In this conversation, Stephan Livera and James O’Beirne discuss the implications of quantum computing on Bitcoin, exploring skepticism towards the perceived threats, the current state of quantum research, and the potential responses from Bitcoin developers. They delve into proposed solutions, the role of institutions like NIST, and the challenges of Bitcoin protocol development. The discussion also touches on user experience, self-custody, and the future of Bitcoin adoption amidst evolving technological landscapes.

Takeaways:

🔸Quantum computing poses theoretical risks to Bitcoin, but practical threats are far off.

🔸Skepticism about quantum computing’s immediate impact is prevalent among experts.

🔸Bitcoin developers have more pressing issues to address than quantum threats.

🔸Proposed solutions like BIP 360 could enhance Bitcoin’s security regardless of quantum risks.

🔸NIST’s push for post-quantum cryptography raises concerns about government influence.

🔸The academic and business interests in quantum computing may hype its urgency.

🔸Bitcoin’s protocol development faces challenges due to a lack of strong leadership.

🔸User experience is crucial for Bitcoin adoption and self-custody.

🔸The future of Bitcoin may depend on how well it adapts to user needs and technological changes.

🔸Bitcoin remains a unique solution to fiat debasement and economic instability.

Timestamps:

(00:00) – Intro

(00:46) – What is the deal with quantum computing and Bitcoin?

(05:05) – Advancing progress in quantum computing

(09:43) – Is the quantum threat to Bitcoin around the corner? 

(11:57) – Quantum resistant schemes developed by Bitcoiners

(15:45) – NIST and diverse views on Post-quantum cryptography

(22:31) – The future of Bitcoin in a Post-quantum world

(29:45) – ‘Matt Corallo Quantum Plan’

(34:08) – A modest version of the Big Blocker view

(38:41) – Covenants and Scaling Bitcoin

(43:35) – Comparing OP_TEMPLATEHASH & OP_CTV

(48:01) – The ‘Slow and Steady’ approach to Bitcoin development

(55:19) – What’s the next adoption pathway for Bitcoin?

(1:04:04) – AssumeUTXO, Utreexo & Floresta

(1:07:20) – Why are so few people running Bitcoin nodes?

(1:14:52) – Closing thoughts 

Links: 

• https://x.com/jamesob 
• https://ctv-csfs.com/ 
• https://www.youtube.com/watch?v=27Qs31E80cA 

Stephan Livera links:

• Follow me on X: @stephanlivera
• Subscribe to the podcast
• Subscribe to Substack

Transcript:

Stephan Livera (00:00)
Hi everyone and welcome back to Stephan Livera podcast. Today we’re going to be discussing whether quantum is BS and rejoining me on the show is James O.B. or James O’Beirne. He is a long time Bitcoin developer, has been working on focused on the custody side of things also ⁓ and was the creator of the OpVault proposal, ⁓ has lots of different views. Today we’ll mainly chat on the quantum stuff, but maybe we’ll get into some of the other, ⁓ you know, covenants in general kind of core and ⁓

protocol discussion but yeah first of all welcome back to the show and ⁓ you know give us your overview on ⁓ quantum stuff what what what is the deal with this quantum computing and Bitcoin

James O’Beirne (00:40)
Hey, thanks, Stephan. It’s good to be back. It’s been a little while since I’ve been on the show, but yeah, I always love coming on. It’s funny, quantum, while it’s been sort of a recent highlight for a lot of people, I was kind of reminiscing the other day about when I first heard about it. And I was in an undergraduate computer science program freshman year, and they had us go to a seminar about this thing called quantum computing.

and the year would have been 2007. So it’s been almost 19 years that I’ve been thinking about this stuff, albeit very passively. But for the first maybe 10 years I had heard about quantum computing, I was very agnostic about it. people who you would talk to who were sort of close to quantum computing were anywhere from alarmist,

to kind of ambivalent to agnostic about it. ⁓ And so I never really knew what to make of it for at least like the first 10 years of knowing about it. ⁓ I always assumed it was something that was practical, maybe on the horizon, ⁓ but still quite far off or maybe it was ⁓ incrementally making progress in the same way that AI sort of incrementally made progress for decades. ⁓ But.

I would say maybe six or seven years ago, through an accumulation of reading, as well as talking to people, anybody with a physics background or with a heavy duty cryptography background, I would kind of pull on quantum computing to see where it was at. kind of the response that I would get, know, roughly directly correlated with, you know, how smart I thought the person was, ⁓ you know, in their given field.

was basically that we haven’t seen any kind of appreciable progress. ⁓ It’s really far off. Maybe it’s theoretically possible, maybe not. ⁓ And in fact, the people who I polled ⁓ who were closer and closer to having a rigorous physics background were increasingly skeptical that it was even a thing at all. ⁓ And so, in the last two years,

of quantum kind of coming to the fore in terms of a quantum, know, a cryptographically relevant quantum computer, posing some kind of a challenge to Bitcoin. That kind of ramped up both my interest and consequent skepticism because quantum is being used as a sort of wedge, I think, to potentially drive the adoption of a bunch of new cryptography.

So anyway, I’ll pause there in case you have any interjections.

Stephan Livera (03:36)
Yeah, okay.

So I mean, obviously, you know, the threat that most people are concerned with is this idea that if a if there’s new advances very quickly in quantum computing that ⁓ and that combined with Shor’s algorithm, which may allow people to sort of back out the private key for our bitcoins and then steal the coins, etc. Especially if you have an exposed public key for that, for those coins or your ex pub was exposed at some point. That’s that seems to be the main threat.

And then of course that has spurred a lot of research and discussion The hunter beast pay to Merkle route Proposal BIP has now been merged as a BIP So there’s there has been some progress on that But on the other hand there are let’s say detractors of Bitcoin who saying our Bitcoin developers are not doing anything But at the same time there are real trade-offs to this as well. It’s not just free It’s not just you can’t just snap your fingers and are now we’re quantum secure like there are there’s real work to be done But let’s just dig into the possibility

aspect of it, right? Because it seems to be that there’s been some recent announcements from Google and some other big quantum labs, and they’re saying things like, ⁓ well, we’ve got this Willow chip, or, ⁓ now it’s not a matter of, ⁓ it’s a matter of engineering capability. Where are you on that?

James O’Beirne (04:57)
Yeah, so if you look at the timeline, debatably the largest number ever factored by a quantum computer is 15. And that was done back in 2001 by IBM. And 15, I guess, has some special structure that was particularly amenable to quantum computers. And so when the next attempt was made at factoring, which was 21 in 2012, that was a 10x effort.

over and there’s even some debate about the structure of that as to whether or not it was a legitimate use of Shor’s algorithm. ⁓ So we really haven’t seen any demonstrable progress past that. There have been these kind of sleight of hand type experiments ⁓ using quantum computing or some physical apparatus. But those are typically handicapped by ⁓ classical

pre or post-processing. ⁓ So it’s not a purely quantum thing. Oftentimes the way the quantum circuit is set up is with some knowledge of the solution. So ⁓ really these are toy sort of meaningless examples. ⁓ You know, while I think I’m not an expert on the experimental state of, ⁓ you know, things like error correction or, ⁓ you know, increases in coherence time and so forth. ⁓ I think

To my knowledge, there remains this truly fundamental gap in terms of the idea of, you know, one of the big fundamental questions is can these qubits, which are basically, you know, probabilistic bits, so in a computer, all computations happen on these deterministic bits in our classical computers, where there’s either a zero or a one state. In quantum computers,

those computers work on qubits, which have a sort of probabilistic state. And there’s an entanglement between all of the qubits in a computer, ⁓ which creates all kinds of problems because maintaining a coherent quantum state is this very finicky physical process that we really don’t have great methods for in terms of scaling everything up. ⁓ And so there’s this…

this philosophical or kind of theoretical ⁓ question that’s basically like, you know, for a quantum circuit of sufficient depth, you know, say something like running shores on even, you know, 128 bit ⁓ ECDSA, does the depth of that gate, does the time spent in the computation kind of outstrip

the physical limits of how long these qubits can maintain coherence. ⁓ And so that’s like an open fundamental question that even ⁓ prompts people to wonder if quantum computing is like a feasible thing in the first place. That’s setting aside all of the physical challenges that would accompany a quantum computer with a sufficient number of qubits to actually make something cryptographically relevant.

