Operation Saylor - Episode 17/120

Hi again and welcome to another episode of the Operation Saylor. This is update number 17, corresponding to November 2023.
If you are reading this for first time, you might want to check Episode 1, where my plan and details are explained. That will get you in context.


  • BTC stack: 1.35916794 BTC
  • € stack: 40.80 €
  • Current total value in €: 46,253.50 €
  • € into BTC: 30,000 €
  • Paid back to bank: 5,859.20 €
  • Outstanding debt + interests: 38.085,13 €
  • Installments to go: 104



Hi again and welcome to another episode of Operation Saylor. This month, I've decided to invest this small space to present some ideas that have been flying in my mind recently around the scaling problem, custodianship and the future of banking. I expect this episode to trigger some people's feelings (I hope you don't choke me @Darthcoin) and, hopefully, lead to some interesting debate.
Let's begin by picturing the final station of the journey we are on: a world where Bitcoin has won the monetary competition wars and has become the standard asset for the entire world. All of humanity saves in Bitcoin and transacts in it. Prices are set in sats, payslips flow in sats, life savings get stored in sats. We made it.
Those who are technical enough know that, unless some amazing breakthrough comes along, Bitcoin's layer 1 can't sustain this on its own. If billions of people tried to perform dozens of transactions daily, the mempool would become a massive bottleneck, and small transactions (small possibly being anything under 6 figures) would simply never make it. And let me stop here and say that this idea of a dozen transactions per person and day is outdated and comes from a Fiat mindset. As Antonopoulos anticipated, we are heading to a future where we transact more and more frequently, to the point of doing so continuously in mathematical terms. Think, streaming sats.
Anyway, I'm derailing. Mempool can't sustain all the transactions in the world. We have a scaling problem.
Now, a scaling problem is not the end of the world. It's just a problem that needs to be solved.
There are several solutions being developed out there to tackle this problem. Lightning is, of course, the best-known one. And I would personally say, the most promising one. Liquid is another option, way less popular than Lightning, but nevertheless a working one. I am aware of the existence of some other ideas, such as Ark, which are in a less mature phase and may or may not someday turn into serious contenders. Since there is no such thing as a free lunch in the engineering design of money and payment systems, all of these different solutions come with their own unique set of trade-offs. You get some advantages by using them (smaller fees, faster settlement, etc) but you have to pay a certain price (locking in funds, trusting a federation, going from a serverless wallet to one that requires a 24/7 server to work, etc). Ultimately, these different sets of trade-offs will compete in the market and gain a certain market share. Survival of the fittest will determine what works and what doesn't.
Now, let's take a pause from my rant and let me poke your brain with a riddle. I have intentionally left a way to transfer Bitcoin out of my explanation so far. I would like you to try to guess what it is before we continue. Let me give you some clues: it usually provides instant settlement. The fees are low or negligible. It is the oldest scaling solution that exists for Bitcoin. And I would bet my ass it moves several orders of magnitude more volume than all the other scaling solutions as of today.
Can you guess it? I'm talking about custodians.
They come in many different shapes and sizes. Some of them are exchanges (Coinbase, Kraken, Binance). Some of them are wallets (Wallet of Satoshi, Blink, LightningTipBot). Some are managed by massive, regulated entities based in countries with sophisticated legal systems. Some are being run on a crappy Raspberry Pi stored in the broom closet of some cypherpunk PhD student. But they generally have one thing in common: given that they are just systems of IOUs, if both Alice and Bob have an account with the same custodian, they can transfer Bitcoin to each other instantly and without fees. As more people join the custodian, more transactions can be done easily. It scales like a charm.
Some people have just read the previous paragraph and are already cracking their knuckles, ready to point out that that's not really transferring Bitcoin, strictly speaking. And that Alice and Bitcoin don't even really own any Bitcoin at the custodian. Not your keys, not your coins, right?. Trust me, I'm with you, but let's see if we can compromise and at least agree on this: custodians are a valid storage and payment system as long as they NEVER fail to redeem their IOUs. If you don't agree on this, you probably shouldn't waste your time reading further because you'll think that I'm talking nonsense. Now let me detour for a bit again and we will come back to this point.
