Jameson Lopp & Peter Todd on Libra: Technical Analysis

WBD144 - Interview with Peter Todd & Jameson Lopp (Banner).png

Where to find the show

iTunes | Google | Spotify | Stitcher | SoundCloud | YouTube | Deezer | TuneIn | RSS Feed

Download Episode MP3 File
The file will open in a new window. Click down arrow to download the file.

“If you asked me to go and make a centralised blockchain, run by a federation so that you could go and make a new type of asset... The architecture I would pick is actually not that far from Libra’s.”
— Peter Todd

Location: Skype
Date: Sunday, 14th July
Project: Casa & Independent Consultant
Role: CTO & Applied Cryptography Consultant

Libra promises to be the first distributed ledger to transition from a permissioned system to permissionless. However, it faces many significant technical challenges, many of which have yet to be solved by other cryptocurrencies.

One of the main issues Libra faces is scalability. At launch, Libra is tipped to offer 1,000 transactions a second, but with a network of potentially billions of users, this will very quickly reach maximum capacity. 

Will Libra be able to solve its scaling issues? Will people care enough about Libra to want to use it? What happens when there are disagreements within the Libra foundation?

In the second episode of the Libra series, I discuss these issues with Jameson Lopp and Peter Todd, as we take a deep dive into the technical side of Facebook's Libra.

TIMESTAMPS

00:04:10: Introductions
00:04:26: Delving into Jameson and Peter’s initial thoughts of Libra
00:09:41: Touching on Facebook’s disregard for privacy and following regulations for Libra
00:11:11: Exploring Libra’s untruthful statements and how Facebook are misleading consumers 
00:16:39: Delving into node operators and assessing the risk of operators being dishonest
00:22:40: Touching on whether the launch of Libra is beneficial to Bitcoin
00:25:11: Discussing whether Libra is part of Facebook’s broader surveillance capitalism strategy
00:31:53: Exploring some of Libra’s scalability issues in more depth
00:37:19: Delving into Libra’s roadmap to a proof of stake system regarding membership eligibility
00:42:26: Discussing usages for Libra and potential for UI integration with WhatsApp and Instagram
00:46:42: Exploring Libra’s smart contract language Move and potential futures for the network
00:51:26: Discussing KYC/AML in Libra and the move to push it onto wallet providers and exchanges
00:56:23: Touching on potential futures of Libra usage, regarding wallets and peer to peer exchanges
00:58:37: Exploring gas in Libra and some of the risks this poses to the security of the network
01:04:48: Are there simpler ways to achieve Facebook’s goals than a decentralised Blockchain?
01:09:43: Final comments and how to stay in touch

 

SUPPORT THE SHOW

If you enjoy The What Bitcoin Did Podcast you can help support the show my doing the following:

If you are interested in sponsoring the show, you can read more about that here or please feel free to drop me an email to discuss options.

SPONSORS

THANKS

A big thanks to my WBD Maximalist Patrons for helping support the show: JP Petit, Logan Shultz, Seb Walhain, Steve Foster, Tony, Gordon Gould, David Burlington, Jesse Powell, Bitcoin Tina, BitHyve and Wiel Menger.

TRANSCRIPTION

Peter McCormack: Hi Peter, hi Jameson, how are you both?

Peter Todd: Great!

Jameson Lopp: Couldn't be better!

Peter McCormack: Awesome! So we're here to talk about Libra, but firstly, I don't know if you both saw, but England won the cricket world cup. It's kind of a big deal!

Jameson Lopp: Yes, England is my favorite team of all of the stick bat teams!

Peter McCormack: Well, I'm having a beer! Anyway, we're going to talk about Libra. I've been through your article, Jameson, it was very useful and Peter, I've been through yours. This is going to be more of a technical conversation I expect. So I'll need my normal hand holding, but I think I kind of got my head around both of your thoughts on it. We'll start with you Jameson first and then you Peter.

Initially just to set us up, it would be interesting to know both of your initial reactions because we all heard about Libra for a long time in advance. Everyone had kind of assumptions of what it was going to be and how they were going to deliver it, but now we know a lot more about it. It'd be great to know what your initial reactions were Jameson?

Jameson Lopp: Initially I was actually pleasantly surprised with regard to a few things. Mainly that it seems like they do have some technically competent engineers who are working on it. But, upon closer inspection of the various white papers they put out, it's clear that they still have quite a bit of work to do and so there's still a lot of open questions and that's part of the reason why people are still talking about it so much.

Peter Todd: I was very much unsurprised by the technology and I'd say, I don't really think it's all that good either. But what really surprised me was the politics. I was expecting them to go and have a specifically US dollar backed coin or something like that. To go challenge leaders of the world with a new currency is I think crazy!

Peter McCormack: Right, so you think they would have just been better off with a dollar pegged stable coin?

Peter Todd: Well given that they're in such political trouble already, I think it's amazingly audacious to make this kind of statement.

Peter McCormack: And if they had been with a dollar peg stable coin, which we've already got a few, they probably would have been in a much better position and maybe not hauled in front of the Senate, I think it's on Tuesday David Marcus is there.

Peter Todd: Yeah I mean it's the kind of thing you would... The only rational thing to be seen to do, would be to go direct attention away from something else bad that you're doing, which is not very feasible thing to do when you're such a large company that's already hated by so many people!

Peter McCormack: Especially conservatives!

Peter Todd: Yeah and Democrats as well for kind of similar reasons in some ways, just opposite sides of the coin.

Jameson Lopp: It seems to me like at least part of the reason why they may have gone with this basket-based approach, is that they want it to seem even more de-centralized so that we're not completely centralizing our currency around one fiat currency, but rather we're going to try to make it as stable as possible to even prevent fluctuations that may happen in the major currencies. But like we said, that may very well backfire from a political standpoint.

Peter Todd: I would push back on calling such a basket, "decentralized" too. It's still centralized around a very small set of currencies. The only way you call that "decentralized" is in the language of geopolitics where nothing's really decentralized, it's always political power.

Peter McCormack: Do you think also it might been more of a... I'm guessing this is going to be a global currency as Facebook is pretty much in every country in the world and that to have a dollar pegged stable coin politically is not a great move in countries that may be dislike the US quite a bit and that by having its own currency, it kind of gets away from it feeling like a US currency and some kind of US dominance.