⁓ applicable. ⁓ You’re talking about the need for thousands of logical qubits, ⁓ which translates into tens of thousands, hundreds of thousands of physical qubits. ⁓ You need multiple physical qubits basically to construct a logical qubit. ⁓ And a logical qubit is kind of the equivalent of a single bit in a classical computer. ⁓

So there are varying reports on ⁓ all the different physical problems associated with this. There are different methods of constructing qubits. ⁓ And I couldn’t tell you what sort of the largest physical qubit arrangement that’s been come up with is because again, there are these different techniques, all of which are kind of fraught in their own various ways. Some things require…

keeping these highly tuned machines at totally frigid temperatures. ⁓ So there’s different physical means of achieving these qubits and they’re all kind of fraught and there’s no indication that we even have the physical means of scaling these things up.

Stephan Livera (09:30)
Yeah, so you’re definitely more on the skeptic side. do you see it as like it might never be a thing or do you see it more like no, it’s like decades out. Like where are you at there?

James O’Beirne (09:41)
I think ⁓ we don’t know if it’ll ever be a thing. And if it is a thing, ⁓ yeah, my belief is it’s many, many, decades out. The point of comparison that’s often drawn with quantum is one that stems from AI. And that’s this idea that, well, you know, for decades, you know, starting in the sixties ⁓ with the Lisp guys and, you ⁓ people have been talking about this notion of ⁓ AI and general intelligence and

really we didn’t have any kind of impressive result until a few years ago when we got these LLMs that, if not being a kind of generalized intelligence, at least sufficiently mimic that to the point where they can sort of fool most people. ⁓ And so the argument goes, well, with AI, it kind of came out of nowhere, so we should kind of expect something similar with quantum.

And well, I think that’s kind of a nuanced discussion in itself. ⁓ There was a ton of incremental progress with AI and, ⁓ you know, AI is this kind of purely virtual software problem. ⁓ Quantum brings in this very thorny physical component, even if you set aside kind of the fundamental questions about whether it’s feasible or not. And I think

We’ve gotten really good as a civilization at dealing with these virtual software problems. It’s a highly constrained environment. We haven’t gotten so good, you know, in the last few decades at dealing with physical problems. I mean, you know, for something as relatively simple as a say nuclear reactor, we can barely put those together anymore, let alone the kind of, you know, metallurgical and engineering advancements that we would need for something like a crypt

cryptographically relevant quantum computer. I think there’s, yeah, there are lots of avenues ⁓ for profound skepticism when it comes to this stuff.

Stephan Livera (11:47)
So when it comes to Bitcoin then, what do you think Bitcoin developers and people building in this ecosystem should do? Should they be working on quantum mitigation, ⁓ these kinds of ideas or not? Or where are you at on that?

James O’Beirne (12:05)
Yeah, it’s kind of a tough question because it depends on what level you want to answer it at. ⁓ From a fundamental basis, setting aside market expectations or worries ⁓ on the part of non-technical people, I would say there are way better uses of our time as developers. There’s kind of an infinite list of things that we could be working on. ⁓

And for me, know, quantum doesn’t even breach the top hundred things when it comes to Bitcoin. ⁓ But ⁓ but we’ve created the situation where ⁓ because there is this kind of doomish hype around ⁓ quantum and its possibility, ⁓ it’s now created, you know, potentially ⁓

significant ⁓ price implications for Bitcoin. obviously, to some extent, Bitcoin is this iterative, self-fulfilling prophecy. if we get to this point where institutions and other sort of large entities are convinced that quantum is just three years around the corner and the Bitcoin guys aren’t doing anything about it, ⁓

You know, that creates a massive headwind for Bitcoin adoption. And ⁓ I’ve been really interested and kind of morbidly curious about the interaction between, say, a delay in Bitcoin’s timeline. ⁓ You know, if sort of the Lindy meter on Bitcoin gets pushed back, whatever, eight years because people are worried about quantum, how does that interface with, ⁓ you know, the security budget and the supply schedule? ⁓ Because that’s a very complex dynamic in itself.

Stephan Livera (14:04)
Yeah, I see.

James O’Beirne (14:04)
So anyway, I’m sorry

to get back to what you were asking. ⁓ Ironically, I think there are some really good dimensions to some of the proposed solutions for these quantum mitigations. know, for example, like in the abstract, I’m very favorable to the content of bit 360 as it’s written. If not only because

from the beginning in Tappert, would have made a lot of sense to just have a spend to Merkle tree type without having that pub key on top.

Stephan Livera (14:38)
Yeah. So I guess in a way you’re

sort of saying some of these BIP 360 concepts might be useful anyway, regardless of whether quantum computing is a real thing.

James O’Beirne (14:47)
Mm-hmm.

Yeah, absolutely. I was much more ⁓ skeptical, hesitant ⁓ when the proposal actually incorporated specific ⁓ quantum resistant schemes. ⁓ Because, you know, when you start talking about putting new cryptography into Bitcoin, that’s kind of a whole other level of severity of change. ⁓ If you’re just talking about introducing a new taproot type that

you know, in many ways just removes code from the existing Taperoot path, ⁓ then, you know, that’s a ⁓ significantly less sensitive change to be making.

Stephan Livera (15:30)
Interesting. Yeah, and I guess the other, probably the other big one that people might want to ask you your view on is, look, James, what about NIST, National Institute of Standards and Technology? They are saying, look, you know, they’re running these, you know, ⁓ competitions or these ways to have people work on post quantum cryptography. They’re recommending by a certain year that people should have post quantum. And we’re seeing some.

instances of the technology industry update to using these kinds of things like VPNs and things like this. I think that might be because of the so-called store and decrypts later kind of concern. So I guess what’s your thought on that? Do you think that they are just overly panicked about something that’s not gonna be real anytime soon or how would you?

How are you thinking about this concept that NIST and others are already doing some form of post-quantum?

James O’Beirne (16:30)
I think when you’re dealing with a government agency, especially NIST, who has a history of trying to, you know, in some cases, you know, maybe they were led into it by other agencies or, you know, maybe they themselves proposed knowingly some faulty constructions like, ⁓ you know, the deterministic random bit generator. ⁓

or even just kind of looking back at other instance like the Clipper chip. mean, really if the government is strongly pushing cryptography your way, ⁓ you should be very, very wary. ⁓ And ⁓ there’s a part of me, I don’t really have substantial evidence for this obviously, but there’s a part of me that doesn’t wonder if, well, ECDSA is working just fine. ⁓ It’s a good algorithm in the classical world. ⁓

⁓ If I were an organization like NIST or the NSA and you know, I wanted to come up with a scheme that was, you know, thought to be very strong ⁓ when it concerned most actors, but you know, maybe I knew about the way that a certain parameter was specified or I was aware of some kind of structure that would allow me to, you know, given whatever information they have, ⁓ you know, safely unlock this scheme if there was a national security risk or

a warrant or something. ⁓ I would be really concerned about that. And I am really concerned about that, ⁓ especially when you kind of ⁓ pick up on the hysterics coming out of that community on this topic. But even setting aside the tin foil stuff, I think there’s just kind of a weird nexus going on between academia,

and business when it comes to quantum because there are a lot of people who have kind of based their academic careers studying quantum computing. There’s a whole sort of credentialing infrastructure there. ⁓ And from my time in academia, I think there are a lot of really smart people and a lot of good work.

going on, but you can also get into these situations where you know, you haven’t really thought hard about the fundamentals of what you’re doing and you’re just kind of proceeding on something because your advisor, you know, is there and is a subject matter expert. And so I can see there being this kind of, ⁓ you know, emergent phenomenon around there being a big scene in academia for quantum computing. There obviously being business interests around, you know.

these companies ⁓ and the two kind of fusing together ⁓ and creating a situation where those three entities, government, business, and academia, are sort of unintentionally hyping this thing up, if that makes sense. ⁓

Stephan Livera (19:23)
Yeah,

okay. And so it seems like, you know, from what we’re seeing and hearing, it’s Google, it’s IBM, it’s Continuum and a few other…

quantum labs and things like this. And of course, known in the Bitcoin space is Stepan Sinjarev. And so he is actually working on a quantum computer at, I believe it’s called PlanQC. So he’s like a sighted researcher in the field, like postdoc, working at Max Planck and things like this. But he also worked on Spectre Wallet, as many of us know. And he worked on, I believe, the underlying library, I think it’s called MBIT, which is underlying like Spectre DIY. So he’s probably one of the unique people who’s actually like a genuine expert

in both sides of this. And I recall seeing him speak at a panel, and this is in November, just of last year, so like two or three months ago. He was saying like, only start worrying about this in like 20 years from now. So you have like this whole range of different views, right? When a genuine expert in the field cited working on this, he’s saying, no, it’s like 20 years out before you even really worry about this. And there’ll be like weaker.