The custodian scaling solution is, by far, the oldest one in Bitcoin's history. Mt. Gox was there way before Lightning was even an idea. Actually, the custodian scaling solution is not even a Bitcoin thing. If you've done a bit of reading about the history of money from authors like Saifedean or Lyn, you already know that the grandpa of Fiat money is gold receipts. Why did those exist? Because gold was hard and expensive to transact with and we couldn't possibly get all the transactions that we needed done by having everyone move around with gold coins and bars. The parallelism with the inability of Bitcoin's base layer to handle all the transactions the world needs to do is funny, isn't it?
Now, if we are to be humble and level-headed, I think we should acknowledge that the systems of custodian's IOUs built on top of gold brought a lot of great things. Much trade was done, and much progress was obtained through it. It would have been difficult for merchants that were thousands of kilometers apart to constantly settle between each other with physical gold transfers. Well, if they had had to settle that way, they probably wouldn't have traded at all in the first place. Gold IOUs made this possible, and thus I feel forced to concede that these systems brought good things with them. As we already know, it also brought terrible things down the line.
And this is where we can grab the Bitcoiner machine gun and shit on banks and governments: gold IOUs don't work because they are inevitably captured and corrupted. The physical centralization of gold in the vaults of a few banks makes it easy for a powerful entity to collude with the banks and take control of the system. Make IOUs slightly harder to redeem over time, bring some regulations that increase friction, add a few sparks of propaganda, and voilà, after a few decades, people will become dumb fucks that use the IOUs even AFTER you have openly stated that you are not going to redeem them for gold anymore. You now live in the Fiat world. I'm simplifying things a ton here, and if Lyn Alden would read this, she would probably have a seizure, but I think we can agree that gold custodians didn't turn out to be a good storage and payment system because they ALWAYS end up failing to redeem their IOUs, mainly due to centralization and perverse incentives.
Now that I've built up all the background and context that I wanted, we reach the interesting part. We have agreed on the idea that custodians are awesome if, and only if, they don't betray their promises. But history teaches us they tend to do so. So the question becomes: what's going to be the custodian's role in Bitcoin's scaling problem? Are they going to be a valid solution? Are they going to disappear into oblivion? Are they going to bring all the evils that came with gold custodians? Are they going to succeed massively and kill all other scaling solutions?
I'll take a pause here and clarify that I'm very humble about what comes from this point on. I don't know shit about fuck, and I'm sure the future will surprise all of us. Yet I still want to present my vision because I think people discard the custodian scaling solution way too often, mostly based on dogma and emotional judgments. In any case, and to make things a bit spicy, I am going to give you my hypothesis before I build them up: I believe custodians will play a great role in a hyperbitcoinized future, I believe they will contribute greatly to Bitcoin's success, and I feel dunking on any custodian just because it's a custodian is pretty stupid and doesn't help us move towards hyperbitcoinization.
If we were to assume that Bitcoin custodians are going to be just like gold custodians were, then the answer is clear: we better steer away from them because we know how the story ends, and it's not pretty. But I think that would be a pretty dumb assumption, since Bitcoin and gold are very different. Let me point out a few relevant differences between them:
  • Handling large amounts of Bitcoin does not require government's consent. Handling large amounts of gold without government meddling is impossible.
  • Large amounts of Bitcoin can be settled in a few minutes at almost no cost. Gold settlement (physical settlement) scales awfully, and probably the other way around: the more you want to move, the more slow and expensive things get.
  • Control over Bitcoin, if the owner consents, can be audited in real-time and pretty much for free. This is not the case at all with gold.
  • Hand- to-mouth users might find it impossible to transact in the future. Onchain will be prohibitively expensive, and sovereign usage of lighting is beyond the ability of 99.99% of people.
  • Ecash allows for custodians that provide extreme privacy while taking care of client funds. People under a custodian's ecash system can enjoy ownership and transactions with other clients of the custodian with perfect privacy. Achieving the same level of privacy with gold would be extremely hard and crazy expensive.