Peter Todd: I think even in that kind of scenario, it would still be better off for them to position this as a payment solution, let's say maybe multiple different currencies and equally with the US, I think the only parties there that really matter that much are the US and maybe Europe. Other countries are just too small for them to not be able to steamroll them.

Jameson Lopp: I think there are a number of questions around, how these end users are going to denominate the value. Are they going to get to choose? Like I prefer to keep my value displayed to me in one certain fiat currency, but on the back end it's actually this complex basket? What are the ramifications of that, especially from a regulatory and tax standpoint?

It seems like the sort of hand-wavy answer right now is that "oh, they're going to be regulated endpoints that are going to handle a lot of that." But then of course they're also touting the protocol itself as being open for anyone to build on. What happens if we end up having less regulated endpoints and will there be flagrant, I guess violation of a lot of regulations simply because people will be using it without worrying about tax rules or other regulations.

Peter McCormack: I mean that's kind of been Facebook's MO since it launched, whenever it was 12 years ago, I can't actually remember. But the MO has always been to flagrantly disregard laws, especially with regards to privacy and then go for their growth, go for expansion and almost pull back afterwards. I mean their privacy to begin with, was terrible and has improved in some ways.

My expectation with this, is they were doing something similar, but it seems like because they're so big now that actually they can't get away with that, that the regulators realize how powerful they are and actually they're going to have to follow certain regulations almost from launch.

Peter Todd: So I actually strongly disagree with that take and the reason why, is because privacy laws, from what I know, tend not to cover anything Facebook ever did until more recently. The way that they described it, is it was very unregulated, that there was nothing actually stopping companies like Facebook from existing, because people were opting in and providing all this information to this company and there just wasn't regulation around this.

I think the better example there is actually Uber, which very much was actually acting illegally in many places, maybe a little less than some people would like to claim. Certainly a lot of places they were flagrantly violating laws and interesting of course, Uber is part of the Libra consortium apparently, so that may say something, but I don't think the comparison to Facebook is apt.

Peter McCormack: Interesting. All right, well we'll work through both of your points. My general take from both of you is slightly different. Peter, you seem to have gone originally down the route that the white paper is quite disingenuous and some of their proposals and some outright lies. Whereas Jameson, you focused a lot more on the complexity of what they're doing and some of their unanswered problems.

But I'm going to start with you Peter and I'm going to start with a tweet you put out. I think this is a quote from their white paper, but you put, "the Libra Blockchain is a decentralized, programmable database designed to support a low volatility cryptocurrency, that will have the ability to serve as an efficient medium of exchange for billions of people around the world" and your comment underneath is, "that is a lie.”

Peter Todd: Yes. Specifically saying that a fiat currency backed system is decentralized, is right there on the face of it, a lie. It just is not possible to back a decentralized system by a fiat "physical asset." You just can't do it. So the moment you go put those two statements together, you're lying. On top of that, the actual technical architecture of it, is also not decentralized. What they're proposing is a federated system, where a fixed set of entities control system.

I'm sorry but you just can't call that decentralized. I'll point out another point, that the reason why I made those comments, rather than say taking the tone that others have taken, is it's a lot easier to go and point out when people are lying, because often it's just so blatant in this industry. I didn't want to go and sit down and spend a couple hours writing an article. I wrote what was easiest to go do, because I had other things to do that day and it was just obvious lies.

Jameson Lopp: There's also a fair amount of marketing taking place here. Even just the use of the word Blockchain, even when they admit that the fundamental structure of the database is not a Blockchain, though there is a Blockchain that gets used in there amongst some of the validator nodes.

But they're using a lot of these terms that have cropped up over the past few years and some of them, they're using them in a sort of aspirational sense and others, like the decentralization term, it's been so overused at this point that it's hard to even say, what is centralization or decentralization, because people are using it in a variety of different ways.

You could probably make an argument that a Federation of three or four nodes is "decentralized", but even just the fact that you're making an argument like that, is kind of a waste of time in my opinion.

Peter McCormack: Well, it seems to me they've cottoned onto two different angles here. The use of something like cryptocurrency, seems to be something that would appeal more to users of Facebook who've maybe heard of cryptocurrencies, they've never really used them, but they know it's a different type of currency.

They've also cottoned onto things like decentralized and Blockchain perhaps to appeal more to the industry side of people, potentially developers, wallet providers, anyone or exchanges who might work with Libra. It seems to me that's maybe what they've done there. But they certainly are marketing terms.

Peter Todd: I don't like people trying to push back on these ideas that centralized, decentralized have fairly rigid meanings and I think what the lesson to be learned there is simply that we have an industry that's just full scammers and people are dishonest. Just because a million people go and claim vigorously that the sky is red, not blue, it doesn't mean the color changes. It's only until you get to the point where the very meaning of red is different and you are actively changing the use of that term everywhere that that kind of logic makes sense. I don't think we're there with decentralized or centralized.

I think they have fairly clear meanings that are used in a way [Inaudible 15:19] as recently and what we have to go and point out as an industry is look, Facebook's willing to lie to us. They would be willing to lie to us about plenty of other things. That shouldn't really be surprising that a dishonest company will go hire researchers who are dishonest and go and publish dishonest white papers and that's just the way this stuff works. By not pushing back on that every single time, you allow people to get away with these things.

Peter McCormack: Do you have a preference for who manages money if you had to choose between Facebook and the government?

Peter Todd: Well, I would say neither, but in most places the government.

Peter McCormack: Okay, interesting.

Jameson Lopp: Well I think perhaps part of the problem here is that they're intentionally using some of these loaded terms like centralization and decentralization instead of specifically talking about, what is the threat model, what is your actual level of security against certain actors within the ecosystem, basically subverting your own actions. If we're really talking about this Federation of private companies, how many of them need to collude against you to make the utility of the system a lot lower for you?

Peter McCormack: So have either of you actually looked at the threat model, because one of the things I did consider with these companies who are the node operators, is that they have a lot to lose by acting dishonestly. A lot more say than a government does. What happens with the government if they dishonestly, we all just kind of suffer it, because we're used to it, but with a company you can actually vote with your feet and scandals aren't great for companies.

Peter Todd: The problem with that is, what dishonesty means, can vary a lot. I think a company, to say just directly steals your money with no legal reason whatsoever is I think very unlikely. But I also think it's not necessarily that likely for even say Bitcoin miners to want to do that, as the legal risks are too great. But to do things that only seem a little dishonest, such as stealing your money on some trumped up government reason that everyone knows is false, well, I mean that seems very likely. Why would you want to go up against a government there?

