Peter Todd, Bitcoin expert and cryptocurrency consultant joins me on the show to talk about Signal’s recent decision to use MobileCoin rather than Bitcoin. Is Signal selling out? We talk about why MobileCoin is centralised, why it was a bad choice, and why they should have used Bitcoin and Lightning instead. We also talk about directions with private messaging going forward.
Peter Todd links:
- Twitter:@peterktodd
- Mastodon
- Website: PeterTodd.org
Sponsors:
- Swan Bitcoin
- Hodl Hodl Lend
- Compass Mining
- Unchained Capital (code LIVERA)
- CypherSafe (code LIVERA)
- CoinKite.com (code LIVERA)
Stephan Livera links:
- Show notes and website
- Follow me on twitter @stephanlivera
- Subscribe to the podcast
- Patreon @stephanlivera
Podcast Transcript:
Stephan Livera:
Peter, welcome to the show.
Peter Todd:
Thank you for having me.
Stephan Livera:
So Peter, I know you’ve had some commentary recently about Signal’s, new shitcoin mobile coin. So yeah, let’s just kind of get into that and maybe set some of the background here. What’s your kind of initial impression of Signal up until now?
Peter Todd:
Signal for the point of view for a user using it, you know, Signal’s convenient to use a phone number to use it, set up kind of automatically works so low, from a more tech point of view, like this is a lot of things that they’ve done that really cut corners to allow that experience be quite as easy as it is. And you always kind of wonder, are these really like quite the right trade-offs good example, being phone numbers, the heavy utilization of them where that’s just dubious for a lot of reasons, but the trade-off there was, it made getting people on the platform very easy. So, there’s been controversy about this, but then you go look at this, a new coin and I think the trade-offs they make aren’t quite so dependable.
Stephan Livera:
Yeah. That’s a really good way to put it because I think the theorist might say, Oh, look, everyone should just be using GPG and, have your public keys out there and do it. But then the reality is would we have gotten to the, I don’t know the exact numbers, but let’s call it 40 or 50 million monthly active users that Signal has. And I guess that’s always that trade-off and I know Signal also had a big jump in popularity only recently because of that recent drama about WhatsApp data being shared with Facebook. And so I guess that’s probably the counter argument would be something like, okay, well, without this fund number, trade off and without making it
Peter Todd:
That’s the counter argument, but then the counter counter argument is it wouldn’t be that hard to give people another way to get on a Signal without a phone number. We don’t have to go to perfect decentralized everything just to reduce some of these risks.
Stephan Livera:
Yeah. And I think Moxie’s argument — Moxie for people who don’t know he’s one of the founders and one of the lead maintainers and contributors of Signal. And I think now I’m not, I’m not obviously not going to be at represent his argument perfectly, but I think his argument was something like we should have a slightly more centralized infrastructure. So long as that gets more people that can get more people in and you can sort of make certain trade-offs around that. What’s your perspective on that?
Peter Todd:
I mean, the funny thing is the, this mobile coin, I think shows the dangers of that centralization more than it necessarily needed to. And why I say that is because centralization with competent people running it, it’s more dangerous in certain scenarios, but it’s less dangerous than it could be. Whereas mobile coin is this thing where hang on a second, why are you poking the bear? You know, why are you adding this very, very controversial thing. Controversial, not just from sort of government point of view, but also even within the crypto community point of view, to the Singh that already the vulnerability of centralization, it’d be like, I mean, after all I myself, I run a centralized platform, open timestamps, it’s absolutely centralized. And part of my argument for why that’s okay is because I’m not going to do controversial things with open timestamps. You know, if I added file sharing to open timestamps, that would really undercut my own arguments for why it’s okay that open timestamps is centralized.
Stephan Livera:
Yeah. And so there’s this whole argument. And I think I probably agree with that, but I guess I’m curious what you think there’s this argument that Signal have essentially sold out their user base by creating and marketing this cryptocurrency.
Peter Todd:
Yeah. Unfortunately I think you can make that argument very easily and what’s annoying about it is they could have gotten from the user’s point of view, they could have gotten all of the advantages of a centralized infrastructure for payments, whether it would actually creating a new coin that wasn’t, that part of it wasn’t needed, but creating a new coin gets them a ton of money and by them. I mean, people find mobile coin, which indirectly sounds like it’ll help Signal, but it’s not a hundred percent clear, but long story short is creating new coin gets a massive monetization opportunity. And it just doesn’t necessarily make very much sense,
Stephan Livera:
A real shame in some ways, because I think many Bitcoin people have been I guess helping new people get into Signal to say, Hey, instead of using normal texting or Facebook messenger, why don’t you get on Signal? And then for some of us, it feels a little bit kind of like a betrayal because we’re trying to build the scene and build the community and then this sort of turning around and it’s unfortunately a sad trend to the wave scene as well.
Peter Todd:
Let me make one thing clear. You can say the exact same thing for ethereum people, it doesn’t matter what coin you happen to go back. The fact that they’re creating this brand new coin with all the problems that has involved with it to then, get money by sort of tying it into this other thing, which we would like to continue to existingly Signal, that’s sort of a betrayal of everyone involved at Signal. Everyone who’s been pushing it.
Stephan Livera:
Yeah, exactly. And like, yeah, not just Bitcoin, but just anyone who’s been trying to, I guess, give people some, whatever privacy trade-off, but still better than what they were normally doing with SMS or with Facebook messenger or whatever they were using before. And I think historically the problem is there’s been a lot of scams in the crypto industry or basically everything other than Bitcoin should be default assumed to be a scam. And unfortunately it seems like the same trend is playing out here where perhaps some insiders get early access to the coin and then dump it onto retail. What’s your view on that and how that has played out over the last few?
Peter Todd:
I think you’re exactly right. Every time anyone starts a new coin, there usually are not good reasons for it. And an example like this, where like you’re not, they claim otherwise, but in fact, it’s mobile coin centralized the way that technology works, with the stellar consensus and their addition of SGX, it’s a centralized system end of story. And they could have gotten the advantages of that by, without creating a new coin. But by doing that, you add in all these really complex legal questions. And also the fact that you’re going to wind up having very big winners and losers in the buying and selling of this thing, like for technical reasons, occasionally you need to create new coins. That’s just a fact, but this is not one of those cases. They did not need to do that. And it creates so many, could just create so much problems that Signal’s kind of attaching themselves to, and like, frankly, I would not be that surprised if Moxie wound up in jail for this, both for the whole sort of financial fraud side, but also the sort of very clear fraud of saying something decentralized when it just isn’t, that’s something that could backfire in a very bad way.
