Comments by "SAL" (@SAL-fs1mr) on "3Blue1Brown" channel.

  1.  @alronthunderdrummer  he means he likes centrally managed databases. Bitcoin is not an efficient setup but it is meant to achieve decentralization where there is no trusted entity - important for an internet money resistant to politics.
    4
  2. If you want efficiency, stick with centralized systems controlled by the few. Proof of work is what's needed to achieve decentralization.
    4
  3. The network as a whole would reject those blocks for breaking protocol rules. Those miners would essentially be wasting all the expensive proof of work computations they did for nothing.
    4
  4. There were 50 unspendable bitcoins created in the genesis block. Whoever has the associated private key will just be able to prove they were the one to make the Genesis block (assuming no one else obtained the private key. But they won't have power to do anything else, they won't be able to change how bitcoin works.
    3
  5. Bitcoin blocks have a limit to how big they can be, which translates to about 100 gigabytes max growth per year. 1 TB in ten years, at which point it would probably be trivial to manage due to technological advances. But even without such advances managing a few TB is very doable with today's tech.
    3
  6. Google still has to make a quantum computer 100 times bigger than what they have now to be a problem for bitcoin (extremely hard to do), and even before that happens, bitcoin could just upgrade to a quantum resistant algorithm, these schemes already exist today, but not fully standardized yet.
    3
  7.  @34joliez  over 90% of all government or central bank issued currency is already digital (numbers in bank accounts, essentially), less than 10% is physical cash. So digital money is possible already. And also based on history, money doesn't have to be issued by governments or central banks. There were times in history when the most traded commodity served as money (sea shells, beads, gold, silver). So bitcoin is digital commodity that is being adopted as money thru free trade.
    3
  8. How is it a pyramid exactly?
    3
  9. Not only would other miners immediately reject a chain with less POW, but so would all users running nodes on their computers. All network nodes check all POW and verify all the transactions that miners include in blocks. You can do this yourself by running a full bitcoin node on your own computer.
    3
  10. Before transactions are included in a block, they sit in the mempool and every node's mempool will be different based on which transactions they become aware of. Each miner might have slightly different ways of choosing which transactions go into a block they found. Some strictly do it based on the transactions that pay the highest fees, others choose to include transactions based on the order they saw them in the mempool. All blocks follow the same rules, but that doesn't mean each block will be the same.
    3
  11. The problem becomes that the person sending the transaction can still produce ANOTHER valid signature to "double-spend" their money (a conflicting transaction). Proof of work solves this double spend problem WITHOUT a trusted central entity.
    3
  12.  @SimonTiger  the bitcoin software is completely open source!
    3
  13. It "gets in" by someone trading a bitcoin for another currency - a market develops to discover the price of bitcoin.
    3
  14. If you secure the network with hashpower, you get the new coins. How else would you choose to distribute the coins in a DECENTRALIZED manner?
    3
  15. There is such thing as quantum resistant encryption, so most likely bitcoin will just have to upgrade to make it immune to any quantum threats.
    3
  16. Not only that, but fees for blockchain transactions can be expensive for everyday small payments. So while the bitcoin blockchain is being optimized for security and decentralization, there is a second layer payment network being developed called the "bitcoin lightning network" where payments are now instant and fees are extremely small (on the order of 1/100,000,000 of a bitcoin).
    3
  17. There can be brief periods when price goes down, but if you choose any 4 year period, the price has only gone up. Btw, 4 years is the best time frame to examine because bitcoin has a supply rate halving every 4 years.
    3
  18. Bitcoin users are already paying fees now, so it is reasonable to expect fees to remain (or even increase with more demand and usage) once there is no new coins to mine.
    3
  19. All the non-mining full nodes would still check the block and reject it for not following the rules. Anyone can do this validation by running a node on their own laptop.
    3
  20. Linked list made immutable in a trust minimized manner thanks to proof of work computations.
    3
  21. There are many problems that come with having to trust a central source. Maybe they block payments from being able to be posted. Maybe the website or central node gets a shutdown then that's it for the system. That is why Bitcoin is decentralized, all the other attempts at digital cash prior to Bitcoin had this weak point of centralization which led to their demise.
    3
  22. It only seems inefficient if you don't understand what the end goal is. If you want your monetary wealth controlled by others (banks) then bitcoin isn't for you. But if you want full control of your monetary wealth without having to trust others, then bitcoin achieves this for you.