These differences between gold and Bitcoin, along with others I haven't listed, directly affect what custodianship could look like in the Bitcoin world. Here are a few observations on the differences I can see between gold custodians + IOUs and Bitcoin custodians + ecash.
  • No regulatory barriers to entry. Anyone can become a custodian. It wouldn't be a trivial thing to do, but it would be way easier than trying to start a bank in, say, the US, nowadays. This is because custodians could easily move to any friendly jurisdiction. Or just not give a fuck at all and operate fully in cyberspace.
  • Bitcoin can move fast. Really fast. This has some consequences:
    • The magnitude and speed of bank runs is a million times more severe than in the gold standard era. Custodians live under the threat of this happening at any moment. The slightest doubt over a custodian's reliability could cause one.
    • Users could spread their liquid funds across multiple custodians and shuffle them around easily. We could even reach a point where you use a simple wallet with a single balance, but the funds are actually distributed across multiple custodians simultaneously. Balances across custodians could be shifted at any time to group funds to make a large payment, to avoid some specific custodian or increase funds in another, etc.
  • Federated custodians are a thing. Trust can be placed in multiple, unlikely-to-collude parties.
  • Proof of assets is doable, and it seems like even proof of liabilities could be a thing at some point (https://gist.github.com/callebtc/ed5228d1d8cbaade0104db5d1cf63939). This means that the chances of a custodian successfully engaging in fractional reserves would be dramatically reduced. And that rugpulls would have to be complete and immediate, or we would go back to instantaneous bank runs.
  • Self-custody of Bitcoin by normal users would still be somewhat of an option when compared to the gold scenario. This is another looming threat to custodians individually and also as an industry: behave, or I'll take funds back to my wallet. A constant competitor of sorts, a bar that any custodian needs to overcome just to be considered. In today's banking and government situation, given that banking is an oligopoly that you can hardly escape from, it doesn't matter if a bank fucks you hard in the ass. Your only option is to go to another bank that's going to do the same thing.
All of this leads me to imagine that, in the future, we will have a healthy market for Bitcoin custody. A market with a lot of different custodians competing, where customers have access to excellent offerings and great prices. A market where trust will be betrayed infrequently, for doing so will be way more expensive than behaving. We agreed early on that custodians are a valid storage and payment system as long as they NEVER fail to redeem their IOUs. I think this will be mostly the case in the future. A lot of people will choose to use them, and a lot of transaction volume will happen within their systems. They will be another tool for scaling Bitcoin, just like Lightning or Liquid. And hopefully, they will bring more good than evil.
I don't think custodians will be perfect. Humans are evil and stupid, and mistakes will be made. We will also probably face cycles of centralization and decentralization: there are incentives for custodians to grow larger and larger. And at a certain size, bad things will probably happen, and we will then face trends of decentralization. An ever-going cycle of custodian boom and bust. Ultimately, issues will never disappear completely. But, if custodians do their job properly 99.99% of the time, perhaps the net impact they make on society will be massively positive. Just as we can acknowledge that gold standard custodians didn't only bring negative things but also positive ones, perhaps we can also imagine that this will be the case with Bitcoin custodians. And perhaps the balance between good and bad will be much better than with their gold ancestors.
I'll close by repeating my hypothesis again: I believe custodians will play a great role in a hyperbitcoinized future, I believe they will contribute greatly to Bitcoin's success, and I feel dunking on any custodian just because it's a custodian is pretty stupid and doesn't help us move towards hyperbitcoinization.
Finally, I believe we should always leave the door open for unexpected technical breakthroughs to come along and change things completely. I don't think many people in 2012 could have foreseen the coming of Lightning and just how positive and impactful it would be. Maybe some new technology comes around in a few years and provides us with better, more sovereign ways of transacting Bitcoin fast, safely, and cheaply.
Anyway, that was quite a rant. I hope you found it interesting. I'll be very curious to hear your thoughts. As always, thanks for reading and see you around next month.

Previous episodes

Beautiful analysis, well-articulated. The sense of what it means to be a 'custodian' is going to permute in a million ways, using all the entities you listed already, and more that will be invented. The gradient of custodianship and sovereignty will be broad and complicated. There's no other way, and thinking it through is useful.