Stephan Livera:
Yeah. Okay. So, and obviously we’ll get into the, should they have used Bitcoin and Lightning stuff, but first let’s talk a little bit more in detail about what some of these components are, for example, what is Intel SGX? What is that?
Peter Todd:
Well, it’s this proprietary security solution offered by Intel that does something quite specific. It allows a creation what’s they call an enclave, which is, an analogy to real-world enclaves sort of sovereign if you will, execution environment within your CPU. And what that means is that within an enclave code can be run that if the security assumptions proof be correct, Intel’s implementation of it works. The owner of that CPU cannot modify or observe what happens in that execution environment. So it’s sorta like having a, a little CPU within your CPU, the implementation doesn’t work that way, but that’s sort of what it’s like. And a really key thing that SGX can do, which is quite hard to get is not only can this code run in that environment, but it can create a signature attesting to the fact that this environment is present in secure in that signature can then prove to a third party that if what Intel’s doing is trustworthy, then the code that they’re talking to or interacting with in some way, communicating to whatever is actually within an enclave and it’s therefore not able to be modified.
Peter Todd:
So why is that important to move upward? Well, that’s part of their privacy arguments. You know, they use these signatures to prove to other nodes that no one can go observe what the transactions are actually doing.
Stephan Livera:
Yeah. And so in some sense, we can say, this is like a trade-off they were choosing to try and say, okay, because we made this trade off, that’s giving us more privacy. And so in essence, one, I guess, counter argument that you’re making here is that you could have used the SGX in another way, rather than creating a, with another coin. That’s essentially what you’re saying, right?
Peter Todd:
Yeah. It’s definitely part of it. And the funny thing about this is the trade offs they make, they made with this are not getting anywhere near as much advantage as you might expect. So what Signal has done is they’ve taken the Monero, I there’s sort of the Monero way of making coins more private, and they’ve actually implemented this. So they already have the sort of privacy that Monero has. And then they use the SGX to eliminate the statistical analysis that’s possible on Monero, but by using SGX, well, your whole system could be shut down by Intel, ultimately because SGX in its current form effectively to run code on SGX, you need Intel permission. So it’s a strange kind of set of trade-offs they’ve made. And the implementation seems to be even more restrictive than that in, I mean, to actually run a full node on this to actually be a part of this consensus, well, they don’t actually tell you how on the, on the repo, in the source code, it’s actually not clear how anyone’s supposed to go do that.
Peter Todd:
You know, they ask you to go send them an email, which I did, and I haven’t gotten a response back. So, is this something that you can actually participate in? I can’t go tell you yet. And the fact that’s even true is pretty suspicious to begin with.
Stephan Livera:
Right. We used to in Bitcoin is open source ethos that anyone can read the code, anybody can participate and anybody can run their own Bitcoin full node. And so as you were planning out earlier, this is you were mentioning earlier how there’s it’s kind of like loosely speaking, it’s like a meshing of Monero and then potentially the stellar and ripple consensus protocol. Can you tell us a little bit about the difference between that versus the kind of the Bitcoin approach, the Bitcoin way?
Peter Todd:
Well stellar and ripple in a nutshell, they’re essentially traditional consensus systems. And what I mean by traditional is there is a set of notes that you trust to maintain consensus, and those nodes choose among themselves, what transactions have happened. And what’s very strange about the way Stellar and Ripple can advertise this is their particular flavor. If this search traditional consensus is meant to kind of allow people to pick different, you know, different trust sets or unique node lists is one way to — there’s one term that’s used for this. The fact of the matter is if you and I pick a unique node list, it doesn’t overlap enough there’s scenarios where we will be at a consensus. So the logic unfortunately, is that everyone has to go pick the same unique node list or you risk serious serious problems.
Stephan Livera:
Yeah. And just to spell that out for people, why is it an issue if we are not in consensus?
Peter Todd:
It’s because then, your view of who owns what money could be different than mine, which allows me to spend the same coin twice.
Stephan Livera:
Yeah. And that obviously is a huge issue for any cryptocurrency that people are essentially trying to come up with you know an agreement on what is the correct or current state of the ledger who owns how many coins and so in this world where just sort of pushing that trust off to somebody else. And if you can’t, become your own, kind of make it so that in this decentralized way, we can all agree. What is the truth? Then you’ve got some fundamental issues here, don’t we,
Peter Todd:
And I’ll point out, I mean, ripple uses the term validators four nodes in this unique node list. And that’s a completely different use of the term than the way Bitcoin talks about validation. You know, when you talk about a validating node in Bitcoin, that node takes information available to it verifies that the Bitcoin rules are being followed and then effectually spits out a yes, no answer, you know, yes. Is this a valid block? No, is it not? And as part of that process, it also maintains a set of all unspent coins, so that’s very, very simple. And the consensus in Bitcoin is defined by most work chain. Now, most work goes back to proof of work, namely, who has, you know, which chain in total has essentially tried to find the most, winning hash solutions. You know, it’s a bit, a bit more complex than that, but, you can think of it as sort of work you apply, but what’s really important is that that definition of what is the most work chain is independent of who actually created it.
Peter Todd:
So the process of coming to consensus and Bitcoin is just getting as much information as you can, but what chains might be out there and then picking whichever one, meets that definition. So it’s very hard for you and I to have a different definition that doesn’t come into consensus, so long as we’re distributing blockchain data around freely, you and I essentially guaranteed to think that the current highest block is the same current block. And that’s just not true in something like ripple ripple. There is no way to decide which version of the chain is correct. Other than to just pick some people who get to decide that that is the unique node list. Inland, let them make that decision.
Stephan Livera:
And basically that’s just unsolving the problem that Fayette money that Bitcoin is doing, because that money is that system where, there are certain people who can, print new money or have access or privileged access to that ledger, where in Bitcoin, it seen like it’s an open ledger and anyone should be able to view it and verify it, and be able to, I guess the important part in some ways is like we’re saying Bitcoin, being able to reject an incoming to say, I’ve, — for example, if I send you something wrong, that’s not Bitcoin. You can say to me, no, Stefan, that’s not real Bitcoin. You, you didn’t send me a Bitcoin.