Jameson Lopp: There's also I think, the issue being that we don't fully understand how the dynamics of this Libra association are going to turn out, because I think there's going to be inevitably a lot of politics and the fact that this thing is not operational, it's probably not going to be operational for another year or so, we don't really know what, I guess the bylaws or other processes of this Libra association are going to be.

As far as I'm aware, none of that is public. But I guess at the end of the day, if there is a sufficient level of voting by members in the association, they could theoretically change the protocol, the network, the whole operation of the system at a whim.

Peter Todd: It's important to point out that at a technical level, that what they're designing isn't auditable and there's been some statements made that make it sound like the actual Blockchain data itself, they're not going to make publicly available. It'll be hidden away in these so-called validator nodes and all you'll get is essentially a signed statement saying, "well yeah, we claim that this is true, but we're not actually going to prove it to you."

Jameson Lopp: I've had at least one of the Libra folks say that, and once again, this is a sort of hand-wavy far future thing, but I've been told that the goal is to make it possible for someone to download and validate the actual Blockchain. But it sounds like that is not even technically possible at the moment and who knows when or if it will be.

Peter Todd: And I'd say it's also obviously not feasible if Libra catches on, because the amount of data you're talking about, it's going to be such a huge amount. You're not going to be able to do this by just walking up to them and asking. You're going to have to go negotiate some agreement and pay some costs and so on. Now this is not going to be a cheap data structure they will distribute widely.

Peter McCormack: I think one of the other differences is that I doubt people are going to be holding serious amounts of money in Libra, like maybe you would do in Bitcoin. There's a serious risk to people losing a significant amount of their wealth. Certainly early adopters or people who have invested heavily in Bitcoin, if something was to happen there, if there was some way of confiscating money or stealing money from the system or something dishonest could happen.

I see Libra more as a tool that people will perhaps load a few dollars in, as and when they need it, but I can't myself see people loading thousands and thousands of dollars into it. So whilst there's risk, it feels like there'll be less of a risk.

Jameson Lopp: You don't think that Libra coin might become bigger than Tether, as a Bitcoin fiat on-ramp?

Peter McCormack: Yeah, I guess so, but then if I was a serious trader wanting to use a stable coin as a Bitcoin on ramp, I would rather use Tether or GUSD or something else than Libra coin, but it's a fair point.

Jameson Lopp: Would really though, like aren't all of those operating under similar types of setups where basically the administrators of the stable coin can burn your coins without your authority?

Peter McCormack: Of course. But I feel like Facebook is, I don't know, it just feels less part of the Bitcoin ecosystem. It feels like it's more of a... It just feels like it will be under more regulations. It feels like it will be kind of tied to the government. I can imagine them having certain arrangements with the government after their Senate testimony hearings, which I don't think the likes of Tether have, but you might be right. But I don't know, it just feels different to me.

Peter Todd: One big difference between Tether and Libra is that at least in the near run, in theory, you can actually audit what Tether is doing. In Libra, the way that they seem to be going is that no one outside of the Libra coalition will actually know what's happening on the Libra Blockchain. In Tether, it's all public. I mean this is why you get all kinds of various Tether tracking Twitter accounts and websites and so on, watching like a hawk every time they go issue some coins.

Jameson Lopp: Yeah that $5 billion or so issuance recently and the subsequent burning was definitely fun!

Peter McCormack: Yeah, what was that about?

Peter Todd: From what I could see, someone made a typo in doing something a little weird to move Tether between different chains and then everyone kind of jumped on it.

Jameson Lopp: I think it was Poloniex came out and said that they screwed up doing a chain swap I guess.

Peter Todd: But it's important to note, that Poloniex said that they screwed up in something that they were doing with the assistance of Tether, which sounds like it was actually Tether making the transactions and they were just being nice and taking responsibility for the screw up.

Peter McCormack: It was quite interesting, I was sat down recently with Brad Stephens from Blockchain Capital and he had an interesting take on Libra being a win-win for Bitcoin. He said that, if it does go ahead, it will be a great introduction to cryptocurrencies and probably lead people more to Bitcoin. But if it doesn't actually launch, if for some reason they can't launch it, the government start banning it, the regulations are too tied, it's another win for Bitcoin, because it got regulated away whereas you can't regulate away Bitcoin, so it kind of proves the Bitcoin use case.

Peter Todd: I think it depends on how it gets regulators away. I mean, the worst case would be that they go set up a bunch of focus groups and so on in the various bureaucracies, that are then out of a job because Libra stops and they want to continue having something to work on.

Jameson Lopp: But at the very least, hopefully it'll keep the administration focused on Libra and their actions and you can at least keep them a little bit less focused on Bitcoin. Even though I think we all agree that if they did try to put pressure on there, that it's not really possible to kill Bitcoin. But regulators can certainly still cause a lot of grief for enterprises that are operating in the Bitcoin ecosystem.

Peter Todd: I would push back on this idea that you definitely can't go kill Bitcoin. If regulators really wanted to, they do have some options such as, for instance, banning the manufacturer of ASICs or forcefully inserting backdoors in them. There are some fundamental issues with Bitcoin's security model that have vulnerabilities in ways that other solutions don't and so far it looks like those vulnerabilities aren't really a big issue.

I mean, it's a very, very major geopolitical risk to go start telling ASIC manufacturers, "hey, you can't go build these chips", because that's a real point of contention among militaries around the world. How do you get access to this? How free is this and so on. But it is a possible road, much the same way that censoring the internet fully is a possible route. Are these things likely to happen? No, but we should be careful about our language and how we go talk about this and what we think is possible or not.

Peter McCormack: So maybe Libra is actually causing problems for Bitcoin and is bringing too many eyes onto it right now?

Peter Todd: I honestly don't know one way or the other. I think they're both possible and it's tough to make definitive statements about that.

Peter McCormack: All right, well before we get into the technical stuff, there's just one other thing I wanted to refer to. I think you both read the article. I know you tweeted out Peter, it was the Eric Wall article, "the problem with the Libra is that it's part of Facebook's broader surveillance capitalism strategy." Do you have worries about that as well?

Peter Todd: I think it's inevitable that Facebook will go use this as part of surveillance capitalism. The credit card companies and banks and so on, already want to do this as well. Probably the most positive thing that can come out of this is by tying such an obviously bad actor to financial surveillance, with any luck we'll get some political pushback on financial surveillance in general.