I hope you don't choke me @Darthcoin
Why should I choke you? I like your "Operation Saylor" series. I even share it with others, to read and learn from your journey / experience. Keep'em coming !
You talk about custodians. What about "uncle Jim" solution? Running a bunch of nodes in a small community can help a lot those that could not run their own nodes / services. And can have quite good trust.
I think more node runners should think about this scenario, as I explained here: https://darthcoin.substack.com/p/bitcoin-private-banks-over-lightning
I don't think many people in 2012 could have foreseen the coming of Lightning
Is true, even until 2016-2017 LN was just a dream. Were some discussions about but nobody had an idea how to do it. Nowadays we are quite far from those times, but still we have a long way into this jorurney. LN is still veeery young.
I agree custodians will and should come in all shapes and colours. If you think about it, stacker.news is a custodian itself, and we all save a lot on fees when zapping each other's posts here because of this design.
I also hope we get community custodians. A family could have it's own cashu mint. A neighbourhood. A football team. An entire city.
Every day. We stray further from God.
Jokes aside, I haven't lost faith in a self-sovereign techno-utopia yet. Being able to transact offchain was critical for that, and we made it possible once we fixed transaction malleability in 2017.
Lightning will definitely not cut it at global scale, but I want to believe that it's just the "0 to 1" technology, and that some people will find incrementally smarter and more robust ways to transact offchain, with better tradeoffs.
Besides that, as you said custodians have been here since the beginning and will probably be a mainstay of Bitcoin forever. Not that I like it, but it is what it is.
I don't think what you've written is controversial. There will be all manner of ways to perform Bitcoin transactions off-chain that make different trade-offs between trust, cost, and ease of use.
Lightning represents an extreme: Minimal trust at the expense of a cumbersome experience (hot wallets, channels, liquidity, force closes, etc.) It's a powerful tool for those who want full sovereignty over their funds, but the idea that everyday people will be transacting on Lightning seems unlikely.
I think the controversial bit of my position is my suggestion on not shitting on custodians and, potentially, embracing them. Well, embracing might not be the right word. More like use them with the right amount of skepticism, coupled with the right amount of pragmatism.
What I generally dislike is the frequent opinion of any custodian being the devil in any situation. For instance, I see bitcoiners that regularly criticize Wallet of Satoshi simply for being custodial. I don't think that's healthy for everyone. And I think these are the same people that wouldn't like my stance.
Thanks for all the thought you put into this :) you highlighted a lot of great stuff..
ultimately i agree wholeheartedly with everything you've said about the custodian role, but i am curious, do you see custodians ever falling out of the picture entirely? Or only if we hit that L1 scaling eureka moment?
I don't think they would ever disappear completely. In highly trusted environments, just a tad of convenience is enough to justify using them.
The most basic custodianship model is the debtor. I lend you a 100 bucks because I trust you, and you will pay them back to me one of these days. You are kind of acting as a custodian.
Now create a social cluster with a lot of trust and probably some person might arise as the central hub that tends to manage debts and accounts. Perhaps it could be the most popular merchant in the local bazaar.
This naturally lends to creating custodianship relations. And I guess that, in trusted social circles where the risk of rug pulls is negligible, it's just fine. Even if they could settle easily. I mean, I don't need to settle with grandma every time we have a financial interaction.
Wow this is great, thanks for sharing!
Well I’ll add fedimint and cashu as custodian solution evolution for communities! Another WIP is Valera and there gonna be more coming
PS: it’s nice to see that blue line crossing the orange just after a year 1/2, congrats proving fiat system wrong
I think ecash and federated custody will both be very important in bringing over safe and privacy respecting custodians.
this is awesome keep up the good work
Thank you so much @pillar for these consistent monthly updates. You're already winning at this installment #17 (inflation-induced debt destruction and BTC going up forever "Laura." ;-))
deleted by author
Thanks to your for reading and for the kind words.
Well, there's no denying in that seeing the blue line go up is pleasing. But in my view, I'm winning since I bought the bitcoin. Because I'd rather have bitcoin in my pocket and debt than nothing.