Peter Todd:
And, you could ask that question in a, sort of a very technical way. Like, what is it that makes my computer reject that fake transaction? Well, it’s because I have a validating note, it’s got access to all the same information everyone else in Bitcoin does, and it can do that same computation, that same validation, which gets to another problem with mobile coins plans. They want to have the system, which would scale to every single user [inaudible], well, that’s a lot of transactions. That’s a lot of data. And if you put all that data on the chain now for me to go validate that data, we’re talking about the number of transactions that if, if it caught on would be totally impractical for me to validate well, what’s their solution to that. It’s actually trust these unique nodes even more, not only are they saying, Hey, here is what the valid block is. They’re also saying by the way, just trust us for what’s actually in there.
Stephan Livera:
And I guess maybe to [inaudible] that maybe the equivalent in Bitcoin is look at all the lightweight client users out there. How would we compare that to, let’s say a mobile phone wallet, typical Bitcoin wallet that is a light client.
Peter Todd:
The way I would compare it is to point out that with Bitcoin wallets, while a lot of them let’s face, it are even completely centralized. You know, they trust the wallet, developers for information, for servers, et cetera, et cetera. A lot of them are not, it’s quite easy for me to run a Electrum server myself and just point my wallets, my phone wallets to my own Electrum server I don’t actually use phone wallets stuff ever, probably doing that, but the wallet I do use most for Lightning, I’ve just pointed to my own Lightning nodes. So I actually get essentially the same security just in a slightly different way. And, it’s something that’s quite easy to do. You know, you can buy kits off the shelf that just do this and just, just magically work. And because of all this variety in ways you can get to the same state, it’s much harder to pull off an attack, whereas the way they’re designing the software for mobile coin, there really won’t be that many people who have the ability to break this consensus, why the light clients, etcetera, etcetera.
Peter Todd:
And that’s just a dubious position for coin to be in. And I pointed out there’s even a political side of this, which for example, if you, if most of the people using mobile coin are for instance, using it with Signal, well, you also get sort of outside the consensus attacks where, suppose some government just says to Signal, well, no, you must go change the Signal software to blacklist some coins or even reassign ownership of some coins again. So we’ll have everyone else now, how do you really withstand those attacks when your user base is so centralized so many different levels? It’s just not a good idea.
Stephan Livera:
Yeah. And so I guess in some ways we can sort of, we can say, look it given your own goals, this is not the best way to achieve that goal. And so I think that’s another vector. You’ve also commented that they gone,
Peter Todd:
The one thing that their plan, I think achieves it meets their own goals, is making money. You know, that really is it. And for every other perspective, this does not make sense, even ignoring like centralization issues and stuff like this. It’s the technicalities just don’t make sense, but by creating a new coin, you get a lot of money and supposedly to date mobile coins going to about $40 million of investment. And I’ve heard rumors that their sort of pitch is market cap of a billion pretty soon. So, that is serious money.
Stephan Livera:
Yeah. So they’re expecting, over a 20 X on that. And so historically this has been a big scam basically in the industry known as the shitcoin waterfall where essentially shitcoin insiders create a coin with very questionable justification paper, thin justifications, they pump it on, or they sort of sell it on and give a deal, family and friends deal to their friends and to some VC investors and potentially some, quote unquote, crypto influencers who then go and shill it to retail. And then once they get it listed on some big exchange, that’s when there’s a big liquidity event. And then they are dumping into dumping onto retail basically. Is that would you say that’s basically a fair characterization of what has happened in this industry, in the quote unquote crypto industry over the last few years and potentially this is a similar what’s going on now?
Peter Todd:
Fortunately, yes. If you asked me to say, all right, what coins can you actually justify for tech reasons, Bitcoin, obviously, someone had to do the first one. I think Monero is an example where, because it’s meant to be decentralized cryptocurrency, it would be very hard to go pull off Monero without creating some kind of new coin for Z cash. You know, it’s the exact same argument for Ethereum ignoring all the other issues that have with Ethereum. I think you can make that argument and then quickly start turning into more and more dubious projects that are kind of copycats of everything I just mentioned.
Stephan Livera:
Yeah. And the other point that has to be mentioned here is if this coin, Bitcoin and Bitcoin is meant to be, well, Bitcoin is decentralized, but this, old coins, if it’s meant to be decentralized, then it should be able to be developed in a way that is decentralized. And you’ve pointed out that’s actually not true, that because of the use of Intel SGX, that’s not possible. Why is that?
Peter Todd:
Well let’s go back to, what is Intel SGX trying to achieve? The purpose of it is so that your node is getting, or, your is getting an attestation from the node. You’re talking to saying, I am running the software and you can trust me to maintain your privacy because this software doesn’t have code in it to violate that privacy. Well, that means that the actual software that’s running on that node, you have really two options. One is, is, is it’s bit, for bit identical and never changes. And two is you have a third party that is allowed to sign new versions of it, but then that third party is appointed centralization. And I haven’t delved into the details of exactly what Signal is doing there, but either option winds up being centralized, either you just can’t change the consensus at all, which then actually has the level of centralization will, all right. What’s something breaks, which inevitably will, how do we manage that? Or you get the level of centralization where someone has the ability to swap out the code, and then they have this privilege access. And that, it’s a very dubious position to be in. And I think it will invite regulatory headaches. That Signal just does not need.
Stephan Livera:
Yeah. And I guess to frame that in a Bitcoin sense, if somebody, for example, hypothetically somebody had proposed some incredible new change to Bitcoin and everyone agreed to it and it was going to bring about some great technology. Well, they could go and code that up and put that out there and say, Hey guys, let’s, soft fork this into Bitcoin and people could all agree, Oh yeah, this is a good change. We think this is, win, win, win all around. Let’s do the change. That’s something we can do in Bitcoin land. But unfortunately this is not even possible in mobile coin land.
Peter Todd:
I mean, I’ve been one of the people Who said over and over again, running different implementations of Bitcoin, has very big, dangerous — because it’s really hard for us to get two versions of software, two different software implementations to agree on doing the same thing. But the fact it’s possible in the sort of extreme cases is so important for the decentralization argument and to actually go from this isn’t necessarily a good idea, but it is possible to no, actually, technically it’s not possible to do anymore. That’s a really serious problem.
Stephan Livera:
Yeah. Okay. And so now I think there was some back and forth in terms of people were going at Moxie saying, Hey, why don’t you use Bitcoin and Lightning? And now at that time, and this is maybe a week ago, a couple of days ago, Moxie’s response to something along the lines of, Oh, it’d be nice to do. If it could be done in a way that was non-custodial and scalable for the mobile users, what would you say to that sort of argument?