    3
  23. The mining nodes want the reward (new bitcoins and bitcoin transaction fees) for publishing the next block of the blockchain (think of it as page in the blockchain ledger). To ensure that only one such page is published to the network roughly every ten minutes, there is a competition to see who can come up the hash that provides enough leading zeroes. By seeing enough leading zeroes, you can be sure it took a lot of computational effort (this is why it is known as proof of work). This is how transactions are "verified", or in other words, added to the blockchain. If Alice lied and didn't have 1000 to send, then the transaction is completely ignored, because such a transaction would be unable to reference past blockchain transactions (that all nodes have record of all past transactions) showing she had the funds to start out with.
    3
  24.  @Evadelover24431  the mining nodes keep recalculating the hash function until they get a result with enough leading zeroes. The hash function has to use inputs based on parameters of the previous block. Once a result is found and published on the bitcoin network, all users verify the correctness of the result to accept the block.
    3
  25. The computing of a block starts as soon as the previous block is mined. A block is mined once a valid solution to the proof of work mining puzzle is found.
    3
  26. That's one aspect of it. The other characteristics that are more important is securing the blockchain ledger using proof of work mining and for the ledger to be widely distributed all over the world.
    2
  27.  @StefanoTempesta  The node software handles all the connections automatically. This node software is called Bitcoin Core - it is opensource and anyone can run it on their computer. For example, the node I'm running on this computer is connected to 17 other random nodes.
    2
  28. Miners will continue the mining process to earn all the blockchain transaction fees.
    2
  29. It is based on the mining difficulty adjustment algorithm, which determines the number of necessary zeroes based on previous blocks. So basically, all miners and nodes figure it out on their own based on the same blockchain data.
    2
  30.  @victoriam6569  the current size of the blockchain for the last eleven years of Bitcoin's existence is about 310 gigabytes. The good thing about the bitcoin blockchain is that it can only grow a certain amount each year, so it won't grow exponentially in size. It is ruining the idea of trusting others, but at least with bitcoin, you have the choice of whether to trust others or not.
    2
  31. What's "real money" anyway?
    2
  32. It is up to the recipient to wait and make sure that transaction got included into a block. Better yet, the recipient can take the unconfirmed transaction (the transaction not in a block yet) and increase the fee to be more appropriate for inclusion into the next block.
    2
  33.  @64standardtrickyness  it comes down to which branch has the most mining power behind it (assuming the transactions are valid). In the "real world", competing branches can't continue for long as the one with less mining will lose out of all mining rewards (no incentive to build on a minority chain). Again, it is important to stress that miners can't simply do anything they want - they still have to work within the consensus rules as all nodes (whether mining or not) are constantly keeping them in check.
    2
  34.  @64standardtrickyness  if it is indeed a fraudulent block/chain (with an invalid transaction), it simply will not be accepted by the network as they check it in real time (nothing to worry about). However, if it is a competing chain with double spent coins ( coins being spent to 2 different addresses), then all you have to do is sit back and wait for a few more blocks to see which branch the majority of miners are building on (that becomes the true blockchain, since it will gather the most proof of work). That is why it is good practice to wait for 3-6 blocks to be built on any particular block that has a transaction of interest to you. It doesn't matter if a miner gets "lucky" for a few blocks, it is not something that can be kept up continuously.
    2
  35.  @64standardtrickyness  By fraudulent transactions (or blocks that contain them), I mean they are transactions that are breaking the network consensus rules. Some obvious examples: a transaction that tries to spend bitcoin without providing a digital signature from the private key (ie. spending bitcoins from an account you don't control) or trying to create a transaction that references an x amount of bitcoin but making the output be greater than x (ie. trying to spend more than you have in your account) - these types of situations are not a concern at all as the network instantly disregards those transactions and blocks that may contain them. The only issue left is double spending (which is what I think you are getting at in your example) - if someone has say 1 bitcoin, then makes 2 different transactions referencing that 1 bitcoin. On the face of it, neither transaction on its own is fraudulent, but both can't be true. If two different miners solve competing blocks, but each includes one of the two contradicting transactions, then both branches at that point are still fair game (there is no false block yet), but one branch is abandoned over time as the majority of mining power will choose one branch to build on (it makes no economic sense for a miner to build on a minority/losing branch). Even if a player with large compute power did go through with trying out your example, they would have very likely lost much more than they would have hoped to gain as they wasted valuable time and electrical resources mining the block or blocks they don't want included in the longest blockchain. What it comes down to is this: if you are receiving a large transaction, don't consider it "finalized" until many subsequent blocks have been built on top of the block that contains the transaction of interest, to ensure that you are not being double spent. For example, if i'm receiving 100 bitcoin, I wouldn't trust it after seeing it in only 1 block (I'm not going to immediately send them whatever they paid for just yet) - I can however be confident it is finalized after waiting 6-10 blocks have been built on top of it - someone trying to double spend that transaction to me at that point would not only need to redo that block, but they would also need to solve all the subsequent blocks as well, which would at that point have cost them much more than 100 bitcoin in electrical power to pull off. Miners make money by playing by the rules and building on the longest chain, not by wasting resources on blocks that don't get included in the longest blockchain.
