 BTC voluntary burning, accidental loss affects over the long term. PU asks, could voluntary burning of Bitcoin inadvertently harm Bitcoin? Users or services might choose to voluntarily burn some of their Bitcoin as a deflationary gift to all other Bitcoin holders. I've even heard people say that when they die, they will not provide keys for all of their Bitcoin to their family. Instead, they will effectively burn some BTC by not providing the keys. This would have the deflationary effect of boosting price, but could it also hurt Bitcoin over the long term? Any serious side effects of burning Bitcoin and speeding up deflation? Alright, first of all, whoever said they will not provide the keys their Bitcoin to their family upon their death has to win the Montgomery C. Burns Car Mudgeon of the Year award. I mean, seriously, supporting the Bitcoin community and wanting to give a deflationary gift, sure. But burning your Bitcoin rather than giving it to your family, either you have a really terrible family. But even so, friends, charities, nothing, really? You're just going to burn it all? So, after that comment, let's actually answer the question. There isn't any serious side effect or risk to destroying Bitcoin, or losing Bitcoin, and creating more Bitcoin deflation for the rest of the holders, or owners of Bitcoin. One of the things that people might think is, okay, but what if we really start running out of Bitcoin? What if we lose out of the 21 million, 20 million of it? Does that then mean that we don't have enough to do transactions? Well, that's a bit equivalent to the other question I get quite often, which is, Bitcoin is only divisible to eight decimal places. What if we run out of granularity? What if a Satoshi ends up being too expensive to do anything with, and we need to have even smaller units? Both of those questions are related. Basically, let's say we lost 20.5 million Bitcoin, and all that was left was half a million Bitcoin for the entire world. What would we do then? And the answer is really simple. We'd start working on subdivisions of Satoshis. We'd start increasing the amount of decimal points that we use for Bitcoin, so that we would subdivide the existing pool of Bitcoin, to even finer units, smaller than a Satoshi. We'd go sub-Satoshi. In fact, this is already happening. The Lightning Network doesn't operate on Satoshis. It operates on milli-Satoshis. How is that possible, you might say? Well, by ensuring that settlement is rounded up, Lightning Network can operate on milli-Satoshis internally, while settling transactions at a Satoshi level on final settlement. That gives the aggregation of milli-Satoshis into Satoshis, gives the Lightning Network a lot of flexibility. This is not the last time this is going to happen. In fact, we are going to see sub-Satoshi units being developed. There are a number of interesting tricks to do that, including some techniques developed by Thaddeus Drijah, Tadge Drijah, who is one of the co-authors of the Lightning Network, a white paper back in 2014, and who has made some incredibly insightful and creative contributions to the state of the art in cryptography and cryptocurrencies. One of the things he developed as a concept, and I hope I am attributing it correctly, is a way to do probabilistic subdivision of Satoshis. If you are interested, look up Thaddeus Drijah, Tadge Drijah, and sub-Satoshi amounts. Very interesting presentation on that. Bottom line, we can go to smaller units. It doesn't matter how much of Bitcoin is lost or burned, it won't have any effect. It is truly a deflationary gift to the rest of the Bitcoin economy. It doesn't matter. If you lose your Bitcoin, everybody gains a bit of value on theirs. I hope it doesn't happen to you, but if it does, you're not hurting anybody else other than yourself. In the case of the Montgomery C. Burns winner of Car Manchin of the Year, you're also hurting your family, because you decided to burn your Bitcoin. Anita asked, sometimes people ask me, why is the limit 21 million Bitcoin? I couldn't find any information why Satoshi Nakamoto has chosen this number. Why not 25 or 18 or something else? Do you know? Someone else added, SJ added, Y8 decimals. The number 21 million is really quite arbitrary. It doesn't have any special meaning, as far as I know. I think it has to do with the fact that if you take a geometric reduction, as is the algorithm that calculates the issuance of Bitcoin, where you have the amount that's issued reduced by half at a set schedule, and you have blocks coming out at a reasonable rate that doesn't create too many orphan blocks, but is fast enough that you can get settlement within an hour, which is every 10 minutes. That's a reasonable first approximation for a block time. The initial reward for Bitcoin was 50 coins every 10 minutes, a nice round number. The amount reward halves every four years, again, a nice round number. If you work that out, that geometric progression produces 20,999,997 points. Just short of 21 million. So 21 million isn't really accurate. It's not 21 million, it's 20,999,997. If all of the numbers I just mentioned, 10 minutes, four years, 50 Bitcoin in the initial reward, the one number that isn't a nice round number is the 21 million. It's actually 20,999,000 in change. So it's almost 21 million, which makes me think that it wasn't the 21 million that was chosen. It was the 50 Bitcoin as the initial reward, 8 decimal places, which gives you 100 million. Again, it's a nice round number, and it gives you enough decimal places to operate, but also fits very nicely in integer arithmetic, so 8 decimal places works quite well. It's easy to do the math in most computer libraries. Finally, every ten minutes blocks. If you make blocks faster than ten minutes, you have a problem of having more orphaned blocks, especially if you have a large decentralized network. If you make it slower, then you don't get enough confirmation in an hour to be certain of your transactions. It seems to me that the 50, the ten-minute, and the four-year halving were the numbers that were chosen, and 21 million wasn't really chosen. Captain Hadakall asks, Do you believe that the number of Satoshis in a fully mined 21 million Bitcoin network will be sufficient to support a scaled number of users and layered applications? Meaning, as the price of Bitcoin increases, will network affect demand? Will the smallest unit be priced out of practical usage? Could you discuss the implications of a decimal place increase? Captain Hadakall has inflation in his parenthesis. Let's discuss this. First of all, I don't think that 8 decimal places is enough. I think we are going to see an increase in the number of decimal places. However, that isn't inflation. Changing the divisibility of your currency isn't inflation, because it doesn't increase the total amount of currency in circulation. All it does is change the units of the currency you already hold in your hand. They are still just as rare. There are just smaller divisions of them. That is a very subtle difference, but it is not actually inflation. You are not issuing more units. These don't get added to the circulation of the currency. All you are doing is cutting them into finer pieces. In practical terms, it is possible to change the smallest units. The most obvious way is to do a hard fork. A hard fork being a discontinuous, not backwards-compatible change to the consensus rules. The structure of a transaction is changed. The structure of a transaction is changed to change the size of the value field in a transaction. Change the unit from a setoshi to, let's say, a millionth of a setoshi. We add nine more decimal places. We subdivide a setoshi, and we have a millionth of a setoshi. To add six more decimal places and correct all the software to recognize that these are millions of a setoshi, would take a change that is not backwards-compatible. That is a hard fork. If you had enough consensus to do that, that is one way to do it. There are actually some possible ways to do it with a soft fork that has been proposed. One of the interesting things to notice is that second-layer networks, for example, Lightning Network, use sub-satoshi units. Lightning Network operates on milli-satoshis, which is thousands of a setoshi. What happens there is people are exchanging thousands of a setoshi, but they can only settle and close a channel at a setoshi level. If the balance of a channel is less than a setoshi, then they can't settle that. Basically, the Lightning Network and second-layer networks like that give you the ability to increase the number of decimals and do smaller and smaller transactions. As long as, when you batch them all together and settle, you settle with the smallest unit being a setoshi. You can actually fix this problem through the second-layer, but you can also fix it with a hard fork, or possibly some strange way of doing a soft fork that changes the transaction structure in a compatible way. Perhaps by introducing a new transaction version number. Hard to say.