Peter Todd:
Certainly I mean, let’s face it. Lightning does have scalability problems at a certain point because every channel in Lightning ends up being a UTXO on Bitcoin. But we’re very far from reaching that point and given the sort of trade we’re willing to make, it’d be very easy to use things like SGX to get excellent performance on Lightning wallets while still letting the users maintain full custody of their coins. And also you can kind of say full custody of the meaning of those coins. You know, what is like, what is the consensus for instance? And I suspect knowing Moxie, I mean, let’s [INAUDIBLE], this and say, all right, let’s assume he wasn’t motivated by money or all that. I would think this would come out of his general philosophy of let’s go and create fairly centralized solutions to solve problems like this.
Peter Todd:
And I think a good example where that toss you up in mobile coin is one of the many things that they’ve centralized around is the actual blockchain data is apparently being hosted on Amazon as three. So, at yet another third-party failure mode by having Amazon S3 be the way that wallets go and, get data from that they actually need. It’s just like what? But if you’re Moxie and the rest of the system runs an S3, you know Amazon ECQ anyway, I mean, you kind of see the logic.
Stephan Livera:
Yeah. So I guess then the challenge then would just be, I guess I’m sort of bringing it back to a Bitcoin sense and thinking, okay, the good Bitcoin user can run their own Bitcoin node and that way they’ve got their own version or their own, I guess, copy of the blockchain data. And here we’re talking about how would the user know if something has gone wrong? They’re kind of stuck, they’re trusting Amazon now.
Peter Todd:
So in this case, I mean, assuming they’ve done their job right. They’re not actually trusting Amazon for correctness, but if Amazon blocks access to it, yeah. Their wallets will stop working. And you know, this is something that could be fixed in comparison to all other problems. It’s maybe not such a big deal, but it’s just sort of telling about, of the philosophy that’s behind this. It’s, we don’t actually care about de-centralization at all. We just want something that’ll go work and we’re going to take a very easy path to it.
Stephan Livera:
And so hypothetically, if they had gone with Lightning and let’s say they had said to their users, we’re going to integrate some kind of Lightning wallet into the app. And then there would still be some kind of channel open and channel close, and those kinds of transactions would still have to hit the chain in a, in the Bitcoin world, in Bitcoin’s blockchain. How how do you have any ideas around how that could have been done? Just I guess, hypothetically,
Peter Todd:
Well, in fact, we have a good example of this which is eclaire’s Phoenix wallet and Phoenix wallet does all of the channel management for you. And if you send money, for example, suppose I have a Phoenix wallet and you send money to me, it’ll manage the opening of a channel from one of their Lightning nodes to my wallet on my phone, in the background. And it just automatically happens. And when I want, when I go to send that money again, well, I’ve already got a channel there and I just send it back the other way and everything magically happens. But importantly, because my Phoenix wallet on my phone actually is running a Lightning channel. There is no way for them to just take money out of that. It’s not custodial at all.
Stephan Livera:
Yeah. That’s a great example. And I think there are wallets out there like Phoenix or Breeze or Muun wallet and I’ve interviewed the founders of these wallets also, or of these applications. And they’re kind of in some sense, they’re like a clever hybrid model that gives you, mostly non-custodial experience while still hiding the technical details and that kind of gory stuff in the background,
Peter Todd:
Phoenix wallet is non-custodial, it’s not mostly non-custodial. It is non-custodial, you know? Yes. They just do it.
Stephan Livera:
Yeah. I guess the only part would just be the swap ins and swap out aspects where it is a trusted swap in and out
Peter Todd:
Really though, because until that swap in happens, I haven’t received a payment. You know, the one part of that, which is you could kind of say as trusted is the time between when I see the payment pop-up in, when it actually confirms, but if you’re willing to do this with SGX, just go in attest to the fact that the equivalent of Phoenix wallets implementation on their servers will never steal my money. You can also use SGX to test the fact that your privacy is protected. You know, I mean, a Signal itself already actually does this for contact data information, they use SGX to attest to the fact that their servers are not leaking, who is in contact with whom. So doing this with Lightning would be very easy, but if you did that, well, then you wouldn’t need a new coin. Would you?
Stephan Livera:
Unfortunately, for those poor, poor VCs who want to get their nice 20 X? So…
Peter Todd:
I guess I should point out if they went with this path, there is still a way to get rid of any of this, which is you just go and charge them fees on transfers, just like any other Lightning implementation, this really isn’t a big deal, but that business model is a lot more boring than we made a billion dollars because our coin sold out.
Stephan Livera:
Yeah, exactly. And so they could have gone and basically done something similar to the, Phoenix model and created their own, well-known Lightning node. It could have been the Signal Lightning node, and then the Signal users, in some sense, become
Peter Todd:
Signal Lightning nodes using this model, it’s, you don’t need just one, you can paralyze this as much as you want.
Stephan Livera:
Right. And so then users could pay to other Lightning network and here’s the other thing it would have far more interoperable. So all the world’s Bitcoin users, and Lightning Bitcoin Lightning users could pay a Signal user.
Peter Todd:
And it would still be just as instant as any other Lightning transaction.
Stephan Livera:
Yeah. Which is, which is a shame in some ways, because I mean, even right now, I’m sure there are, Bitcoiners and Lightning users out there who have used Signal just by literally copying and pasting a Lightning invoice into Signal. And just using that to pay. Like, I’ve done that to here and there.
Peter Todd:
Certainly I’ve totally done that many times. It works just fine.
Stephan Livera:
Yeah. So it feels like a big disappointment to say Signal, basically sell out and start shitcoining when they could have just done this simple Lightning option. I mean, maybe it’s not simple, but it would have been, it would have been work. Of course let’s be clear, but I think, I mean, my perfect.
Peter Todd:
In comparison to creating their own coin. It would have been simpler.
Stephan Livera:
Yeah.
Peter Todd:
When you go look at how much work they had to do to create mobile coin within the SGX environments, which were, you know, needs to have all your software rewritten, this is not trivial. This is a lot of work.
Stephan Livera:
Yeah. And you were pointing out that as well on Twitter that they had been, I think implementing this since October of 2020 right?
Peter Todd:
I don’t think that was me pointing it out and it would have been even earlier than that.
Stephan Livera:
Oh, I think you were saying, — yeah, it wasn’t that you found it. I think you were just kind of raising awareness to that. And I think you were saying you’re giving credit to someone else, but I think you were saying they had added mobile coin payments integration back in October 2020.