James O’Beirne (20:16)
Mm-hmm.

Stephan Livera (20:28)
Other systems and cryptography that breaks down before you know they get to Bitcoin You know to me it sounds like well, that’s pretty reasonable And then there are other people like ⁓ you know there some other talks and things where people sort of saying it’s like you know 20 years out before Bitcoin is You know at risk sort of thing, then you have other people saying no no look look at their timeline It’s coming in two to five years. We need to like do a coordinate a big soft fork now Yeah, it just kind of I guess to me the way I’m seeing it is like

I’m not an expert, just an ADIQ podcaster, but it seems to me like if it’s a real thing, it’s more like 15 or 20 years out.

James O’Beirne (21:08)
Absolutely. That’s my understanding as well. And even I remember Hunter Beast, who’s proposed BIP360, very nice guy. ⁓ But I remember him, you know, kind of making a lot of noise on Twitter about this experiment some other guy on Twitter had done where he allegedly broke 5-bit ECDSA. ⁓ But it wound up that, you know, that wasn’t…

at all true in a quantum sense. And there was a lot of classical processing done to kind of arrive at that result and a lot of ⁓ guessing, you know, that doesn’t scale well with the number of bits. So I just think we’ve, you know, at a fundamental level, we’ve seen no basis whatsoever that this is anywhere near around the corner. We’ve seen no basis whatsoever in my thinking that this is a decade away, two decades away. ⁓ So.

I think it all does come back to that in a sense. ⁓ But yeah, again, I am all for DIP360 as written, but it’s really ⁓ nothing to do with quantum.

Stephan Livera (22:12)
And so then, I guess ask it in a more pointed way then. Let’s say, hypothetically, it advanced like, okay, let’s wave our magic wand. Let’s say BIP360, the community rallied around that and said, yeah, let’s do this soft fork. But then maybe in the future, it’s actually, no, now let’s do a soft fork to actually bring in.

you know, shrinks or Sphigs Plus, or, you know, an actual, you know, and then at that time there might be a conversation around, well, are we going to raise the block size or have some kind of quantum witness discount to compensate for the fact that these quantum signatures are bigger? And then also it’s also a more real thing. There are real costs that we will have to pay. So for example, we will no longer have HD hierarchical deterministic wallets, know, things we’re going to lose music to and frost. We’re going to have silent payments, you know, theoretically, like some of these things would have to

to change and even our hardware wallets might need to be updated and upgraded to deal with quantum signatures or post quantum crypto. So let me ask you this way, do you think, would you be for or against that, let’s say that second step?

James O’Beirne (23:20)
I think it’s an absolute nightmare. ⁓ If you just even look at how long it took to roll out SegWit, debatably Taproot hasn’t even been fully rolled out in many ways. Most major custodians I’ve worked with don’t support ⁓ spending Taproot outputs they can send to Taproot. But ⁓ we all know that the deployment process for Bitcoin ⁓ takes a very, very long time.

in a full ecosystem sense. And a proposed change like that for quantum ⁓ would take a very long time and a lot of effort. But that’s not even my fundamental objection. Okay. I look at cryptography the way that I look at, say, pharmaceuticals. ⁓ It’s allegedly old nurses wisdom that you should not take a drug that hasn’t been on the market for more than eight years because

You know, these are very, very complicated things that we just don’t fully understand. And really the only meaningful test harness is reality. Cryptography to me is the same way. ⁓ You know, you can do your reasoning on paper, but until you deploy a crypto system and put some actual ⁓ risk behind it, ⁓ you just, don’t know what’s what your unknown unknowns are. And I’m very concerned that

The quantum froth is going to lead to a situation where we even consider committing new cryptography to Bitcoin that hasn’t seen that kind of testing and battle proofing. ⁓ I think that would be a horrible risk, even if it’s just opt in, because if you can convince enough people that there’s reason to add this cryptography to Bitcoin, it’s a very short leap to convince a similar number of people that they should, you know, encumber their coins using these schemes. And then what happens when

you know, we find out that we didn’t understand the scheme as well as we thought we did, even if in the case of like Sphinx, it’s based on SHA-256, which makes it sound very safe. But, you know, there’s a big difference between being convinced of the efficacy of SHA-256 as a cryptographic hash algorithm versus SHA-256 composed, you know, together to make a signing algorithm. ⁓ There’s kind of a big gap there. So,

I’m really, really averse to the idea of ⁓ adding any new crypto to Bitcoin unless we really, really need to. And I don’t think there’s there’s anything near a justification for that for quantum.

Stephan Livera (25:51)
Yeah, okay. And so we’ve spoken about the bit 360, Hunter Beast and Ethan and Isabelle, think the other author on that. So that’s, I guess, one idea. There’s also this Matt Corrello plan. So I don’t know how much you know about that or if you’ve kind of looked into that, the gist of it, are you familiar with it or no?

James O’Beirne (26:00)
Mm-hmm.

I think maybe you or somebody mentioned it recently. It’s the idea of cloning Taproot and creating a new ⁓ witness version that’s basically just Taproot. But when you spend to that witness version, you have ⁓ sort of no expectation about being able to use the key spend. Is that right?

Stephan Livera (26:34)
Yeah,

sort of, yeah, yeah, so it would be, think, so the way I understand it is like, we would have existing wallets, you know, include this kind of post quantum tap leaf in the script path, while people are still using the standard, you know, tap root stuff today. And then the idea is inside of that, people can do whatever they want. And it’s not even revealed on chain today. So it wouldn’t actually impact, but, but,

post, let’s say, hypothetically Q day comes and the quantum computer is here and everyone agrees. And I guess then at that point, the idea would be to have a soft fork to disable the key path such that those users can now spend using their quantum, know, tap leaf fancy script path into whatever the new, you know, quantum scheme would be. at least in that way, it’s not.

taking up massive signature space, at least today, and it would allow, but of course there’s all these moving parts of like, it a gradual break or is it a sudden break? And then can the community agree, yes Q day has happened and it’s time to now do something. But I guess as you understand that, let’s say the Matt Corrello quantum tap leaf plan, are you for it, against it, or do you have any thoughts?

James O’Beirne (27:35)
Mm-hmm.

It’s inoffensive to me. mean, it’s not horrible. ⁓ But I just think encouraging people to do things like that with all the added complexity just really isn’t merited based on what we know. It’s certainly something that you could resort to. It’s a plan to have. And maybe that’s the way that Matt meant it is a kind of contingency if, you whatever, five or 10 years from now, we have some new evidence that, no, this is actually getting closer. ⁓ You know, maybe that’s a fine halfway point, but

I just think there’s a real risk in getting people riled up about this stuff rather than saying something like, you know, look, we’re technical people. have good understanding of cryptography. ⁓ and we really just don’t think this is a problem here. Here’s maybe our list of contingency plans. If we are presented with evidence, the contrary, but you know, I think it’s our job as engineers. ⁓

to make those kinds of judgments and basically give people our best assessment of what the trade-offs are. And when one side of the trade-off comes with tremendous complexity and something like that, ⁓ that really sounds simple, for businesses and individuals and the wallet ecosystem is a pretty big lift.

I just think we’re better served by saying, okay, we could do that in theory, but let’s wait.

