 Christopher asks, why were 21 million bitcoin identified as the maximum supply number? Can this amount be changed since the code is open source? Or are they somehow inherently embedded into the code in a way that it cannot be modified? This is really a great question, and it's one of the things that started my journey into the source code of bitcoin back in 2012. I had heard, like most people who are new to this space, that there can never be 21 million bitcoin. I was wondering if I would be able to open the code and find somewhere this number, 21 million, and then, based on this number, that I would be able to find somewhere in the code, like an if statement that said, if the next block is greater than 21 million, or if the sum of all of the rewards greater than 21 million bitcoin stops issuing bitcoin, or something like that. But that's not how it works. Interestingly enough, the 21 million bitcoin, although it exists as a constant in the code, is not checked on a per block basis, as you would expect. It's an emergent number that emerges from the way that the reward or block subsidy is issued. So, how did we arrive at 21 million bitcoin? Some people think it's actually a mistake, that it was by accident, and the desired number was 42 million bitcoin. Not that it really makes any difference, because the number of bitcoin that's going to be issued is an arbitrary number. It doesn't really matter. As a unit of measurement, it doesn't matter. In the code, what you will see is the block subsidy being checked with every block, by every participant in the network. What the rule is, is to look at how many blocks have lapsed since the Genesis block, divide that by 210,000 blocks, and therefore calculate what the current block subsidy should be, based on the number of halvings that have happened. So, we start with 50 bitcoin per block as subsidy, and then after the first 210,000 blocks, there is halving, which is when block subsidy is divided by two. And then you have 25 bitcoin issued per block as the block subsidy, and then 210,000 blocks later, or approximately four years later. There's another block halving, and it goes down to 12.5 bitcoin, which is where we are today. In 2020, in the summertime or springtime, there is going to be another block halving, which will be the third block halving, sorry, the fourth block halving. And we're going to go down to six and a quarter bitcoin per block. Now, it's embedded in the code in that it is checked for each block, in that every node checks that the reward or block subsidy that exists in a block cannot exceed the expected current block subsidy based on how many blocks have lapsed. So, right now, every participant in the network checks every block and expects to see 12.5 bitcoin and block subsidy plus fees as the maximum number that's in the coin base of a block. And if it is greater than that, the block is invalid and is rejected. So, you can't create more bitcoin because everybody who receives a block will check that you haven't exceeded the block subsidy amount plus the amount of transaction fees. These consensus rules being enforced by everyone who's participating, not just miners, is what keeps everyone honest and prevents the miners from inflating the bitcoin supply. So, you could change it in the code and if you change it in the code, then your node that's running that code would evaluate things differently. And as a result, your node would accept perhaps a block that had greater than 12.5 bitcoin subsidy today. And that block would not be accepted by anyone else at which point your node and only your node would go out of consensus with the rest of the network and effectively be kicked off the network because as soon as it tried to propagate this invalid block, it would get kicked off the network or banned. This happens, it's a mechanism to check. So, you can change the code in your node and that will result in your node getting kicked off the network because it's not following the same rules. Theoretically, the developers could change the code for the bitcoin core software or perhaps many of the other clients that run on the bitcoin network and then whoever was running that new code would effectively get forked off the network. So, if you imagine some miners downloading that code and some miners refusing to download that code that changed the block subsidy, it would result in a fork and a new bitcoin network would be created that would not have a 21 million supply and I guess most bitcoiners would refuse to upgrade the code to make that change because it would violate the monetary principles of bitcoin. As a result, you'd have another fork of bitcoin. Let's call this bitcoin inflation and bitcoin inflation would have a supply greater than 21 million and would it be bitcoin? According to most bitcoiners, no, but you get to choose the rules you run on your node. Developers can write code, but nobody has to adopt it. So, the way the consensus rules work is this balance of power where every node that runs on the network, whether it's mining or not, validates all of the rules and it validates all of the rules by running the software and depending on what software