Look, it's good when an entity like Facebook, everyone realizes is evil so to speak, goes and surveils you financially or at least has proposals to do so, because then some of the laws can then get used to fight the existing bad actors such as VISA and MasterCard.

Peter McCormack: There's another issue there as well for the US government to consider, in that if they do regulate Libra away, they're kind of handing something to the Chinese with WePay and WeChat. I've had it a couple of times recently, I don't know if you have, but I've actually had people ask me if I would pay them using WeChat.

Peter Todd: I think it depends on how they regulate it away. Maybe the best thing that would happen is that they regulate away not only Libra, but also WeChat, WePay and so on. I mean obviously they're not going to make it go away in China, but they can at least stop it from infiltrating the Western world, which you are beginning to go see happen. Australia's an example. You do see WeChat and WePay on logos at stores.

Peter McCormack: Anything to add Jameson?

Jameson Lopp: I think that it benefits governments to have the ability to control the companies that are controlling the movement of money. So really what I was trying to think more about is that you can't get rid of all of them, what are you going to do? Are you going to have the central banks build systems like this and make everybody use those? I don't really see that happening.

So it seems like it would be easier for governments to allow, whether it's the traditional companies to continue operating the same or to work with these new consortiums of companies to try and get back doors and surveil as much of the flows of money as possible.

Peter Todd: Actually in Canada there is an effort at the central bank to implement something that at least at a technical level, at least somewhat similar to [Inaudible 28:02], to essentially create a digital currency that the banks could end up using to represent movements of Canadian dollars and to see that then push into the retail sector, I actually don't think is very unlikely.

I think the main reason why that certainly not happening quickly and in some countries is not happening at all, is more pushback from banks. I think I could definitely see situations where the governments are interested in having that happen.

Peter McCormack: Well, we need to track and stop the terrorists!

Peter Todd: Well, I'm more just thinking that banks already provide that to government. Soto go and take the next step of disintermediating a lot of the banks from the government purpose of providing bank accounts, doesn't seem that unlikely to me. It may actually be a tool to make politically getting rid of cash easier because then you can go make the claim, "well, all right, everyone has a right to a bank account", which of course they won't, as governments will pick and choose who can actually access banking services. But it's a politically more palatable solution to getting rid of cash, than saying, "oh by the way, all these private companies now can now hold you hostage."

Peter McCormack: Okay, so let's get into some of the technical details. Let's make the assumption that Libra will happen, that it will launch, they will pass whatever regulatory requirements there are and let's just look at some of the technicals, because you've both covered this, especially you Jameson in quite a bit of detail. Both of you actually identified that there will be a scalability issue here. What are the main issues with scaling that you see Jameson and then you can tell me your side, Peter.

Jameson Lopp: Well, I mean, they're still doing all of these transactions, "on chain", AKA in a database that the transactions have to be broadcast around all of the nodes in the Federation and at least have to be temporarily batched up into blocks, so that the validator nodes can come to a consensus about what the different state changes are in the system. So you're going to be using fair amount of bandwidth and disc I/O and of course disc storage because you're sending all of this data around to everybody who's operating on the network.

So they were fairly, I guess conservative, as they put for the minimum specs for a node in the white paper, which still seemed feasible for like a regular human to potentially run a node on the network, I think going for a single core and a few gigs of RAM and a few terabytes of disc space. They were, however saying that it's probably only going to be able to do like a thousand or so transactions per second and of course, they also seem to only really be worrying about keeping up with the current state of the network there.

As far as I recall, there is no real talk about requirements of syncing and validating the entire state of changes from the beginning of time, because I believe the assumption is that any new node or validator that joins the network will just find a recent point in ledger history and validate the signatures on that from the other nodes on the network and say, "okay, we can just start from here and assume that everything up to that point was correct", because it's a fairly trusted Federation of known entities.

Peter Todd: Yeah, I would agree with all that. In addition, I would add that certain very low level cryptographic mechanisms that they're using right now are actually even worse than Bitcoin in terms of scalability, because they don't actually batch transactions into blocks and they don't currently have a mechanism from what I've seen, where it could actually properly generate state changes in parallel.

So an example of that is if you have a hundred different people spending their money, in theory, you should be able to distribute those transactions from those hundred different people, across multiple different computers and multiple different cores. In the low-level cryptography that they're currently talking about with the way their Merkel tree transactions work, it doesn't actually let you do that because every transaction operates on a single state change.

Now these are not unfixable problems, but I think it goes to show that they're just not taking any of this stuff seriously and not trying to build a scalable system from day one. I'll bet you the reason why they're doing this, is they don't actually have long-term plans to decentralize Libra or do anything like that. If you have this unscalable centralized system, that's good enough and it's makes a lot easier and that's totally okay, if you expect you to have say, 10 different people in control of the entire chain.

Jameson Lopp: And this of course doesn't even begin to get into all of the questions around the smart contracting system and how that's going to work from "gas standpoint."

Peter Todd: Yeah, I've seen in corporate clients of mine, this tendency to just reuse Ethereum tech over and over again because it exists and it's relatively well understood. It might not be very good, but who cares? I mean this is effectively a tech demo and I think the smart contract part of it, is mostly something that they're kind of just adding on to give, A, something for the technical people to go do and B, hopefully divert some attention from the real political issues around Libra.

Peter McCormack: Have they not reached out to you Peter and asked you to consult on the project?

Peter Todd: Heck no! Actually I don't actually know anyone who has claimed that they're working on it other than people listed by name. So it may be something where they kind of went around to academics and tech people who they knew would be okay with their political ambitions.

Jameson Lopp: Yeah, I actually do know someone who is working on it, though I won't name them since I don't have permission to do that. But I do believe from my understanding that they went around to a number of the "Blockchain companies" in the space. Basically what Peter was saying is they were looking for the folks who would be at least somewhat amenable to working on a project like this, but still had some general idea of how the tech in this space works.

Peter McCormack: All right, well we'll come to the smart contract stuff shortly. One of the things I wanted to ask you about Jameson, is that one of the parts of the white paper says that over time membership eligibility will shift to become completely open and based on the members holdings of Libra. So firstly you said that like proof of stake, when you talk about the members holdings of Libra, are you talking about any user or are you talking about the validators?

Jameson Lopp: As far as I can tell, this was specifically referring to Libra association members, which ought to be synonymous with the validator node operators.