Peter Todd:
Oh, okay. Yeah. So sorry, I should say so Signal added support to their server backend in 2020, but when I’m talking about the amount of work being done, I’m talking about more of the effort the mobile point needed to do so. All the software. Yeah. So Signal itself one way or the other wouldn’t really need to do that much work. the wallet itself. Obviously we would be fair bit of work, but that’s not work. You saved by creating your own coin. That’s yeah. That’s work that has be done no matter what.
Stephan Livera:
And I think it’s probably a point where maybe someone who doesn’t like Bitcoin or Lightning, they could come back and say, well, what about all the privacy trade-offs of using Lightning? You know, Lightning is not perfectly private and maybe there’s a lot of, they could say, look, a lot of the large Lightning nodes can probably de-anonymize, you know, a decent amount of the Lightning network traffic it’s possible to run…
Peter Todd:
Which gets back to, if you have SGX, SGX can do all of this much better. And in fact, there’s a particular way where I should say what you for failure mode, where I think SGX plus Lightning has much better privacy than what they’ve influenced with mobile coin, which is if SGX is broken, the problem with mobile coins use of SGX is you still have all those statistical analysis being possible in that you can still look at the blockchain and say, all right, now that we’ve served on masses, who might’ve paid, who, but with Lightning, that data just no longer exists, a Lightning transactions, a femoral, the moments it’s happened, the actual data associated with the fact that happened can be thrown away well, or I should say may not exactly moments, but you know, you get the points within a few minutes. It’s it’s, it would be completely okay to throw away that data because the transactions happened and it’s now reflected just a balance maintained between two different computers, how, how they arrived at that balance is irrelevant.
Stephan Livera:
So I suppose there is a maybe, maybe this would be another steel man for the antique. Now I’m obviously I’m a bitcoin and and Lightning promoter myself. But I guess the anti Bitcoin and Lightning argument here might be something like, look, there are still channel probing attacks possible on Lightning. And by trying to infer the channel, balance in certain Lightning wallets and potentially looking at what’s going on on chain in terms of the on chain footprint of a Lightning wallet, by looking at the channel opens and channel closes that it’s possible, or they’re using a more advanced surveillance that it is possible to do that kind of surveillance on Bitcoin and Lightning. And I guess obviously I don’t fully agree with that,
Peter Todd:
But yeah, in this scenario, what we’re trying to do is maintain privacy between end users, doing transactions to a hypothetical Lightning and SGX implementation. You can make that implementation, just not be vulnerable to any of these attacks. You know, if you, if you’re, if you’re running big Lightning nodes within SGX, and they’re connected to a whole bunch of different users trying to do any probing attack, we’ll just get you no information at all, because you can’t access the state of the channels between those users and that SGX node, nor can you access the States of the channels between the different nodes in that group. You know, cause after on this scenario, you’re not going to run one node for every single Signal user. You’re probably going to want to skill it out a bit and have some redundancy, but the communications between those nodes, there’s no reason to give out any information about the status of those channels. You don’t have to. So all these attacks just disappear and we’re back to the requirements of saying, well, we’re trusting SGX to say that there isn’t code running, that’s recording the transactions, which I think is a pretty good, perfectly fine argument because the transaction data could be discarded. And that’s what makes SGX, I think a reasonable trade off to make. Someone should just go do this. I mean, screw mobile coin. Like this just makes sense. It would actually be quite useful.
Stephan Livera:
It’s a good point. That’s a good way to frame it. And so essentially these Signal users in the hypothetical world, let’s say they took on Lightning and they did it like a Phoenix style. They could pay each other, other Signal users and a way that was much more private, obviously with certain trade-offs of trusting SGX. And then they could have also had that benefit of being interoperable with the rest of the Bitcoin and Lightning world, by being able to pay Bitcoin and Lightning invoices, using Signal hypothetically that there’s the possibility they could have had.
Peter Todd:
And on top of all this, their payments would actually be a little faster because just has lower latency than any consensus system ever could.
Stephan Livera:
Yeah, because Lightning payments can be done in a few seconds. Right.
Peter Todd:
Well, much less than that. Remember a Lightning payment is done by the time the information is transferred between the nodes in the road. When you’re talking to the system where you’ve got a bunch of users using this with, within themselves, that latency is down to liberal packet latency. You know, I could make like this way, if we were connected with Lightning wallets through a third-party and that third-party had ample CP and so on, we can make Lightning transaction effectively as fast as a packet takes to go back, between me and you and that’s milliseconds, you just cannot have a consensus system that maintains global consensus and milliseconds like that. You know, this, the speed of light’s too slow. So Lightning actually has the advantage there.
Stephan Livera:
Yeah. It’s pretty damning in some ways I think it’s, and it’s not to say that, Bitcoin and Lightning is perfect and there’s no trade-offs, I just think it would have been, would it not have been better for them to just do the work and grapple honestly, with the trade-offs a bitcoin of Lightning than to go start a shitcoin.
Peter Todd:
And if they don’t like Bitcoin, whatever use the same technology for like at USD token or Ethereum or whatever, the technology of Lightning isn’t dependent on Bitcoin. In fact Lightning wallets on light coin. If I remember correctly do exist, I don’t know anyone ever uses them, but you’re not actually tied to Bitcoin. The currency, this technology is better than anything that they’ve implemented. And, in a point like look at it as, say a retail example. If you want to use your Signal wallet to send someone money in sort of a retail situation, the fact of the matter is the speed of light slow enough that I would much rather that data package go, locally within my province. And you could do this with Lightning cause you didn’t go set up Lightning nodes that are actually physically located close to me like that that is actually a noticeable improvement in experience versus having to wait till the whole globe comes to consensus. You know, if we kind of don’t notice it that much because the world isn’t that big compared to the speed of light, but you put some retail, one the moon, and you’re going to pretty quickly complain about this.
Stephan Livera:
Yeah. Interesting. Yeah. It’s a good point. You make about not having to get global consensus on every little payment and in the Lightning world, we can I can pay you or you can pay a Lightning use very quickly. Now I guess the [inaudible] would be something like, okay it still requires channel state updates and that might, three or four back and forth in terms of channel, state updates. But I think even if you net all of that out, it still works out far, far faster. And just a lot more, it kind of long-term sustainable. Now. It might also be a point there that over time. So let’s say Signal has 50 million monthly active users. Let’s say if they were to get, go another 20 X and get to whatever Facebook or one of these other messenger sizes, then all of those channel opens and closes having to go on Bitcoin’s blockchain might actually become a little bit more difficult at that point. Wouldn’t you say?