Stephan Livera (29:21)
Yeah, so wait and see. Yeah, it’s an interesting ⁓ point. I’m kind of inclined to agree there, but yeah, I just don’t know what are people gonna think. And then the other, I guess the other point with that, the quote unquote Matt Corello plan, let’s say, I don’t know if there’s like a proper name for this. Like I didn’t have to say with him, I put it out recently, but I guess the other thing in his mind was,

because people can use like, because a lot of the coins are secured.

like people have a bit 39 seed phrase and there could be like some kind of ZK proof to prove that they still have the seed, you know, the seed phrase for it and then use a ZK thing to then allow them like hypothetically if you had disabled, you know, the key path spend that those people could still recover their coins. So then in practice, it would just be like some of those early miners and maybe Bitcoin core wallet users and maybe some of the big exchanges and custodians and okay, they would do something fancy and do some other thing. ⁓

like some kind of commitment thing, pre-commitment scheme. So I guess that’s one thing. But at the same time, this would also be, mean, assuming quantum is real, all this stuff, it would still be kind of chaotic because then we would have to think about,

James O’Beirne (30:26)
Mm-hmm. Mm-hmm.

Stephan Livera (30:37)
How long is it gonna take for people to transition over? And are we gonna have a bigger block size to allow that, to allow more people to actually transition out? And how big is the ZK proof? I think ⁓ David Vexler, I believe, had an idea for it, but from what I read on a blog post, it looked like that ZK proof might be 500 kilobytes. that might be a non-starter. If every person has to put 500 kilobytes on chain, maybe there’s some way to do a proof so it can be a smaller amount that goes on chain and not the full 500 kilobytes.

James O’Beirne (30:43)
Yeah.

Yeah.

Yeah.

Stephan Livera (31:07)
But yeah, it’s just, there’s a lot of complexity around that, isn’t it?

James O’Beirne (31:11)
Yeah, 100%. And, you know, again, I don’t think there’s anything wrong with some of these like, pro-volactic schemes where you’re hiding something in the tap tree for later reveal. I think the only qualifier there is that, ⁓ you know, ideally it shouldn’t have spend power today because what you don’t want to do is turn your tap root metadata into this toxic waste where if anybody…

gets a hold of it and there’s some kind of a fundamental issue. it, you know, say you’re spending to, ⁓ you know, like an op success or something in anticipation of a new crypto scheme. If that gets loose.

⁓ all of sudden now, you your coins are at risk of theft even before anything has happened. if it’s a scheme where it’s like, okay, well, we’re hiding this thing in the tap tree and then at some point, maybe we can make some consensus change that turns it on and, you know, makes redemption given a ZKP or something possible. I mean, again, like, you know, my position is go nuts, have fun. ⁓

But if I was working at a company, like ⁓ in the capacity that I am now, ⁓ I would not advise that.

⁓ because it’s a ton of work for something that’s highly, highly speculative. I mean, all of this stuff, know, the analogy that a friend of mine came up with is like, it’s kind of like running around and being really, really fearful for the airline industry that like teleportation is two years away because we have like the AIs thinking about, you know, how to do teleportation. It’s a…

Stephan Livera (32:43)
you

Tea Day is coming.

James O’Beirne (32:50)
Yeah, exactly. Yeah.

For me, you know, again, after years of kind of ruminating and reading about this stuff, you know, albeit cursorily, but like it kind of strikes me as a similar thing. Now, granted, we don’t have a teleportation industry, really. So maybe it’s not totally analogous, but I think that’s like the level of real threat that we’re dealing with here or approximate.

Stephan Livera (33:18)
Yeah, okay. Well, I’ll let ⁓ listeners make up their own minds on that. If you’ve got nothing else to say on the quantum specific side of things, let’s talk just kind of Bitcoin protocol. know you maybe, you have, let’s say, your own views about how Bitcoin protocol development, let’s say, should have gone versus how it has gone. I know you were kind of flirting a bit with big blocker ideas.

but also really interested in the Covenant side of things, obviously, as you were involved with the CTV and OpVault and ⁓ some of these ⁓ related discussions. I guess, let’s hear from you, of Bitcoin protocol side of things. Where are you out there? Like, what’s your overarching view?

James O’Beirne (34:09)
Sure, yeah. know, Stephan, I am, as you know, a contrarian and that’s, I think, many of us are, obviously, if you’re in Bitcoin, you have maybe a disagreeable temperament that leads you to, you know, constantly be prodding at things. And I’m certainly that way, as many of the listeners may know already, but…

I’m glad that you brought up the big block stuff because I do find that really interesting. And actually, you you said something earlier about the ZKP proof size that makes that somewhat of a relevant question. Hopefully not too relevant anytime soon, but, you know, maybe, maybe the, the talk that I’m most proud of ever having given was at the last Opnext. think that was 25. I gave a talk where

Basically, I wanted to sort of reevaluate with fresh eyes a modest version of the big blocker view of the world, which is to say that maybe instead of doing, maybe second layer scaling isn’t going quite as well as we all thought it would. ⁓ And so what does the world look like where we kind of turn back to something like the block size as a method to scale?

And the way I went about this is I basically modified Bitcoin Core to give me a bunch of timing information. I was sort of putting a clock around every major operation that happens ⁓ between receiving a block and storing the UTXO, all that stuff, because I wanted to get a sense of how ⁓ my consumer hardware on residential bandwidth, how much headroom I basically had to work with.

And I was very, surprised at the results, which I won’t try and, you know, go too much into. You can go watch the Opnex talk if you like. It’s something called something like a modeling block size constraints. But the long and the short of it is that I was very, very surprised with the amount of headroom that my, you know, thousand dollar computer connected to gigabit internet had. And it prompted me to kind of evaluate this larger question about like, what’s actually the trade that we made?

in terms of, ⁓ you know, being really principled about small blocks. ⁓ And whether that was just kind of this exercise in, you know, fetishizing like Raspberry Pi node runners and, you know, these like hyper constrained situations where you have to beam, you know, blocks on satellite. And, you know, I was, I was absolutely one of those guys in the, in the block size war. So I was as small blockers as they come. ⁓ But

You know, to be honest, the things that I measured really forced me to reevaluate and kind of juxtaposing that on how things have gone with lightning. mean, people forget that in the original lightning white paper, know, Dreyer and Poon basically say like, hey, the only way this stuff works in the way that we want it to work is with a block size increase.

Stephan Livera (37:16)
Right,

I can’t remember the exact number, but it was like 120 megs or something like this, right? That was the number they gave back in the, I think, 2015 or 16 Lightning white paper.

James O’Beirne (37:20)
Yeah, yeah, yeah, yeah.

Yeah, yeah, yeah, that’s right. Exactly. And so, you know, I think it’s an idea that’s, that’s really gotten a bad rap. It’s fallen out of favor. mean, obviously like the, you know, they’re laughable, ⁓ like parameters, like the, you know, the BSV guys doing gigablocks or like that’s yeah. Yeah. Like that’s completely ridiculous and cartoonish, but, I think like writing off the idea of ever touching the block sizes, ⁓ is, not only like,

Stephan Livera (37:40)
Right, the giga-mags.

James O’Beirne (37:53)
ill-advised but maybe unrealistic ⁓ depending on what happens. So anyway, yeah, so that’s something that I was kind of looking into just as a fun contrarian exercise. ⁓ But to your general question, how do I feel about Bitcoin protocol development? ⁓ I’ll be honest, when you asked me to come on, I was kind of surprised. like, I’m like washed up and semi-retired at this point. What does he want to talk to me for?

because, because of, ⁓ you know, I think I, ⁓

I had this experience ⁓ navigating the process of proposing a consensus change. I wasn’t really thrilled about the way that Bitcoin development as a human system is set up to kind of process and evaluate that. And a lot of people think, I would expect that that really burned me out and jaded me, but

The honest truth of it is that, ⁓ you know, from the very beginning of my development career in core, which was like 2015, it’s really just been an uphill battle to do anything useful. And that’s for a variety of reasons. think ⁓ one of the things that people forget is that in as much as Bitcoin is an experiment in crypto economics and distributed systems and ⁓

Cryptography, it’s also an experiment in human organization in terms of how the development works. ⁓ you know, my belief is that for many years, maybe up until Taproot, ⁓ Bitcoin operated under this sort of implicit hierarchy where there were a few guys who really, you know, if they wanted to go a certain direction and it was like plausible and there weren’t any obvious problems, like that’s the direction that things went. And…

What happened was that those guys kind of abdicated in the sense that they stepped back and, you know, in some cases explicitly said, Hey, I’m not going to do this consensus stuff anymore. And that created this ⁓ vacuum that I think everybody thought was going to be filled by some kind of a like leaderless meritocratic, you know, cypher punky online human organization. And I think that was kind of naive in hindsight. And I think it.