Peter McCormack: Okay, and then you said this is similar to ETH proof of stake, but that also has its own problems that haven't been solved. What are the main issues with proof of stake that haven't yet been solved?

Jameson Lopp: Well the longest running issue is kind of the nothing at stake problem, though that might be somewhat additionally complicated with Libra, due to the fact that these Libra coins are also backed by fiat and other assets that are being managed by the members of the association.

But even beyond that, there's kind of a question of, well how would the Libra association balance it if one member started to hold an inordinate amount of the assets. This is kind of where we go back to one of my earlier questions, which is how are the politics and the bylaws and the other internal functions of the association going to work, because they could very easily have major ramifications on how the technical aspects of the system work.

Peter Todd: I would point out, that from a tech level, this notion that anyone with Libra could participate in the consensus, needs far better technology than what they're proposing. The simple reality is if it doesn't scale, you're not really going to be able to feasibly run nodes if Libra actually catches on in a big way. If you can't do that, you're not participating in the consensus, simple as that.

Jameson Lopp: Well they could very easily just do what people are doing already, which is a farming out their stake to other people to basically do "pools of stake." So they're not going to run nodes, just like a bunch of people who are hashing on Bitcoin don't run nodes, they just let somebody else take care of that problem.

Peter Todd: Which I'd point at a server terminology thing. I wouldn't say that someone in that position is really participating in the consensus, when they have no real way to run a node and much the same way as they've argued before that people with hashing power are not truly mining Bitcoin unless they have the ability to run a node, at least double check what's actually happening.

Jameson Lopp: Yeah I think for a lot of those people and entities, it's really more of just an investment, than an actual interest in participating in the ongoing evolution of the ecosystem.

Peter McCormack: No one's going to be able to become a validator unless they're authorized anyway. It's not like Bitcoin where anyone can join this, you'll have to be authorized anyway.

Peter Todd: Well from a sort of tech economic point of view, that's definitely a true statement because as long as it's backed by some real thing, the people who are actually in control of the actual assets that Libra is backed by, they're going to have to have some kind of legal agreement which says, "well, basically if you validate or see something crazy, we're just going to ignore your Blockchain and figure out some other way to figure out what the ownership of the assets we hold in trust actually means."

Jameson Lopp: This is kind of the crux of the issue and the vagueness of some of the descriptions of how the system may work in the future, is because I recall the white paper refers to that issue and says that they're exploring, I think it was "market based options", to try to get away from having just a handful of those entities that are going to be maintaining that basket of goods. I'm not an economist, but I have no idea how that would work, it seems very pie in the sky. I'm not aware of any system in operation that does anything like that.

Peter Todd: A long-term business plan to deal with this basket of goods might be to go make it go away. You could imagine a system where eventually they decouple the Libra currency from having an asset backing bias and then when they do that, they get an opportunity to make a ton of money by selling off assets.

Peter McCormack: Right, so it's not backed by anything, but then they're just like the government and able to print Libra at will.

Peter Todd: Exactly, and just like the governments when they go do that, you make a ton of money because now you have a whole bunch of assets that you can get rid of.

Peter McCormack: Wow! Okay, well that's kind of scary. Also potentially some of the vague statements are to allow them to get to a point where they are live, it is a system that's operational and whether or not it is decentralized, centralized, it won't really matter because it's not like the majority of the users of Libra are going to be like Bitcoin users who kind of do a lot of research.

They find out how Bitcoin works and they go down the rabbit hole. The majority of Libra users are going to be standard Facebook users, WhatsApp users, who use it because the UX is great for them to buy something. They're just not going to care about these things that you guys care about, that you write about.

Peter Todd: I think it's maybe too early to speculate what we will actually use Libra for. For all we know, what might actually happen is a whole bunch of brand new money transfer systems pop up that are using Libra in the backend. They go say that they want to have Libra bring finance to the entire world, but for all we know that may be complete misrepresentation for what their actual business plans are,

Jameson Lopp: And who knows! I fully expect that a hyperbitcoinised world would be one in which the vast majority of people don't know how Bitcoin works or even care, much like the vast majority of people don't know how any of the current monetary systems work. They just want to use it.

Peter McCormack: Yeah, but there's enough people like you Jameson and you Peter and other people involved in Bitcoin, keeping it honest in the background, keeping it as decentralized as possible. Whereas with Libra, because it's corporate owned, it doesn't have that same kind of culture and group of people who care. All they will care about is Facebook and Calibra itself.

Peter Todd: Let's suppose, for sake of arguments, that somehow 100% of the people who own Libra in the world, cared deeply about de-centralization. Even in that scenario, because Libra is fundamentally controlled by a central set of entities, those people wouldn't matter. Nothing they did could the fact that Libra was centralized, other than sell their Libra and exit, at which point, that initial condition is no longer true and it's still centralized.

Peter McCormack: Also it'd be interesting to see if people actually care for Libra, if they want to use it. One of the difficult things with Bitcoin and spending it, is you do have to do that transaction in your head, what's the conversion? What am I actually spending here? I know it's likely that the Libra wallet will show you how much Libra you've got and it's going to do the conversion into your local currency.

But if say they integrate this with Instagram and you're looking at a photo and you have the opportunity to buy from it, firstly, are they going to force people to accept Libra? Because if they do they almost certainly will have a lower conversion rate if they don't offer say PayPal and credit cards alongside it and will people actually want to use it?

Peter Todd: Well one thing they can do is make the basket of currencies totally fixed. Then in the UI because it's fixed, essentially hope that the changes aren't too much and get away with ultimately lying to people by just making the UI localized, so that you go see some dollar value. Now I suspect you might get away with that for a bit, at which point... Like Bitcoin wallets, tend to go show your amount of Bitcoin or your currency equivalent, like US dollars and so on. I personally usually have my Bitcoin wallet showing how many of the currency I care about, I have worth. 

You can get away with that in a Bitcoin wallet, because it's so well known for its users that Bitcoin is a separate asset. I suspect if Libra did that to avoid the UI problems that you're talking about, they would get a lot of pushback. So to me this suggests that either they haven't thought this through very fully or they really genuinely want to push an entirely new currency and I think that would raise a lot of questions about what their actual business plans are.

Peter McCormack: Well their big problem at the moment is dropping users on the Facebook network. Their model is, I don't know what, probably 90/95% advertising model. I think they lost something like 22% of users, something crazy, maybe 12% or 22% of users in the US, so they've got to do something otherwise they're in big trouble and moving into money seems like an option for them. Okay, so let's move on to the smart contracts part.

