Establishing a transaction fee market. There is, and has been, a fee market for Bitcoin; it was not important until we reached critical capacity. Once you reach a point were capacity is constrained, fee markets come into play. Fee markets will happen, and if you don't build a primary fee market, people are going to build a secondary fee market.
How do we create the possibility of fulfilling transactions at much greater volume for people who can't afford the current level of fee? Or if it continues to rise? The answer we seem to be arriving at is: secondary and tertiary layers above Bitcoin, the most promising of which is Lightning Network at the moment. Micro-transactions, nano-transactions.The Bitcoin blockchain is a very valuable commodity for applications; the fees make it so that it is not economical for everyone to use it as a back-up platform.
No matter what capacity you create, an engineer will create an application to fill it. Fee markets help us make choices about which transactions are important. There are two ways to make choices: through authority or through a market. The market says "if the person sending the transaction thinks it's worth paying a fee, then it's a legitimate transaction." Because of this, adoption in some areas will require higher-level algorithms.
Transcript
[AUDIENCE] I have a question about Bitcoin governance. I would like to know your point of view about... establishing a free market for Bitcoin. I am very concerned about transaction fees being [too high] for poor people all over the world.
I know there is a portion of the community that want to establish a fee market for Bitcoin, that could raise... fees for transferring money, and I want to have your point of view on this subject. Thank you. [ANDREAS] There is and has been a fee market for Bitcoin.
It was not important until [recently], when we reached the point where capacity was critical. Fee markets don't matter if you have unlimited capacity or [until] you reach the point where the capacity is full. Then fee markets come into play. Fee markets are not a choice [we have].
[It is not a question of], "Should we do a fee market or not do a fee market?" Markets happen. You can say, "Don't make a market," and markets will happen anyway. Fee markets will happen. If you don't have a primary fee market, people will build a secondary fee market...
by using up all of the [block] space and then reselling it in a fee market. Fee markets will happen whether you like it or not. The question is, how do we fulfill a much greater volume of transactions for people... who can't afford the current level of fees, or if they continue to rise?
The answer we seem to be arriving at, is through the use of layers above Bitcoin. Secondary and tertiary layers. At the moment, the most promising is the Lightning Network, which is a system of payment channels... that have settlement into Bitcoin, but allow for a much higher rate of transactions per second, with millisecond latency and very small fees.
It would enable not just microtransactions, but what I call nanotransactions, with very small or no fees. If that is used, we will have [more] capacity problems again later, and we will find another scaling solution. If that [solution] is used, we will find another scaling solution, and keep doing that for twenty-five years. We will be as successful as the internet has been, in terms of continuously evolving to scale for...
new applications. Most people find it difficult to understand that [because] the Bitcoin blockchain is a highly secure, very robust and decentralized platform, bitcoin is a very valuable commodity for applications. If you have zero fees, then everybody will use it to backup their computer files on to the blockchain. Why not?
I would back up my computer. I would dump gigabytes on there. Great. It is replicated internationally with an entire reward system to make sure that...
everybody will keep my data backed up forever, and it is free! So what would happen? We would use [up the] full capacity. The fees make sure that it is not economical for such use cases, which will restrict its use, but if you increase the capacity of the blockchain, the applications would simply expand to fill that capacity.
Applications on the Bitcoin blockchain behave exactly like a gas. Do you remember physics? Gas always expands to fill the container. The blockchain is the container, applications are gas.
If you gave me a blockchain today and said, "It can handle a billion transactions," most people would think, 'I don't know what we could use a billion transactions for.' Computer engineers would think, 'I wonder if I could use a billion transactions per second for this application...' I once had someone ask me, "Do you really need to have a gigabit connection at home?" Yes! I am a computer engineer. If you give me a gigabit connection twenty-four hours per day, I will have it pegged at one gigabit up and one gigabit down, full time. If you give me a ten gigabit connection, I will make a backup of the entire internet.
A one hundred gigabit connection? I will back up every virtual reality, 4K, or HD movie there ever was, then serve it to the entire world [for streaming]. There is no limit. No matter how much capacity you create, an engineer will invent an application to fill it.
The fee market is how you stop that from happening. You basically say, "We need to make choices about... which transactions are important or not." There is two ways to make choices. One is through an authority figure.
The other is through a market. An authority could say, "I think this is a good transaction, and this one isn't." You need to choose who makes that decision. The problem is, if they decide for everyone, that is not a very good way of doing it. A market-based solution says, "If the person sending the transaction thinks it is worthwhile by paying a fee, then it is a legitimate transaction.
Finished." That is the way you keep it open, for people to [add] applications. That will [initially] cause some problems for [Bitcoin] adoption, especially in countries... where a microtransaction is not one euro but a rupee. For that, we will need to build higher-level algorithms.