That vacuum hasn’t been filled and the protocol hasn’t meaningfully progressed since Taproot. And, and really there are no signs that it’s going to, ⁓ despite some very credible proposals, you know, some serious work that’s been put forth. ⁓ you know, me and Jeremy spent a lot of time on CTV. mean, Jeremy, especially for years and years and years. And, ⁓ I started out as a, as a sort of CTV, misunderstander critic. didn’t get it.

when I was first presented with it maybe in 2019 and then you know when I started working on custody and vaults and things like that I finally got it and it clicked that was in 21 and You know there were there were a lot of well-intended people you know working on that proposal and just You know the way that it was received by many who are high status and Bitcoin ⁓ you know didn’t strike me as productive and

⁓ really, I think it’s an interesting time to kind of re-examine all this because, ⁓ you know, I ultimately wound up kind of organizing, ⁓ what, what some perceived as an ultimatum. It was really just meant as kind of an open letter of this, you know, ctv-csfs.com, which is still up. You can go read it if you, if you aren’t familiar, but the idea was that, you know, many of us who wanted to keep, you know,

pushing forward this idea of covenants in a concrete way, ⁓ were taken aback by ⁓ how little progress had been made. so, a lot of very, very smart, long-time contributors signed this letter basically saying, hey, ⁓ we recognize that Core, or least theoretically doesn’t like totally run Bitcoin, but at the same time as the most popular implementation.

If a consensus change is going to be even evaluated, we need some engagement support from core. So we’d really like you to review these pull requests and merge the substance. know, guys like Andrew Polstra signed that letter. ⁓ A lot of Hashrate signed that letter unintentionally, really. ⁓ And that was back in June. ⁓ You know, and there was sort of a six month, hey, you know, can we can we get this done by then? That would be

put us at December 25, it’s now February 26. So that effort sort of failed. And ⁓ that was about as composed an effort as I could muster. ⁓ What we got in response was sort of a counter proposal from Sanders and Poinso that, you know, in plain terms, nobody really paid much attention to and just kind of served to fracture the conversation.

Stephan Livera (43:22)
Right, this is the template hash

⁓ proposal. So in your view, how does template hash compare with CTV and CheckSync from Stack?

James O’Beirne (43:31)
Yeah, Template Hash is basically just kind of a less featureful version of CTV in the sense that ⁓ it’s Tapered only, which is a major constraint for custodians because it turns out that if you’re a serious custodian, probably you’ve got an HSM deployed at the moment and ⁓ almost, well, no HSM on the market has native Schnorr support. So ⁓ I’m not aware of a single

Tap a single custodian at scale. That’s, you know, supporting Tapper, maybe, maybe Bitco. think actually, ⁓ I know somebody over there who, who yes, yes. I didn’t want to say, not knowing if he was public about that, but yeah, yeah, yes. ⁓ so, but even then I, I’m not sure if that’s, I’m not sure how mainline that is within a bit. Go. He’d be able to tell you, ⁓ I can just say that, you know, a of the custodians I’ve.

Stephan Livera (44:10)
Right, because they didn’t Brandon Black work on their music too.

James O’Beirne (44:30)
I’ve looked at work with, ⁓ you know, don’t really have plans to support Taproot because there isn’t a compelling use case given how hard it is. And so, you know, with template hash being Taproot only, you have certain uses that that just kind of rules out in terms of doing things like ⁓ vaults and presigned transactions or replacing the need for presigned transactions. The other thing is that ⁓ one of the…

the appealing uses of CTV was kind of this off-label use that the BIDVM guys ⁓ had come up with, which was essentially committing to ⁓ other inputs within the transaction. And Template Hash strips that out. The other thing that Template Hash does, and maybe the most fundamental objection that guys like me and Jeremy Rubin had, is that it commits to the annex. And we don’t…

really know what the Annex is for yet. And so I just kind of think it’s premature to lock ourselves into that.

Stephan Livera (45:31)
⁓ yeah.

I mean, you correct me, you probably know better than me, but I thought there were some forms of LN symmetry that use the annex. I thought that was one thing that apparently might be used for, and I guess Greg Sanders had done some work on LN symmetry, so I don’t know, maybe it’s that, I don’t know. Do you know?

James O’Beirne (45:54)
Yeah,

yeah, I think that maybe was the motivator was that he had applied it somehow. ⁓ He wanted template hash to work with Alan symmetry. My objection was, you know, these are sort of orthogonal concerns. You could leave CTV as is and, you know, simply introduce some opcode that’s like, you know, allowing you to commit to the annex if you want to. ⁓ But I’m sure there may be cases where you don’t want to, for example, you know, one of the proposed uses of the annex may be

if you ever get to cross input script aggregation, ⁓ there may be some metadata that you need to shove into the annex. And so I just think it’s kind of an unnecessary design choice to make that commitment if you don’t need to, rather than decomposing it. So anyway, it’s not really that important because the upshot is that the conversation kind of fell apart because…

You had all these people who were very enthusiastic about using CTV and, you know, the sort of core ivory tower guys came along and said, well, we don’t, we don’t like CTV, but we like this, this other thing that, you know, we, we think does a better job. ⁓ but unfortunately, you know, people’s bandwidth is limited and, and, ⁓ you know, I think there was, it’s, very hard to reproduce that kind of Lindy accumulation of review and.

⁓ understanding that CTV had had. And again, it kind of comes back to the fact that this consent manufacturing process that used to be very robust in Bitcoin, that was very robust during the SegWit, Taper days, you for better and for worse, that just kind of no longer exists ⁓ because you don’t have as strong a leadership hierarchy within the technical community.

Stephan Livera (47:47)
I see. I’m curious your view on, let’s say, like, think about like an AJ Towns, where he’s got this kind of slow and steady philosophy in terms of how, you know, or he’s written a bit about how he thinks, let’s say, consensus changes should be pushed forward in Bitcoin. So as an example, there’s this Inquisition, which as you know, but for the listeners, it’s a Cygnet, which is sort of like a testing ground for various Bitcoin ideas and…

I think even right now actually there is a bit 54 consensus cleanup, which is now just recently active in that Inquisition Cignet. And so I’m curious your thought on that. Like is it just a matter of like slow and steady? Is it just a matter of like it just has to sort of take a longer time necessarily?

James O’Beirne (48:35)
Well, I think nobody’s saying fast and loose, obviously. But yeah, ⁓ there are different gradations of slow and steady. AJ and I have worked together a bit. ⁓ We didn’t have a great working relationship for some reasons. ⁓ to my knowledge, ⁓ right now he’s primarily focused on ⁓

Stephan Livera (48:41)
course.

James O’Beirne (49:03)
like a Lisp dialect that he’s ⁓ working on as a kind of maybe alternative to simplicity. ⁓ And I just look at those kind of changes as, and this might surprise people that I believe this, I look at those kinds of changes as inappropriate for Bitcoin at this stage because…

I think Bitcoin should be done with these engine lift style, massive overhauls that just introduce reams and reams of new code and substantial updates to the scripting model. I think those days are over for Bitcoin. think like we have and should ⁓ ossify past those things because those are truly existential changes to the scripting environment. Adding something like say CTV or…

Check template verify, I’m sorry, check contract verify, which is Salvatore’s ⁓ great proposal. Those are more akin to ⁓ adding time locks rather than something like SegWit or Taproot, which were really these like.

Stephan Livera (50:05)
Right. You see them as marginal

improvements rather than wholesale, take out the engine and plug in a new engine kind of changes.