You've written about this Jameson, they're going to be using their own programming language called Move. Have you had a look at this? Do you know much about it? It seems to me that again, that's another thing that's pretty vague and I think you said that your expectation is that it's going to be quite restricted with what you can do with it to begin with.

Jameson Lopp: Yeah, I tried to look into it and I couldn't really find much information. I seem to recall that, I think it was Elaine Ou, who tried to write a smart contract and did not have a very good experience trying to do that and it's just because it's so early. I think the only things that they've really written and made available are just simple transfer and transaction functions right now. I don't think it's really possible to do anything interesting on the platform.

Peter McCormack: And Peter, your worry is that the smart contracts in Libra will have similar failings to smart contracts in ETH and actually it's better in a UTXO model?

Peter Todd: Well I think the bigger question actually is, will you actually be able to use the smart contacts at all and from the sounds of it, at least for the foreseeable future, they're not going to allow people to write their own ones. There will be a fixed set of contracts that you're allowed to use and nothing else. In general, the UTXO model has definitely been much more successful at writing smart contracts that actually work.

But I'd also point out that of the smart contracts that you actually need, with a little bit of clever crypto, you don't need much more than just signing and it's amazing how much you can go do with some clever ECDSA maths. Ted [Inaudible 49:00], he's done a lot of work on this and it may be the case that they don't actually need any of these complex languages to do the vast majority of what people want smart contracts for.

Peter McCormack: So you're thinking, that might be like predefined functions?

Peter Todd: Yeah. Facebook has made statements along the lines that you won't actually be able to write your own code.

Peter McCormack: Interesting. I guess they want a system that other people will build upon to grow the network. It's kind of interesting. What do you think the expectation is Jameson, with the smart contracts? Is this just to allow people to create wallets, maybe to create more complex e-commerce environments? What do you think?

Jameson Lopp: It wasn't really clear how far they wanted to go with it, other than the fact that it did seem clear to me that they want to support a variety of assets. Now does that mean that they want any developer or at least any authorized developer to create their own or be able to create their own digital asset on this platform? Maybe, it's hard to say how widely used they want the platform to be outside of the Libra coin. I guess it kind of goes back to our earlier point that I think Peter made, that some of this extra stuff may just be diversionary tactics.

Peter McCormack: So what, you think that they might allow you to create almost like colored coins on Libra?

Jameson Lopp: Well, because the white paper specifically states that the Libra coin itself is not a native asset. It is in fact defined using the smart contracting language as the first of supposedly many assets.

Peter McCormack: Interesting! So I guess they could be considering NFTs or in game tokens?

Jameson Lopp: Yes, theoretically.

Peter McCormack: Or even localized stable coins!

Peter Todd: Yeah, it could end up being something like Blockstream Liquid, where there's tons of different stuff. Although, and I'll say this for Blockstream Liquid as well, the idea of putting all these things on one single chain, doesn't really make that much sense. My explanation for why they talk to the smart contract functionality is probably, in the nicest explanation, just a marketing thing where they knew if they didn't include it, that'd be made fun of by a bulk part of the community, so they had no choice.

Peter McCormack: Okay, interesting. So let's talk about the KYC/AML side of things. Again, going back to Eric Wall's article, it seems like they're attempting to push the KYC/AML onto wallet providers and exchanges exploiting a loophole in that. What do you think is going on here, because I think it was in Eric's article, he put a screenshot from Libra and law enforcement, where exchanges and wallets will need to follow applicable laws and regulations.

Do you believe this is just them trying to exploit a regulatory loophole or do you think there's anything in this in that, because they're trying to create a globally distributed cryptocurrency, that trying to create something that covers every regulation, in every local market is actually quite difficult?

Jameson Lopp: Yeah, I mean it would be insane to try to do AML/KYC stuff at the protocol level. So I think they at least realized that. Now the bigger question I think becomes how tightly controlled is the access to developers or to other entities to create wallets and other software that may be subject to those type of regulations. How possible would it be, for an entity to create like a rogue wallet or rogue application that did not abide by any of the laws of whatever jurisdiction it was operating in.

Peter Todd: So we've seen statements from them, really pushing back on the notion of having any privacy on the core chain itself. One way to interpret these statements would be to say that they know what they're doing is going to be up against governments and law enforcement around the world, who would rather have full control over this and they're taking the next best option, which is to make a system where there's no privacy.

Yes, in theory you have the freedom to do things and if I were in a position of say creating a multi-jurisdictional currency, I would probably punt on the issue of control, back to the local jurisdictions where wallets would conceivably be enforced and so on and then reduce my exposure so to speak, by just making everything public. Does this actually give regulators what they want to know, no, but it may be a reasonable compromise.

The accounts based model rather than UTXO based model, might be part of this because accounts fit better to this notion of having zero privacy. Accounts like Ethereum has significantly less privacy than Bitcoin in practice, because people use it for accounts. So everything is absolutely tied to one single account.

Peter McCormack: I think it says in the white paper, "the Libra protocol does not link accounts to real world identity." Jameson, you pointed out that you think this sounds good, but you also said, "does that mean that developers could build privacy tools on top of Libra?" I mean my expectation is that it's unlikely, but what do you think?

Jameson Lopp: Well, there are people, at least in the Ethereum space who have built hierarchical deterministic wallets or who have built mixing contracts and whatever. So theoretically, if the smart contracting language move was good enough and allowed extremely arbitrarily complex types of interactions like that, then developers could write similar privacy enhancing functionality. Would even be possible to do like ZK-Snark type of stuff?

I think some people have messed around with that with Ethereum. That's why I think the openness of the platform to developers is going to be a big question, because you know that whatever they open up, somebody is going to try to exploit it and try to improve the system to actually make it more usable.

Peter Todd: See, I strongly disagree with your argument there because I don't think you need any more openness and people are able to do transactions without permission. You don't need smart contracting to get very good privacy on these kinds of systems. I mean at minimum, a couple mixers is good enough to seriously cause problems to law enforcement. It just is not difficult to get privacy.

It'd be nice if, for instance, you could run ZK-Snarks on these things, but it's not a requirement at all. It's more than enough to simply have people get access to the coin and I suspect that might be part of their strategy around communication. To say all these big things about how they want to be legally compliant, when they actually have no intention to. Yet simultaneously because getting information out of this Blockchain is going to be so easy, you can still make money off advertising and selling people's personal data and so on. So from their point of view, they've got a lot of bases covered.

