I’m not such a high profile person why would someone want to attack me or steal my identity. And people often say this to me and I’m not really quite sure what to respond. I guess one of the things is of course we don’t really know what an artificial intelligence and this sort of thing is capable of in the future with the data has accumulated on you. What do you normally tell people, what’s your sort of way to convince people that having good operational security, keeping your privacy matters under wraps and also being careful about your data and what you share with whom and what companies and what’s your what’s your way to convince people I guess that that’s a good idea.
Jameson: So it’s like I said it’s kind of like the spoonful of proactive measures is worth a I guess a pound of trying to fix things up. It’s because there are a few things that we know. One of those is that information wants to be free. And basically any service that you give your data to over a long enough period of time it’s almost inevitable that that data is going to leak. It might leak due to what we’ve seen with Facebook of like partnerships in accidentally allowing partners to see data and and then those partners might leak it in other ways. Or it might leak because they get exploited somehow and someone managed to get a big data dump and put it for sale out on the darknet. But that’s the first thing that I try to tell people at the very least you might want to worry about identity theft because that’s so common at least in the United States. But then for more of the actual physical security and operational side of things you don’t know who you might piss off and especially if you’re active on social media. It’s just not possible to fully comprehend like the thought processes of everybody else out there who’s on the internet who might read or hear something that you say and then what they might do as a result and so I believe that the vast majority of people are “good moral people” who will not harm others to help themselves in most situations but it’s pretty clear that there are a small number of people out there who have a sociopathic tendencies or who will do things that we generally consider to be immoral and that’s what I’m worried about. And for me that’s because my audience size and my reach has grown to the point that there are a non negligible number of those people who are likely to come across what I’m saying and you get triggered by it. But while the likelihood of something like that happening for the average person is probably lower, you never know. And so it’s just a form of insurance against a somewhat unlikely but still possible event. It’s like the Justine Sacco lady that I had in my post where she made one bad tweet and as a result it impacted her career and her life. And her reputation is basically ruined at this point.
Sebastien: Yeah I think this fear of escalation as a very kind of U.S. centric type of idea. I think in the U.S. people will want to protect themselves in part because of this fear of escalation. Whereas in Europe people would want to protect themselves more as a preventative measure against companies that might misuse their data or data leaks or this sort of thing. I feel like here in Europe at least like fear escalation is quite low I don’t think people have much of a fear of that. They might say something on Twitter that will piss someone off to the extent that they might get physically harmed or threatened or something like that.
Brian: The one thing that stands out to me. So Twitter sure. One of the things that I found striking, often you have people on Twitter and I think in the crypto space this is very common, who just seem like horrible people, they’re so aggressive and then I met some of these people in real life and it’s a huge difference. They’re actually a nice reasonable person, or seemingly like that. So I think it doesn’t seem, I mean I could see it happening but I don’t find it so concerning. But then this is an area and I think you’ve talked about that too, of basically people saying okay let’s target crypto users and go in and try to extort them or steal their funds. And I remember reading a while ago there was some guy in Norway who was doing some Bitcoin trading and somebody went into his house and tried to steal the Bitcoins and then killed the guy.
Jameson: There have been dozens of those incidents and in fact we just saw a guy tweeting earlier today about his friend in Oman I believe was physically robbed and assaulted and then I saw another piece of news pop up about someone actually being murdered in Japan after meeting someone at a Bitcoin meetup trying to find like more source material on that. But that’s part of the problem I guess with being an early adopter in this space is that it’s kind of a paradox where it’s not a good idea to talk about money and wealth and assets but we also have an incentive to talk about these networks because we want them to grow, we want to get more people to come into the networks and expend their own resources to build the networks and evolve them. And so as soon as you start talking about being interested in these things then you’ve created a point in time where an attacker might go back and look at your history and say ‘oh they’ve been talking about Bitcoin since 2010 or Ethereum since 2015’ or whatever. And then you know the attacker starts extrapolating well you know they could have you know this many millions of dollars and they probably don’t have bank level security so if I’m weighing my options of where I get some easy money you know do I rob a bank or do I go find this crypto person who probably has a bunch of money under their mattress and a hardware wallet and I just need the five dollar wrench attack. And so that is one thing specific to people who are in the crypto space is that you know we’re we’re talking about these highly liquid bare assets and if you are going to go down the path of being your own bank you have to actually understand everything that is involved in being your own bank.
Brian: So I guess just before we wrap up maybe we can briefly talk about something that ties in very nicely here. So the company that you’re CTO of Casa, are you building basically these self custody solutions for Bitcoin. Is that also presumably one of the scenarios that you try to protect against like this five dollar a wrench attack. Or can you talk a little bit about like what this product looks like.
Jameson: Yes. So the first service that we started offering at Casa is the key master service which is basically a vault product. It’s a three out of five multisig Bitcoin wallet. And what’s different about this wallet, there’s a few things. One is that it is mostly backed by hardware devices and we support off the shelf hardware devices like treasure and ledger. And the premise is that it’s not only multi sig it’s multi device and multi location. So we’re building in a level of redundancy and robustness and minimizing any single points of failure to every aspect of the system that we can in order to protect not only against theft but also against loss. And when I say loss I generally mean something happening where the user screws up and they can no longer access their keys and basically all of the money is gone but nobody has stolen it. In my experience and from some of the analysis that has been done we estimate that twice as many Bitcoins have actually been lost than have been stolen. So it’s the fact that users are generally not I.T. experts or even if they are technical like myself, it’s just annoying like nobody wants to go through really boring data backup and backup integrity testing checks and all of this other stuff. Nobody wants to spend even an hour a year doing that. And I was spending one to two days every year refreshing my own cold storage setup which was this custom thing that use secret sharing and sharded out these encrypted file containers across various people that I semi trusted and thinking through all of the different attack and failure scenarios is exhausting. So we’ve basically created a very user friendly app on IOS and Android where if you can read the screen and follow the workflows on the screen then it’s really as simple as plugging in your hardware devices that you buy and and following our guidelines for how to initialize them and test them and do health checks every now and then. The one thing that we did that had not been done before is that we actually got rid of the need for storing recovery seed phrases. So with our solution when you actually set up your wallet we tell you not to write down the seed phrases and that is by design because users are terrible at security and if if the user has to keep a seed phrase secure then that’s this whole other basically iceberg of security knowledge that needs to be ingested by them.
So by getting rid of that they can instead just think of their security in physical terms, where are my physical hardware devices, distribute them in different access controlled locations and that’s a lot easier to reason about. And so this is generally what we’re trying to do at Casa. We also have other products, one of which is the plug and play node product. And we’ve got a few other things that are coming out pretty soon but we’re from a very high level trying to bring usability to the masses when it comes to security as we want to decrease the level of technical knowledge that is required to operate within these systems to get that maximum level of security. And so as a result our mission is just to help increase personal sovereignty. And so it’s a very broad mission and we’re going to be trying to attack it from a number of different angles, key management is just you know the first most obvious one.
Brian: Thanks so much for joining us today Jameson. It was real pleasure speaking with you. Hopefully we can have you back on at some point. I think it is a lot of stuff that we could dive in a lot deeper and maybe have more focus especially along OpSec and the whole security thing is a massive area.
Brian: Thanks so much for coming on and we will of course link to many of your blog posts which really make for excellent reading so please keep up the fantastic work there.
Jameson: Will do. Thanks for having me.