James O’Beirne (50:09)
Yeah. Yeah,

yeah, exactly. And know, Taproot was done in such a way that it’s very, upgradable. ⁓ know, Taproot was, you know, in many ways built as that framework so that, okay, we don’t have to do this giant lift anymore to sort of change the world in terms of scripting. So I look at Simplicity, I look at Bullish, which is… ⁓

⁓ AJ’s thing and I’m again, I’m just very skeptical because if you know simplicity is supposed to be simple but if you if you actually go and look at the code number one, it’s horrifying and number two ⁓ You’re shifting the question of what is consensus into what do the jets do and which jets are you offering and to make you know certain operations efficient

Like say SHA-256, you you need this like jet for that, which is an escape hatch because simplicity itself is this hyper simple, you know, constrained language. So you’re really just kind of like moving the goalposts into what gets a jet, what doesn’t, you know, how consensus compatible are the jets and so on and so forth. I’m not saying there wouldn’t be benefits to simplicity. I do think there may be, but that’s not incorporating the price of the complexity of the changes, which I think are substantial. So anyway, ⁓

know, in terms of AJ’s slow and steady thing, I mean, I think that’s easy rhetoric to give. ⁓ But the reality is that ⁓ Bitcoin does have this natural hardening process, which should exist. There should come a day where we’re really not touching.

the scripting of Bitcoin. ⁓ I think that day actually may already be here just because of the organizational realities of how Bitcoin changes. you’ve got guys like Michael Saylor saying, you know, like, no way, get the developers away from this thing. ⁓ And, ⁓ you know, that used to really, really rankle me. And I think I’ve kind of hit ⁓ a zen.

moment with it more or less because and partially thanks to you and other people on you know other other other friends of ours I’ve kind of realized that the the limitation isn’t really like like yeah Bitcoin itself is technologically limiting but but really the fundamental limiters like

people and the way that people want to use Bitcoin. you know, the reality is like most people just don’t want the responsibility of doing self custody. And I thought for a long time, I still think ⁓ that it’s my job as an engineer to make that easier, to make it better, to introduce tools that, you know, let people take…

possession of coins without really, really worrying about it or without it being a really, really hard process. You know, it’s my job to try and future proof the protocol when it’s like the monetary backbone of the world and, you know, ⁓ increasingly limited numbers of people can actually get access to holding UTXO themselves without a trusted intermediary. I look at that as my job, but I also understand that reality is what it is and…

Bitcoin may just kind of have run its course in terms of these script level changes and, ⁓ you know, we’ve got what we’ve got and we should make the best use of it.

Stephan Livera (53:28)
Yeah, interesting. so, ⁓ obviously, setting aside things like there’s like this timestamp bug in like 80 years from now, whatever, like, assuming everyone’s going to be, you know, cool to do an upgrade for that. That’s not a big deal. But certainly, because that’ll be a hard fork. But from what I’ve heard, it’s like the other side of that will actually not create a block because like it’ll run out and anyway, that’s kind of a small thing.

James O’Beirne (53:37)
Mm-hmm.

Stephan Livera (53:58)
If you like, cause yeah, cause there is this question of should Bitcoin protocol development really go hard down this pathway of like CTV or LNHance or CCV and some of these ideas that enable things like, you know,

LN symmetry or make it easier to do things like arc and these various forms of L2s that then enable better self-custody. And on one side of it, it’s like, I mean, obviously you and I obviously would like more people to self-custody, but at the same time, as I’ve mentioned before, it’s like, where’s the real ⁓ bottleneck? Is it actually just like a social and economic thing that…

Even now, if we look on mempool.space right now, how costly is it to get next block? You can get ⁓ one sat per V byte. In fiat terms, it’s nine cents to get next block confirmation. So it’s all like, we can talk about all these people doing self-custodial lightning and things like that, and then there’s actually just like all these people just using.

James O’Beirne (54:50)
Yeah. Yeah.

Stephan Livera (54:59)
either custodial things or L2 stuff like lightning and arc and spark and ecash and feddy and know liquid and whatever. So I guess where do you sort of land on that? What’s the next you know adoption pathway?

James O’Beirne (55:13)
It’s a great question. ⁓ You know, I had this experience recently where a few years ago I set up this ⁓ home security system, these cameras that, you don’t talk to the internet. They’re cheap Chinese cameras that I’ve network isolated. It’s totally self-hosted, all this stuff. ⁓ And I love it. So I made a Twitter post a few weeks ago saying, I this thing. It’s cool. And the response was overwhelming. People were like, man, this is so cool. I really want one.

And I thought, okay, well, maybe I should productize this or something. But then it occurred to me that, you know, I spent like a good two weeks of difficult, you know, like running cat five through your house is not easy. And ⁓ it dawned on me that a lot of people think that they want a self-hosted camera system, but probably what they really want in terms of a user experience is like throwing up.

three wireless cameras and then having their Samsung TV display a picture in picture when someone’s at the front door, which I totally don’t have. It’s this crazy hackneyed setup that works for me because I’m highly technical. And so it was just this experience of realizing, you know, it’s this classic issue that people say they want privacy. People say they want self custody. They say they want to be resilient. But when it comes to actually kind of doing the work to make it happen, it’s very difficult. ⁓

So yeah, I really thought that things like OpVault and CTV could create this situation where you ⁓ could just have a cell phone and maybe an exchange account and you would have near cold storage level security just receiving through your phone. That was kind of my big golden dream. ⁓ And then I thought, okay, well maybe then…

people will actually take possession of their coins in larger number. But, you know, I think the reality is that even then ⁓ people want like to be handheld. They want a great user experience. They want to be able to call somebody and say, Hey, what’s going on? You know, this isn’t working or they don’t even want to get to that, that, that situation. And so the exchanges definitely offer that in terms of future tech, ⁓ you know, I’ve tuned out a little bit in the past few months. mean, I know some of the arc stuff was very exciting.

I mean, in fact, ⁓ it, ⁓ is the company called Second? Steven Roos and Eric, yeah, yeah, yeah. Lovely guys, lovely guys. just, you know, couldn’t be nicer, smarter guys working on this stuff. And so I’m really happy that they are, but Steven was one of the big guys pushing for CTV, CSFS, because he said, man, it just, it reduces the interactivity requirements. It allows us to do things in arc.

Stephan Livera (57:41)
Yes, yes, Stephen Aroos and Neil would find and the guys, Eric, yeah.

James O’Beirne (58:04)
⁓ that are really, you know, create this better user experience that people are actually going to want to use. You know, I think to, to, absent some kind of tail risk, you know, black swan event that really puts us in a, in a bad situation, ⁓ you know, legally or geopolitically in order to get people using Bitcoin natively for things like payments, ⁓ you know, or even storing their own wealth.

⁓ you need to create a user experience that’s like on par with credit cards. And that’s very difficult, super difficult. and so I kind of, I’m glad those things are being worked on as a, as a sort of, you know, ⁓ pro-volactic for, for if that day ever comes and, know, we’ve got this, this, this backup system that’s like works well enough. I think that’s great work to be doing, but I just think we’re going to, we’re going to probably continue to be disappointed.

by systems that roll out and it turns out that, you know, they’re either more proprietary than we thought or, you know, they don’t work as well in practice as we thought, which I think, you know, is my assessment of how lightning has gone. think hype for lightning in like 2020 was so strong, so big. And I think almost everybody in varying degrees feels a little bit of a hangover from that because it

It didn’t quite materialize maybe in the way that we thought. Now other people say it works great and they’re probably right. ⁓ But I think expectations were high and so we should rein that in for any subsequent layer twos.

Stephan Livera (59:44)
Yeah, I think it didn’t happen the way we all thought it would. And I put my hand up as well, like I was obviously a promoter of lightning. I still am to be fair. But that vision of everyone having a node in their own home and connecting to it over tour and paying. Now I did that, but most people were not. And it just kind of…

it’s sort of becoming more like a professional’s game. It’s more like a professional’s game of running lightning and being an LSP and it’s not really seen as like a quote unquote everyday pleb.