Peter McCormack: Can you envisage a scenario say where a hard wallet provider supports Libra, and if I was to buy, say a Ledger and Jameson, you were to buy a Ledger that isn't tied to anything and I was able to transfer Libra to it, that I would then be able to transfer it to you anonymously?

Peter Todd: The question is, can someone have that wallet at all without getting permission from the Libra Federation. That's your fundamental question. If the answer to that is yes, then someone can go build a mixer and similar technologies and then we can go have proper privacy. If the answer to that is no, none of that is possible. 

So far, I don't really know what the answer to that will be. But I do know that in their technical docs, they talk about having authorization just to connect to the Libra Federation at all. So that might be a tool to really lock those down, where you'll have to set up a form, to even talk to the Federation nodes and the transactions. I just don't know.

Jameson Lopp: I definitely think that there has been a few statements saying that it should be possible for peer-to-peer transfers to happen inside the network between wallets that don't have AML/KYC on them. But of course that's not going to be possible until there are a wallet's other than Calibra, because we know that Calibra, as a wallet, is going to be doing AML/KYC.

Peter Todd: Well, I put nuance on that statement. It's not a question of whether there are wallets, it's question of what does the tech actually implement? Currently what they've actually implemented appears to be a system where you have to get permission to even submit a transaction. Will that be true in the future? I don't know. But what they've actually implemented has the property.

Now there could be a world where there's tons of different wallets around and tons of people can write them, but you still don't have permission to submit transactions, until you fill in some government forms to provide your ID. Just because someone turned the wallet doesn't mean that the credentials to actually connect to the Libra Federation are freely available.

Peter McCormack: Hmm, okay. Well let's talk about gas. I've always felt like gas and transaction fees is kind of a poor UX, as somebody who has been a UX designer, not knowing the fees you're going to be paying for a transaction. You wrote about the gas side of things, Jameson, I'll take the quote from the white paper that, "many parts of the core logic of the Blockchain are defined using Move, including the deduction of gas fees to avoid circularity. The VM disables the meter and the gas during the execution of these core components." Now your statement is that this is pretty dangerous?

Jameson Lopp: Anything that disables the metering of gas could potentially be exploited as some sort of runaway, infinite loop type situation, a sort of denial of service attack that causes the nodes to lock up for a long time. We have seen similar issues, well at least in Ethereum, there were issues because of miss-pricing of gas for a while. So there's also questions around, how are the different operations going to be priced within the system.

Peter McCormack: I mean it's going to be a difficult thing for users to get their head around because most people are used to traditional payment mechanisms of any... You know, if you consider VISA or MasterCard fees as gas, that's something that's passed onto the retailer and it's not something that the user themselves ever has to think about.

But in the world of cryptocurrencies, it's been flipped, so that the user has to think about this and I've always kind of felt that it's poor UX. Do you think they're going to flip this onto the users or do you think they will follow the kind of traditional payment rails where it's flipped onto the person accepting?

Jameson Lopp: Well I was less concerned about that as I was just with the general security of the system and how susceptible it is to edge cases, people who are trying to make these nodes fall over. But it did sound like the whole issue of the fees and how the dynamics of fluctuating gas prices would work, sounded very similar to the way Ethereum has worked.

There were also actually some conflicting statements between the protocol white paper and the Calibra white paper for example, where the protocol white paper, was saying that, "there will probably be times at which we have to automatically adjust the fee market to make it expensive for people to spam the network." But then on the other hand, the Calibra wallet team seems like they're prioritizing trying to make it as cheap as possible for people to use the network. So it'll be interesting to see how that dynamic plays out over the long term, if this system becomes operational and becomes widely used.

Peter McCormack: It's kind of a weird scenario though, if it ends up becoming something that's used primarily for eCommerce across the Facebook ecosystem, because you're not really going to want to discourage payments?

Peter Todd: In Canada, interact payments wind up charging the person making them and no one seems to care that much.

Peter McCormack: Are they fixed?

Peter Todd: Yeah, I think they tend to be usually like 5 cents per transaction for most people's type of bank accounts. Some of them will go have a bunch of free and then go pay out of your monthly fee for your bank account. I mean that part of the UI, I don't think it's actually a big issue. I think the bigger issue is just when the price changes unexpectedly and in a centralized system, that's not that hard to prevent because you just get enough capacity to stave off any foreseeable attack.

Peter McCormack: Well, we have something similar with booking flights in the UK, it's primarily there. If you use a debit card, you don't have a fee, but they will pass a fee onto you if you use a credit card. But it's usually a fixed price and it's quoted. So here's your price paying debit card, and here's your price, paying by credit card. Okay if I pay by credit card I've got to pay an extra £2, £3 whatever it is.

Peter Todd: Remember in an account space system, the price is fixed per transaction for the type of account that your wallet is using. So every transaction you make will have a fixed amount of gas that it needs to use, thus a fixed price in terms of Libra. With the one exception being, that if they go and raise the price of gas to try to go stave off an attack, which I don't think in a centralized system is all that likely, because they'll just get enough capacity, that that will be a very rare emergency occurrence.

Peter McCormack: And getting enough capacity is easy enough?

Peter Todd: Sure! Pick an amount where the amount of money an attacker would have to spend to spam you, is sufficiently high that there's better ways they could go and attack you. I mean PayPal doesn't really seem to have this issue and they probably solved it by taking some very large transactions per second and buying enough servers to handle that.

Jameson Lopp: So this may very well become another question of how the association itself operates. If they do find themselves to be bursting at the seams with so much demand that they can't keep up, then I expect it'll probably become a political thing where the organizations that are involved have to decide what they're willing to pay to operate these validator nodes.

Peter Todd: Remember the Internet already has this issue with internet routing, because the way the routing protocols work is you have to have a copy of the full routing table if you want to be a core internet routing node. Periodically we've had cases where the number of routes goes up and a whole bunch of the internet routing hardware becomes obsolete, because it simply cannot maintain all the routes in memory at once.

The internet as a whole seems to have handled this with too much politics, so I suspect from an external observer, we're not going to notice much. Internally, people may very well get kicked out of the association for failure to spend enough money, but for external users they probably won't notice any real changes.