you're running, you are either following the same rules as everybody else or not and if you're not following the same rules, you're running your own version of bitcoin and you will effectively fork the network. If you fork it alone, you're just basically going to get banned. If you fork it with a sizable number of miners so that the forked chain can continue, you create yet another bitcoin fork. There's 45 of those already. Most of them are relevant and have almost no value and as a result, it doesn't matter if people do this. Ian asks, Bitcoin caveats regarding total supply. You've all heard the phrase, there will only be 21 million bitcoin. In fact, the correct phrase is, there will always be less than 21 million bitcoin because we never mint 21 million. In fact, the amount minted is 20,999,999 bitcoin, 997 and change. In fact, 21 never gets reached even if all of the bitcoin amined and none of them are lost. As Ian asks, considering all bitcoins lost, destroyed, or burned, can we extrapolate that the amount of bitcoin in circulation will never actually even approach 21 million? Ding ding, absolutely right. You got it. In fact, we can estimate that already at least a million bitcoin have either been irrevocably lost through loss of private keys, burned to public burn addresses. What a burn address is, is an address that is obviously generated without private key in such a way that you know that it is impossible to find the corresponding private key and no one has it. We can talk about that in the subsequent section, how burn address is created. The bottom line is, it's an address that is known to be without a private key. Therefore, when you send money to that address, it's unspendable. Several other mistakes have caused unspendable bitcoin. There's also a little quirk of the protocol, which is that a miner doesn't have to take the full reward, meaning right now, the block subsidy is 12.5 bitcoin. They don't have to take the whole 12.5. The rules actually check to see that coinbase is less than or equal 12.5 bitcoin plus fees. It doesn't check how much less. You take six bitcoin and not take the other six. You could just put a block subsidy of six bitcoin in your block subsidy, in your coinbase transaction today as a miner. What would happen then? Well, you'd mine six bitcoin instead of 12.5. The other six and a half can never be mined again. You lost your chance. That was the block. It's gone. People will see that as a valid block. They will mine it. They will accept it in the blockchain. That six and a half bitcoin will never be mined. Effectively, now you've gone from 20,999,999 bitcoin to 20,999,993 bitcoin because you just threw away six bitcoin as the maximum supply has now been reduced. Because of some mistakes that have happened in coinbase transactions that went to unspendable public keys and weren't quite the correct amount, etc., we're less than the subsidy. We already know that this has happened. It's going to be less than 21 million and perhaps even significantly less. Theoretically, by the time we get to 2140, maybe half of all bitcoin has been lost. At that point, you've only got 10 million in circulation. Does it matter? No. It makes no difference. The number itself, 21 million versus 10 million, means nothing. The purpose here is to ensure that from a monetary perspective, in terms of pure economics, the amount of currency is ultimately capped, creating a hard asset, a form of sound money that cannot be issued no matter what the price on offer is. You cannot increase the issuance. In the physical world, no such thing exists. Even gold, which is extremely scarce, if you increase the price of gold enough, it becomes profitable to mine gold in ever further and more expensive places, including asteroids. So the amount of gold in the universe is absolutely not capped, but the amount of bitcoin is. And it's the only asset that is completely capped. And as a result, it has this unique digital scarcity that is qualitatively different from any other form of scarcity. But whether it's 21 or 10 or 5 million bitcoin doesn't matter. You may think, of course, that, well, if there's only 10 million bitcoin left at the end of all of this, after all of the issuance is done, how is that enough to run a larger bitcoin economy? Well, the trick is that, in fact, what can happen in bitcoin is a further subdivision of the smallest unit. So we could start it dealing in units smaller than a satoshi, a thousandth of a satoshi, even a millionth of a satoshi. You just add more decimal places. You might think that's inflation. It actually isn't. If you think about it, if you subdivide units and add more granularity, it still doesn't change the total amount that is available. It just means you can divide it more. So, for example, if I say I'm 170 centimeters in height, if instead I say I'm 1,700 millimeters in height, that's not changing my height. I'm still the same height. I'm just measuring it in finer granularity. And so adding decimal points to bitcoin doesn't change the total supply. That's a great question, and it's one that comes up fairly often. But I like that twist in this particular case from Ian.