⁓ thing nowadays. ⁓ Now, you know, maybe we’re just kind of in a local bottom, you know, like maybe there is an element of growth, like maybe we’re going to see a lot of FETI instances come up and that’s going to be like one way that like lightning gateways are around there and you know, maybe we’re seeing…

improvements in the self-custodial, like at least the mobile UX, right? Phoenix is amazing. I use Phoenix like it’s my daily driver. Wallet of Satoshi have a Spark self-custodial version now, which works pretty well. The ARC guys are working and they’ve got their own kind of version of it. The Bull Wallet has kind of liquid, but Francis is experimenting with ARC in the background there. So, it’s almost like in one way, maybe we scaled too well, loosely speaking, right?

because so many people are now using these different L2s or whatever you want to call it. I know some people get stickler about like is liquid an L2 and so on, but whatever. Like if we just consider them broadly as L2s, they’ve brought the fees down a lot and people are using them now. And Coinbase and Binance and OKEx and Bitfinex and all these big exchanges, they have lightning now. So in that sense, we won, but it’s still ⁓ not the…

James O’Beirne (1:01:29)
Mm-hmm. Mm-hmm.

Stephan Livera (1:01:35)
It’s not as, let’s say, trust minimized as maybe ⁓ the initial dream was.

James O’Beirne (1:01:43)
Yeah, well, it’s very much like the development of the internet, right? I think if you and I were hanging out in 93, you know, we’d have these grand visions about everybody having a mail server in their house and, you know, it being this, great decentralization of data. And it’s a bit of a sobering comparison because if you look at the way the internet functions now, I mean, obviously it’s, it’s magical in many ways, but in other ways, everybody’s either going to like TikTok, Facebook.

X Instagram and that’s, you know, in some ways like the entirety of the internet for 99 % of people. Um, so yeah, I think, uh, it is that principle that running infrastructure is very hard, especially reliably. And so if you have say a FedEment installation, um, you know, or, or an arc, uh, an ASP, um, you have to make sure that stuff stays up.

And that’s not an easy problem. I’m speaking as a ⁓ former DevOps guy, know, still current DevOps guy from a household. And, ⁓ you know, even for someone like me, it’s challenging. ⁓ So yeah, maybe at some point we’ll get some kind of a technological bailout and ⁓ there will be some incredible way to do this stuff. But I do think that things are going to centralize just as a course of nature. And ⁓ maybe that’s okay.

The AI stuff is very interesting because I do think it brings a lot of power to ⁓ individuals to run these kinds of things. ⁓ know, people who maybe otherwise like weren’t programmers, don’t really know how like Unix works, but they can get the AI to, you know, teach them essentially how to, how to do this stuff. think that’s kind of cool, but it’s, that’s probably a marginal phenomenon rather than some kind of ⁓ a huge phase shift in the way that computing works.

Stephan Livera (1:03:35)
Yeah, it’s almost like the limiting factor is agency, right? As in personal agency, the kind of person who will go and try and use this tool now to figure it out, whereas maybe before it was like you had to be a real, quote unquote, hacker man to get in there and figure out, here’s how you do your command line stuff and whatever. I guess another thing we should chat about is obviously you were the guy behind Assume UTXO. And of course, maybe some of this ties into kind of the big blocker stuff as well.

because these approaches like Assume UTXO and U-Tree XO, some of these ideas, maybe they can be sort of combined in such a way that quote unquote grandma can have an easy validating node that spins up really quickly. So do you want to just comment a bit on that? in terms of like, obviously you were the guy behind Assume UTXO, do you want to give us any updated thoughts on that and what it means going forward with Bitcoin node syncing?

James O’Beirne (1:04:33)
Yeah, I couldn’t be more bullish on that arrangement. ⁓ I actually am kind of maybe working on something in this department, but ⁓ you’re exactly right. The union of, assume UTXO and UTreeXO and pruning, ⁓ well as technologies we’ve had for a while, like ⁓ compact block filters. Yeah.

Stephan Livera (1:04:58)
compact block filters that bit 157,

158.

James O’Beirne (1:05:01)
Yeah. Good memory.

Good memory. yeah, the union of, of that tech creates a really, really exciting node environment, I think. And, ⁓ I’m sort of almost amazed that nobody has gone at it. I might have to myself because I think when you combine those things, you create a situation. Yeah. Yeah. You could, you.

Stephan Livera (1:05:21)
Yeah. It is pretty impressive. Yeah. I mean, I’m curious, have

you seen the Floresta project?

James O’Beirne (1:05:28)
Yeah, I have, have. And I think that’s a really cool direction. I’m totally supportive of what those guys are doing. And I think there could be something very exciting there. I’m a little bit of a rust bear. was kind of poking through their cargo dependencies file and there’s just a lot of stuff in there. And when it comes to, you know, things like node software, I’m very, very conservative in terms of the dependencies that.

that I take on. So, but, in general, I’m, I’m, I’m highly supportive of stuff like that. And I think, I think that’s a very exciting prod project with, with some very exciting features behind it. But even if, you know, what’s funny is like, even if core itself was packaged with slightly different defaults, like if it was set to say, retrieve and assume UTXO snapshot, the latest one from whatever Bitcoin core.org, which again, isn’t, isn’t a security assumption. It’s not a trust assumption because the

hash of whatever is being retrieved is hard coded in the source code, it’s checked against whatever snapshots being loaded in. And then you configured pruning, ⁓ UTreeXO is a little bit space age and a bit more of a kind of a scaling feature, but even just those things, you know, you’d have a lightweight ⁓ node up in operation for most people, I think within like an hour. So I think that would be really, really exciting.

⁓ And yeah, maybe you’ll see some developments from me there. Maybe not. We’ll see.

Stephan Livera (1:06:57)
Yeah, and so I’ve got a few directions to go here, but as an example, mean, last I checked, maybe one or two days ago, I looked up these numbers. So on the kind of global node stats, I think this is Luke Dasher’s stats actually, apparently there’s about 88,000 Bitcoin nodes around the world, and of that 88,000, it’s about 24,000 that are reachable on, I think, Bitnodes dashboard. So.

You know, isn’t it kind of funny that we have, you the estimates for Bitcoin adoption is something like 100 million, maybe 200, maybe 300 million, something in that range. And then out of all these people, it’s only, you know, call it 100,000 people who are running a Bitcoin node on, know, why is that in your view? Is it just, again, the agency aspect of it? Like most people are happy to just trust.

James O’Beirne (1:07:41)
Mm-hmm. Mm-hmm.

Stephan Livera (1:07:50)
Trezor, Ledger, servers, or their exchange, this kind of thing.

James O’Beirne (1:07:55)
It really is. It’s like, well, what benefit am I getting out of this on a daily basis? You ⁓ you and I know there’s an abstract benefit, but it’s very abstract and it’s, hard to quantify in your daily life. And, ⁓ you know, again, I, just, want to reiterate, keeping computers up and running reliably is difficult, you know, managing a network connection between say your home computer and your mobile phone. It’s, it’s not hard, but for most people, it’s just a domain. They, have no.

experience with familiarity with they don’t they don’t really know what’s possible what what isn’t how hard it is how hard it isn’t and so Yeah, I think it’s kind of the same. It’s you know It’s it’s the same problem is like ⁓ why aren’t you know more people writing their own lightning note or Fetty mince? ⁓ It’s it’s that same sort of thing

Stephan Livera (1:08:44)
What about this concept of just mobile only? Because we are kind of moving into this world where there are people who don’t have a laptop or a desktop, they just have a mobile. And maybe the UTRIXO thing of like having a compact state node or a floresta thing, which is like, you know, the proverbial grand mark and just have a phone that is a validating client, let’s say something like that. But where are you at on this concept of mobile nodes and mobile, yeah, just mobile nodes?

James O’Beirne (1:09:07)
Mm-hmm.