Peter McCormack: So the interesting thing is that, if I try to summarize what we've discussed is that this is highly complex and they're facing a number of challenges and a number of technical problems that other cryptocurrencies and Blockchains have tried to solve, that haven't. That actually there's a lot of risks in this with them in launching this. For what they're actually trying to achieve, there's probably much simpler ways of them doing this.

Peter Todd: Well, if you me asked me to go and make a centralized Blockchain run by a Federation, so that you could go and make a new type of asset backed digital currency, the architecture I would pick is actually not that far from Libra, especially if I was under time constraints. If you told me to go do that in say six months, I'd probably pick some off the shelf, Ethereum or something similar software and just reuse it, which is kind of what they've done so to speak. I think you'd do a lot better if you actually cared for scalability.

But to me, this type of architecture doesn't seem that crazy and it's easy to imagine how it would have come about, given how big corporations work and how teams work and so on. So I'm actually not surprised and because it's a centralized system, it can always upgrade it later, they don't have to stick with what they had on day one, they can always change things later, especially in a centralized system where clients are validating.

I mean at the extreme, if they upgraded all the tech back end, they could even run a separate fake Libra validation chain, so to speak, that created fake data structures that looked like they were like client proofs essentially, even though they were completely invalid, to support older clients and trick them into accepting transactions. I mean, this is what you could do in a situation where clients don't validate. I'm actually not seeing too many show stopping technical issues there, especially given they could fix them.

Jameson Lopp: I think that in general, we agree that the system, at least as it is described to be operating right now, is functional, can "work". The bigger questions are these lofty goals, the ambitions they have to try to solve a lot of problems that no one else has solved before.

Peter Todd: Well, I would so far to say that the notion that they're going to make Libra decentralized, requires them to go do something which is indistinguishable and which is to go and make it no longer asset-backed. To go put that in day one just is really dishonest and there's no getting around that.

Peter McCormack: I don't think any of us really and truly expected that Libra coin would be a decentralize, permissionless system.

Peter Todd: No. There's just no reason for them to do that.

Peter McCormack: Yeah, my expectation is that it's a marketing term. Whether that's a marketing term to, I don't know, people who already operate and work within the cryptocurrency space or whether it is a... I don't know! I expect it's just some kind of virtue signalling to... I wouldn't say people like you because you guys will tear it apart, but to the broader cryptocurrency ecosystem.

Jameson Lopp: Well if I recall correctly, I think that I've seen the timeline as something like five years after launch to try to reach some of those lofty goals of proof of stake and improving the basket market mechanism or whatever that would be. Five years is a very long time. So it's almost one of those things that you hear a lot of like the ICOs of 2017 talking about, "we've got our like five year roadmap and this is where we're going to go". Will this system even be functional and operating five years from now is a big question, much less like are they going to be able to solve some of these crazy problems that seem highly unlikely.

Peter McCormack: Well that's very much the Ethereum MO, that these lofty problems that they've had is always a few years away. Essentially keep some kind of blind faith while people put up with and accept things which aren't really great. So I guess they're just kind of, like you say, they're just following what everybody else does.

Peter Todd: It would be fascinating to hear some conversations, those talks with Ethereum developers and other leaders have been having with their lawyers about what on earth [Inaudible 01:09:38] public given that they don't really have clear ideas of where they're going to go next.

Peter McCormack: All right, before we start to close out, is there anything else that stood out for you that you'd want to discuss here now and raise?

Peter Todd: Nothing comes to mind to me. I think we covered it pretty well.

Jameson Lopp: I think that's about it. I mean there's a lot of open questions and that's why I think there are a lot of potential paths this could go down. I've talked to some of the people who are working on the project and they do seem quite genuine in their aspirations. Only time will tell exactly how lofty these goals really are, like how much are they actually going to be able to achieve.

I think the biggest disappointment was how vague a lot of the stuff in the white papers are and how early it was released. It does seem like the reason for that was probably more political and just trying to get something out there in the public, to let politicians and regulators argue about for a while. Whereas the rest of us, and especially like the technical folks are more just sitting here scratching our heads trying to fill in the gaps.

Peter Todd: It is interesting the strength of push back. You look at how there's a bill apparently in the US Senate or US Congress to go and ban tech companies with revenues of $25 billion or more, from being able to participate in finance at all, being able to issue these coins at all. I mean it's quite remarkable just how strong that actually is and people would go pushing around bills like that.

Peter McCormack: Well they fear losing control.

Peter Todd: Well, if you fear losing control, sometimes the better thing to do is don't advertise to your opponents that you fear that.

Peter McCormack: Yeah! I mean if I was to try and summarize this conversation, it feels like to me Peter, that your main view is, stop lying, stop being so disingenuous, stop conning people, because we've experienced so much of this already with other scammers in the space. For you Jameson, it's slightly different in that you seem to be more on the, can we have some more information? Can we ask some clarity? Because without this we don't really know what you're up to. Is that a fair summary?

Jameson Lopp: Well, if this is a scam, then on the bright side, it seems like the only folks that are really going to get scammed are the folks that are buying into the Libra association with their $10 million fees!

Peter McCormack: Yeah, I guess so.

Peter Todd: And also as much as I'm critical of them for dishonesty, if they advertise Libra honestly, I honestly really wouldn't care that much. If you want to honestly create a centralized corporate coin that's backed by a bunch of US dollars. So what, go ahead! I mean it sounds like a more reputable version of Tether and I have no issues with Tether, other than the fact that they're policing your money because those assets are likely to get seized. But if that's a risk you're willing to take, go for it, have fun!

Peter McCormack: Would you use Libra?

Peter Todd: I wouldn't trust it for very much money. I wouldn't trust it for more than I could afford to lose, but hey, if it works for my payment needs, I would probably use it once in a while.

Jameson Lopp: If there was a non-AML/KYC wallet, then I might find a use for it. But I'm definitely not going to be signing up for any new services that require that.

Peter McCormack: Yeah, well that's exactly what I expect from you Jameson, without a doubt! Okay. Well listen, I don't need you guys to close up and tell people how to get hold of you. You've both been on the show before, I'll share that out anyway. Jameson, you had a couple of people catch you up, but you're now back in the lead, that's your fourth appearance, so appreciate you coming on! Also I appreciate you coming on Peter and yeah, hope to catch up with you both soon.

Peter Todd: Thank you!

Jameson Lopp: Thanks for having me!

PodcastPeter McCormackComment