Peter Todd:
Well, the funny thing is my answer to that would actually be screw Bitcoin, just do it on a theory. And I’m sort of being facetious because actually Ethereum has higher fees than Bitcoin does right now. But if I was, Look at it this way, if they’d hired me as consultant, I probably would have given to them an answer like that. Like, let’s go, just pick some coin. This isn’t actually getting that much use that we can sort of do this sort of thing and get away with it. Or even for that matter, like if we were to do it on a brand new coin, doing that with Lightning technology would actually make more sense. Because you reduce the size of the shared state. And just from a maintenance point of view, shared consensus state is really tricky because it makes running the servers hard.
Peter Todd:
You forget this, but suppose you have a server that goes down and you need to bring it back up again in a Bitcoin environment where the speed of computers is fast, roll to the blockchain. You can do almost any operation with a sufficiently fast computer in a matter of minutes to maybe few hours, you know, you need to rethink the whole chain because, Oh shoot, you just realize there’s a bug and you don’t quite know where the Val invalid transactions are. You can do that because blockchain data just isn’t that big compared to the fast computers you can get. But you try to pull that same thing off with a visa scale coin with a shared consensus in suddenly you’re wondering, shoot, how long are we going to be down for? You know, can we even afford to validate the stuff? Can we afford to check for this condition? We found it, it just makes all the administration that much more tricky and that much more high stakes.
Stephan Livera:
And I guess also some listeners might be thinking, okay, so we’ve spoken a bit about the idea of the Lightning model. What about the side chain model? Would that have been an appropriate alternative or how should we think about that?
Peter Todd:
Well, the reason why you do a side chain or I should put it to, I mean, when you, when you talk about a side chain, you could ask, well, what type of side chain are we talking about? Are you calling say liquid side chain? Are you calling, something more heavily tied to Bitcoin aside chain? I mean, if you’re talking about the sort of liquid model, then I think that’s getting very close to, I just mentioned as it’s, if you’re going to have a separate coin or a new D new currency or whatever, you’d much rather have something where it’s say a chain with somewhat more, maybe, somewhat say less restrictions on how much data can be, but using Lightning to then actually make it really scale properly. But if they really wanted to say well, put it this way, supposedly we’re dead set on creating their own coin for the consensus side of it.
Peter Todd:
Tying that back to Bitcoin to come to consensus would probably make sense. Unfortunately, that’s just, unfortunately that’s just not going to happen because I think they’re there. So the group of people who would want to use environmental costs as an advertising point. Yeah. I like that. That’s part of their argument for mobile coin. We’ll, Oh look, you know, we have no environmental cost. And so many of these arguments are kind of because they don’t talk about Bitcoin. They pretend to Bitcoin is a system where every transactions on chain and that’s just not true, or they pretend that for instance, energy, you scales with transaction falling, which is also not true. But even though technically doing that kind of model might make more sense. So against a lot of failure modes, there’s just not going to do it. They’d much rather have this, the best possible centralized system you can get where you can sort of kind of claim it’s decentralized, even though it really isn’t. And it’s, it just invites so many ugly questions from regular regulations. I would not want to be I would not want to be involved in that project right now.
Stephan Livera:
Yeah. That’s a real shame. And I mean, I wonder, could they have done some kind of advanced DLC type approach where they do some kind of fiat denominated payments but in the background it’s actually, Bitcoin DLC is going, is there any way that could have been a potential?
Peter Todd:
Well, interestingly this is one of those things where again, SGX could potentially play a role because you could use it to a test to the fact that for instance, prices were accurate or, coins would be held in certain States or so on. And there’s a lot of things you could do with that. But I think going back to this notion that how the — what coin the tech is on top of doesn’t really matter that much. If you just partnered with some institution to maintain, USD in the background, that would be great too.
Stephan Livera:
I guess the challenge may be something like, Oh, well, there might be concerns later down the line that users would have to go KYC. And the whole point of Signal is that you can be, an activist under an oppressive government tyrannical government, and you need to be able to, send money or receive money under these adversarial conditions. And if we use this, then maybe that’s exposing too much KYC requirement or putting too much KYC recliner.
Peter Todd:
Well, I’ll bet you, that was part of the idea of creating a new coin with, but you notice how it’s not, that like this demo software is not available to people outside of the UK. They’ve already run into these sort of problems.
Stephan Livera:
Yeah. That’s a good point as well, because they’re starting, I think the trial is starting with UK users or at least currently, and maybe they’ll expand that out, but then maybe they will run into these exact same hurdles of, Oh, you need to do KYC and you need to do know your customer and ask them for blah, blah, blah. And then all of a sudden now the privacy that was meant to be given to these people is actually not, it’s not a reality.
Peter Todd:
For them, it worse. You’ve now invited all of us attention to Signal itself
Stephan Livera:
kicking the hornet’s nest before you’re ready.
Peter Todd:
Yeah. I mean, frankly, if I was talking from the perspective of a Signal of just Signal users, I would say the idea of adding a coin to this just doesn’t make sense under almost any circumstance. You know, like if people need to move money around on Signal, just copy and paste the Lightning transaction, maybe add some very minor supports to like call out to another wallet or something, but, putting this much integration and inviting all this KYC, all attention is just not a good idea.
Stephan Livera:
Yeah. And so let’s talk a little bit about funding in general for projects, open source projects. Now Signal has a foundation. I think they were given a donation of some funding for that.
Peter Todd:
No. Oh, they weren’t actually, they were given a loan.
Stephan Livera:
Oh, I see. Right. I actually thought that was… Okay. So do you know, are you able to elaborate more on that and what’s the funding model there and what you would — how should they approach that?
Peter Todd:
Well, unfortunately I can’t, because nobody knows the terms of the loan as far as I know now, this could mean that it’s structured as a loan for, tax reasons or something and it’s completely harmless, or it could mean that whoever gave them the loan actually expects that to get paid back somehow with revenue. So, that’s just, it’s a big open question. It’d be much better if they did not have that. And they just had the straight donation. And I think for something as important as Signal getting that, getting those donation funds is probably not that hard sequel is not that big of a project. It doesn’t have that big of cost to keep it running. You know, if you compare it to say a Wikipedia, I think Wikipedia needs a much bigger budget than Signal does. And Wikipedia seems to be doing just fine on that. So it’s, it really makes me wonder is this really needed to get funding? And I think the answer is probably actually no.