The hardware there is getting really impressive. I mean, it’s been really impressive for a while and…

Bandwidth doesn’t seem to be a limiter by any stretch of the imagination. So I think it’s technically feasible probably, you today ⁓ Even if if you’re just saying okay, we pick a sort of aggressive ⁓ Assume UTXO height that’s you know, relatively recent or whatever ⁓ in terms of just receiving the block and Checking the SIGs. I haven’t done any benchmarking. You would definitely need to do that. That doesn’t strike me as a ⁓

as an inconceivable thing, it’s probably, you know, ⁓ very easy from a computational standpoint for these phones. What ⁓ winds up being the kind of unexpected difficulty with something like that, which I found out a few years ago, is that mobile operating systems are very hostile to ⁓ programs just running in the background. Yeah, exactly. And so it’s like, it’s almost it’s funny because that’s not like any kind of fundamental limitation. It’s just sort of the way the platform operates and

Stephan Livera (1:10:08)
Right, the battery saving stuff, yeah.

James O’Beirne (1:10:19)
So I think that’s probably where most of your your trouble would be ironically if you were willing to pick a kind of aggressive assuming TXO height

Stephan Livera (1:10:20)
Yeah.

Yeah, interesting. it may just be, I mean, that’s probably why most of the mobile wallets that I know are just kind of the Electrum server model. So they’re just calling out and saying, you know, feed me my balances.

And maybe there’s a reason for that, right? In one sense, yes, we can say there’s not a lot of, let’s say, we can go out and morally lecture people that you should be self-sovereign. The finger wagging is not really going to do it for most people, right? They’re not going to care. They’re just going to do whatever is practical and cheap. Or another example would be like, people love to complain about economy class long haul flights, but guess what? People pay for it. People do it because they would rather not pay. And I don’t blame them, but.

James O’Beirne (1:10:56)
Mm-hmm.

Yeah, yeah.

Yeah.

Stephan Livera (1:11:11)
It’s, know, people aren’t necessarily gonna admit these things, but that might be a big driver for them. Like just the price, it’s more expensive to fly a business class or whatever. ⁓ And can we blame them for that? Like, and so guess my point is like with the phone thing, yeah, we can say, well, Apple and Android are really aggressive about battery saving, but could that be because the end users want full day battery life? So is it really, you know?

James O’Beirne (1:11:33)
Yeah, yeah. The average case for that

is better. Yeah, it’s like that’s a rational decision for them to be making. It’s a good feature probably. Most apps shouldn’t be running in the background, ⁓ you know, unmitigated. So, no, I totally agree. I think the sad truth that I’m waking up to is that for that kind of fundamental shift in self-responsibility to happen in any appreciable amount.

it’s gonna take something really catastrophic, Like, ⁓ you know, a major exchange is gonna have to go under or, know, people need to feel pain to, and that too is a rational adaptation, right? Like, I mean, like, not everybody can be like you and me ⁓ in that, you know, we’re like thinking about these eventualities and like kind of preparing and like, you know, spending a bunch of time, you know, worrying about our, say, privacy footprint or whatever, like.

Most people don’t have time for that. just like they have to take out the trash and go to their job and make money and feed their kids and whatever else. ⁓ And so for people to actually rationally reallocate attention to something like Bitcoin custody or even paying attention to Bitcoin. mean, people are incentivized to pay attention to Bitcoin because their money is obviously evaporating in front of their eyes. So was like, OK, well, this is this this might bail me out of that very thorny problem.

But in terms of actually taking self custody, it’s like there’s no hair on fire thing in recent memory for most people. Now you and I, you know, remember when Mt. Gox went under, we watched BlockFi go under. Events like that, you know, should make an impression on people, but ⁓ I think it’s gonna really have to be something ⁓ big and scary and probably painful to kind of like shake people to their core.

Stephan Livera (1:13:20)
Well, I mean, nowadays

it would have to be something went wrong with Coinbase custody, right? I mean, or like a BitGo or, you know, one of these like massive custodians. ⁓

James O’Beirne (1:13:24)
Bye.

Yeah. Yeah. And what

would that do, man? What do you think would happen? Like, if Coinbase custody goes under, do you think that the economic world is just going to kind of let that fly like Dow style?

Stephan Livera (1:13:39)
Well, mean, Bitcoin would

survive, but yeah, would be a pretty, you know, disruption.

James O’Beirne (1:13:44)
I, yeah.

I think there would be some very uncomfortable proposals put forth if something like that happened. ⁓

Stephan Livera (1:13:54)
You reckon

they would try to roll back or try to, you know.

James O’Beirne (1:13:57)
I

could believe that. I wouldn’t even know who would propose it necessarily, but I could absolutely see people saying, we can’t let this, like, and maybe, I don’t know, maybe, yeah.

Stephan Livera (1:14:06)
Yeah.

But remember, there was some

CZ or someone floated the idea to, ⁓ for Binance, right? When Binance had something and ⁓ someone, I can’t remember who.

James O’Beirne (1:14:19)
Didn’t Jer- I thought Jeremy

proposed some kind of like very crazy scheme for like a reorg attack that would, you know, incentivize miners to org out the bad blockers. Something like that. I seem to recall there being some kind of crazy scheme, scheme pitch along those lines.

Stephan Livera (1:14:38)
But I just I mean to be honest, I think it’s unlikely to happen But I mean who knows but I guess on the positive side though. I mean look Bitcoin adoption is growing. Yes I mean we’re down from the high right now So, you know people are kind of scared the fear and greed index is sort of at a low point now and people are kind of a bit bearish ⁓ You know, I think people just have to like learn to zoom out and understand like I mean here’s how I’m thinking about it and you know, I’ll get your response on that but the way I’m seeing it is

James O’Beirne (1:14:41)
Yeah. Yeah. Yeah. Yeah.

Stephan Livera (1:15:07)
look, we’re gonna get a lot of the things we wanted, not all of them, and not when we thought it would happen. When I was a noob, I thought it was all gonna happen so quickly, and obviously now, 13 years on or whatever, it’s a long process. And nowadays I’m kinda looking at power law in terms of adoption rates and things like that. And so I’m just kind of, I guess, resigned to that aspect of it, but.

I do think Bitcoin is going to change the world. think adoption is going to carry on. It might be a few more decades, maybe it’s two more decades or whatever in that range. And it will have a real impact on economics and the world, even if it’s not the ideal vision.

And I’m still hopeful that we get some kind of a CTV or something like that. And maybe not right now, maybe people ⁓ will eventually come around to that. So I guess that’s where I’m at. Where do you wanna land and have any closing thoughts?

James O’Beirne (1:16:11)
I’m happy to say I agree with all that. ⁓ Although I’ll put my signature pessimistic bent on it. ⁓ Look, fiat debasement is realer than real. Everybody knows that. Everybody knows that. ⁓ Gold and silver have seen their day in the sun. ⁓ But as someone who’s like reevaluating those things periodically after being a Bitcoiner, ⁓

You just can’t look at them seriously in the same way. It’s like, okay, you know, can China like fabricate gold? How do I know this thing is real? I don’t have a mass spectrometer. If I want to sell it, I’m going to take some huge spread right on the chin and, you know, trust UPS with however much money. It’s like, it’s just a nightmare. And ⁓ there really isn’t any alternative to Bitcoin. ⁓ It’s really the only halfway plausible way out.

of this like storm that’s headed for us. And so I still really appreciate it in that sense. You know, ⁓ I think it’s it’s still working for that use, even not like in short term literal terms. ⁓ It’s still from a technological standpoint, kind of the only plausible thing that can, you know, operate on the other side of whatever’s coming unless. You know, unless, yeah, we really do revert to some kind of bizarre gold standard where

sure if you’re being inflated ⁓ and, you know, countries are like carting around, you know, giant planes of gold to do net settlements and stuff. mean, maybe that happens, but by comparison, Bitcoin just seems to be such a better medium for that kind of value transmission. ⁓ So it’s still a beautiful thing. It’s still absolutely ⁓ one of the coolest things to work on. ⁓

Yeah, it’s tough as an engineer when you kind of see what the possibilities are and what the things you could be getting are, but you’re not getting. ⁓ Trust me, you want some level of disgruntlement among your engineers so that they’re working to get more for you.

Stephan Livera (1:18:24)
Well, yeah, let’s leave it there. So yeah, we covered a few things Obviously the quantum bearish nurse as well as you know Bitcoin protocol aspects listeners. Make sure you follow James online. You can find him at James OB And of course the website ctv-csfs.com James thanks for joining me today

James O’Beirne (1:18:46)
Thanks for having me on, Stephan. Always a pleasure.