    2
  36. Your understanding of how bitcoin works is flawed. If someone produces many blocks, it doesn't mean they are trusted, they still have to produce blocks in accordance with the protocol rules. Checking and rejecting invalid blocks is done extremely quickly, even my tiny raspberry pi computer can do that! Global usage of Bitcoin is scalable on second layer protocols. This is how millions of transactions per second can be achieved. For example, there currently is a decentralized second layer protocol for instant Bitcoin transactions called lightning network - over 99% of lightning transaction data doesn't need to be stored on the blockchain.
    2
  37. You've identified a key limitation in that the whole network has to store all transactions for all of time. This problem is prevented from getting worse by have a restriction on the number of transactions that can happen on the blockchain. Scaling is achieved on 2nd layers such as the bitcoin lightning network where we can have 100,000 times more transaction capacity and they are instant and practically free.
    2
  38. Yes, each of the transactions are verified by their digital signatures. Anyone can run a network node on their own computer and do a complete verification of the entire blockchain (no mining is needed to do this). Mining just wins the miner the ability to make and publish a block, which prevents double spending from happening.
    2
  39. More or less, yes. Sometimes a miner doesn't see a transaction that just occurred right before they mine a block and sometimes they can't include a transaction in their block if it is too full. In either case, the transactions just sit in the memory pool until a subsequent block picks them up.
    2
  40.  @fasti8993  great question. The software is completely open source, meaning that anyone can contribute to development and everyone can verify the reliability of it - though, if you can't code, you would have to instead trust the very many independent programers and security experts who do regularly test out the node software.
    2
  41. Nothing really. If you want to pay a fee worth hundreds of dollars to the bitcoin miners, go right ahead.
    2
  42. Actually, the required mining power is independent of the number of transactions happening. And "corrupting" the miners won't achieve anything because anyone can be a network node using a regular computer (no mining necessary) to verify the integrity of the entire blockchain - so you are not trusting the miners. Also, the miners don't have any real power. We saw this play out in 2017 when the majority of miners tried to push a contentious hardfork upgrade, but the rest of the nodes on the network enforced rules to make their new blocks invalid. The miners are just workers for the users of the decentralized peer to peer bitcoin network.
    2
  43. It isn't much of a stretch to assume that the vast majority of miners will build on a singluar legitimate blockchain. Miners aren't the only network actors, it's possible for pretty much anyone in the world to run a full copy of the blockchain, checking every single detail down to every byte of info, including all the proof of work produced by the miners, and their not going to follow a garbage blockchain ledger. This creates an extremely robust schelling point on a global scale.
    2
  44. Not the current Google computer, no. They might develop a powerful enough quantum computer in a decade or two, but bitcoin would have likely already upgraded to a quantum resistant algorithm by then.
    2
  45. Work here means electrical energy expended to secure the network.
    2
  46. likely will over time just push them to adopt quantum resistant algorithms.
    2
  47.  @RedValkyries  in a very extreme case, it would be possible. But hardware isn't the only factor, you would need to account for cost of electricity - paying for the electricity to power 50% of the mining capacity would be a very huge cost. In other words, if they want to be profitable, they would need to continue to play by the rules of the network instead of trying to attack the bitcoin network.
    2
  48. The point here with bitcoin is that these payments are digitally checked by every single computer on the network, no room for human errors of miscalculation.
    2
  49.  @MikeLisanke  the correct term I should have used is "temporary ban", not "blacklist". A normal node can quickly ban a connection from a misbehaving node if they detect too many repeated requests or if they misbehaving node is trying to relay an invalid transaction. But you are correct about nodes being able to change IP, run over TOR or a VPN, but in which case, the misbehaving nodes can be dealt with quickly again. I'm sure it is explained with more detail in the bitcoin GitHub, but it is also mentioned on the Bitcoin dot org website, look for "p2p network guide". This banning defense is a feature of the p2p network, not blockchain. The blockchain is just a large file, it is only aware of mining nodes whenever a new valid block is published. I would recommend you check out a book by Andreas Antonolopous called "mastering bitcoin" - free on GitHub since it is an open source book.
    2
  50. All nodes on the bitcoin network have an exact copy of the blockchain which show all addresses and associated funds. Nodes are also checking transactions in real time, seeing if they are correctly spending funds they have or not. Each nodes does a full blockchain scan once then keeps a running list of UTXOs (unspent transaction outputs). Yes, public keys will be known by everyone on the network, but that doesn't mean we know who the identity of the person is.
    2