Stephan Livera:
Yeah, that’s a shame. And there’s obviously in the open-source world, it’s difficult to like have one true answer of how it should be done, but there’s different ideas out there. Some of them are just donation only some of them are maybe they do a freemium model where they have the product free, but then the, some users can just pay for a fit for some additional features. And then those paying users are in effect subsidizing the free ones. I guess that’s another potential option that they could have gone.
Peter Todd:
Yep. And the facts matter is if you have something that works as well as Signal does and does as much good as Signal does, it’s not that hard to get donation money. And there’s a lot of projects that run purely on donations. Yeah. That’s a shame because yeah. You know, and that’s including for paying developers too. Yeah.
Stephan Livera:
Yeah. And so what are the main costs for these kinds of foundations and organizations? Is it mainly around developer costs and then say server and hosting costs and potentially some legal foundation fees and things like that?
Peter Todd:
It honestly depends a lot on what project you’re talking about for Signal — given Signal schools of scaling up. They may be in a position where hosting costs are actually more of an obstacle than than development costs are. But frankly, it’s hard to know cause it really depends on you. How exactly are people using Signal? If I send you a text message on Signal, the cost of bandwidth for that is will, and truly trivial. Now, if I go send you a minute long video, I’m doing the equivalent in terms of resources of, hundreds of thousands of text messages. So how people use Signal, that kind of open question and frankly, I don’t know, but this is again, one of those things where getting sponsorship for that hosting doesn’t sound like it should be that much of a challenge. You know, I’m not going to say it’s trivial, but given where Signal is, if you say, gone to Microsoft or, Google or Amazon or something and partnered with them for hosting costs, that’s probably doable. You know, CloudFlare seems to be the sorta sorta company that might be willing to go and do this kind of stuff or, more niche things. It’s it just, it just doesn’t feel like this is an insurmountable obstacle.
Stephan Livera:
Yeah. Okay. So turning more to what should Bitcoin is do because
Peter Todd:
Maybe, I’ll just make one final point too, is that the fact that they now have a coin and they have the sort of taint attached to it makes future sponsorship a lot harder to achieve.
Stephan Livera:
Yeah, that’s right. Because they’ll say, well, Hey, you started your own shitcoin why do I need to pay you now for it? Exactly. Yeah. So from a Bitcoin perspective, we have probably seen this happen and look, I mean, it’s not just Bitcoin is right, but anyone who’s sort of interested to use privacy technologies, privacy messaging, they might’ve been using Keybase, but then Keybase, had some ripple shitcoin integration sorry, Stellar. Sorry. Yeah. And then, there’s other messaging apps out there I’d like people to talk about, what is it, Threema and then you can self host what used to be, was it riot matrix now, elements, matrix, what are some of the options that, Bitcoin people or anyone interested in privacy? What should they be doing? Or, I mean, should they just continue using Signal for now, but then be on the lookout for something better?
Peter Todd:
Well, I think in general people have to assume there’s a good chance Signal gets shut down over this. You know, I’m not going to say it say a 50% chance, but it’s certainly more than zero. And it’s probably in the single digits. You know, what, what Signal is done is really risky and they’ve opened a hornet’s nest that didn’t need to happen. And also this may be something where Signal becomes available on a per country basis. India is an example, they tried to ban cryptocurrencies. Will you be able to dance, download Signal in the future in India? Maybe not. You know, I mean, even if, even if they say, well, the version in India doesn’t have it enabled or so on, it doesn’t necessarily mean the government’s going, gonna accept that as an explanation. They may still say, well, app stores may no longer offer Signal and unfortunately Google would probably comply.
Peter Todd:
So it’s an ugly thing. And this for alternatives, it depends on what you’re trying to do if you’re just talking amongst close, with small group of technical people, self hosting stuff is options and so on. But if you’re talking to your a less than technical family members I wish I could give you good options right now, Signal Signal is your kind of your best option even still, but you just sort of be the best advice I give is sort of be on the lookout and see what comes in the future.
Stephan Livera:
Yeah. And actually one more point while you were — you just got me thinking as well in terms of distribution, right? So many applications nowadays rely on Google and Apple for distribution to the user. But in the more, let’s say adversarial environment, that user may not be able to rely on that. And then that is why they use alternative app stores and F droid or side load apps, or potentially have the dot APK file, which they are downloading and installing. That’s probably another area.,
Peter Todd:
First and foremost, I need to point out all of these solutions are only doable in Android. Apple just doesn’t allow this stuff in any reasonable way. So right there, yeah. Get an Android phone rather than Apple.
Stephan Livera:
Well, there you go. And even then there’s, more of a push, some Bitcoiners are into using some of the alternative more hardened operating systems, right? So graphene OS some people use copperhead and some people are into CalyxOS. I guess the idea then is, Signal the app should be usable in those, in, in that kind of scenario as well. You shouldn’t have to be trusting and using the Google play store to get access to these things.
Peter Todd:
And that’s one of the things where Moxie is famously for centralization, and he is pushed back very hard on people, even just having alternative Signal clients, let alone federating the protocol. So he kind of has a history of this stuff, and it’s made a lot of people dubious believing even before this mobile coin stuff.
Stephan Livera:
Yeah, that’s unfortunate. So I guess any, I guess, closing thoughts to leave listeners with, what should they be thinking about in terms of their private messaging and potentially for private messaging with their less technical family and friends?
Peter Todd:
Well, you know unfortunately maybe the thing that say is this could be like email, we don’t need to have a system where there is sort of a one size fits all for how you use your messaging in much the same way as you can go use Gmail for your email. And I can go and run my own server, which is exactly what I do. You know, th that’s advantage of federated protocols and Moxie’s arguments against that as well. Then you can’t advance the protocol as fast, et cetera, et cetera. But at this point, I think messaging is pretty well understood. It would be pretty reasonable to create a federated protocol doing the things that Signal does in doing it in a proper federated way, but remains to be seen if anyone will put in effort to do that. And even if they do we’ll catch on,
Stephan Livera:
I think that’s an interesting way to put it. So essentially it’s like even when people were angry at Twitter for all the censorship and so on, and then there was a push about using Mastodon, it’s probably an equivalent to that. It’s like this idea of, let’s try to look at these federated options that give people a way to interact and interoperate using different clients and different software such that you don’t have to tell people, Hey, get this one app. And maybe in the messaging world, maybe that ends up being something like the elements matrix style setup.
Peter Todd:
Exactly. Will it happen? I guess we’ll have to go see and unfortunately, a very important ingredient to it, which I think you mentioned before is we need phones continue to be able to, you know, have general purpose software installed on them. And right now our sort of last holdout of commonly used phones is Android. Apple. Doesn’t really allow this in a reasonable way. And I don’t, you know, unfortunately I don’t have good answers to this. If this goes away, it is just not feasible to create phones these days and expect a reasonable percentage of population, half them, unless this is possible from the get-go and crazy enough, we probably need legislation solutions to this to assure that you can actually run custom software and phones. That’s something that could go away.
Stephan Livera:
So what about if hypothetically people are listening and they want to fund projects that are going in this direction, does that mean they should be looking at things like, you know, Liebrum or pine phone or, you know, those kinds of projects that are trying to be like a Linux phone or an open phone?
Peter Todd:
I mean, I think those things are valuable, but they are not by themselves a solution to this problem, as much as I want people support those things. I think we also need to put pressure on Apple and Google to, in case of Apple opened up their phones again, in the case of Google continue opening it. And that also part of that putting pressure probably means talking to your local politicians and making sure they’ve realized this is an important thing. And it kind of fits into Rachel repair, a movement as well.
Stephan Livera:
Oh, okay. Yeah. yeah, I’m not as familiar with the right to repair. What is that?
Peter Todd:
Well, it’s effectively a well, so a good example is in most countries you have a right to repair your car and not only are automator makers not allowed to, for instance, cancel your warranty for repairing your car. They’re also required to provide technical information about cars and also require the car to supply spare parts, et cetera, et cetera. And that effectively means that you’re able to modify and repair your car. And in the tech world, that’s going away. It’s much harder for people to, for instance replace batteries on iPhones. Then the equivalent thing for a car, you know, nothing stops replacing like the gas tank on your car, but with iPhones, they’re rapidly adding security features, which makes us impossible. And the idea of rates repair laws, just to force phone manufacturers, continue making this, making these things possible, as well as depending on, what proposal we talked about for repo supply the same manuals for repair as are available anywhere else in the supply schematics and so on.
Peter Todd:
So there’s a bunch of different proposals and w the sort of general umbrella term Rachel pair. But I think the name kind of sums it up, sums it up like the right to actually modify your phone and fix your phone and get access to how it’s working. And rather than just being this glass brick, that you have no visibility into it from a security point of view, this is very important. You know, even experts have very hard time analyzing phones to know what they’re actually doing. And phone of course, is a microphone and a camera. You know, you could be walking around with a bug in your pocket, recording everything you go say, and it’s very hard to prove that it’s not happening. And the level of lockdown in modern phones is quite extreme.
Stephan Livera:
Yeah, I see. Yeah, so I guess I wouldn’t want to see the answer being, yeah, let’s call that for favorable government regulation. I think, I think the answer would be more like we have to actually just put money into building the alternative. And I think maybe that alternative is not clear and easy to see right now, but it given enough time and enough people getting concerned about using these things using technology in a more privacy preserving way, then maybe there’ll be enough. And also as obviously number goes up over time we might see more funding into suitable projects, I guess that’s how I would.
Peter Todd:
Well, I mean, the thing I pointed out there is the nature, like the level of resources that go into building a modern phone from scratch is so that the entire world’s economy can barely support, more than a handful of manufacturers, right? It’s, it’s just something where we run into the problem that earth itself is too centralized. And it’s, it just keeps getting worse year after year, because as we get more and more advanced chips, the amount of infrastructure required to build those chips just keeps going up. You know, I mean, Intel, for example, could actually drop a manufacturing business and that would effectively leave us with one company in the entire world. That’s able to make chips from scratch, TSMC, and there is no competing with that. Like, unfortunately, these are things where I think libertarians get a bit naive about the scale of these problems. There aren’t good solutions to this, if just going your own. And it’s very easy for governments to cut off access to those solutions. Anyway, pure isms phones, they depend on third parties to supply them. And the ecosystem of those third parties is remarkably centralized and there is no way to fix this problem with the level of resources that effectually turn you into a government anymore.
Stephan Livera:
Yeah. So I think that’s probably the open question. I think of it more like it’s not necessarily it’s not to say that every market in order to be a healthy market must have, 30 high, all equal competitors. I think some in some markets, it will just naturally tend towards a few smaller amount of larger players. And I think from my point of view, it’s not saying, Oh yeah, look, if the government wasn’t there, everything would be perfect. It’s more just to say, we’re probably going to, by inviting government in, we’re probably going to create a lot more problems than we’re solving.
Peter Todd:
Look at this way. When we get to the point where these things are, the centralized, we can’t avoid inviting government in because these companies and at working with government anyway, there, there is no avoiding this one way or the other chip fabs get in bed with government. You know, I mean, TSMC, there are political football between Taiwan, China, and the U S and really the rest of the free world. Like it is very easy for TSMC to get into. I mean, concretely, there is a very genuine worry that the Chinese government invades Taiwan to seize TSMC. And once that happens, you’re screwed because it’s now the Chinese government that has, that’s controlling the chip industry, they’re, they’re not going to care about your libertarian ideals. They’re going to say, well, no, purism tough luck. We’re just not gonna supply you with chips.
Stephan Livera:
Yeah, no, that’s a fair point. I think I think we’re just going to have to,
Peter Todd:
Obviously governments are more decentralized than the chip industry right now, because there are a lot more governments out there, the chip manufacturers.
Stephan Livera:
Yeah, for sure. Yeah, I think, I think that’s correct. I can’t disagree with you there, so we’ll just have to wait and see, I think what we’ll see is over time, we’ll just like, see it, but it’s not going to be something solved tomorrow, but I think at least for now, we’ve sort of spoken a little bit about, what were some of the trade-offs of Signal Signal’s decision and their choice in starting a shitcoin or, using this shitcoin as opposed to, doing the Bitcoin Lightning thing. So, yeah, probably we’ll probably close it off there. Any final comments or and of course, where can people find you online?
Peter Todd:
Well PeterTodd.org, my not often updated website and @PeterKTodd on Twitter. And I do have Mastodon on Mastodon.petertodd.org as well. So, that’s where I am. And I always like to say, if you can’t figure it out, my email address, you’re not a very good stalker.
Stephan Livera:
Well, thanks very much, Peter. I enjoy chatting with you and certainly very educational.
Peter Todd:
Thank you. Talk later.