On 12/26/2017 4:04 PM, Steven Schear wrote:
Clearly they are not, yet. The problems are somewhat multi-dimensional and the way forward isn't assured. If some counters wanted to scale a "Bitcoin-like" chain to handle, on-chain, the average transaction volume of PayPal (about 120/sec.), quite a coup, it would require (by my reckoning) a block size of about 64 MB. The Bitcoin Unlimited people are planning to test blocks much larger. Clearly, this would likely result in considerable miner concentration, unless, as Garzik's recently announced Bitcoin United (which includes a "our" Escher feature) takes over Bitcoin Corey's mantle.
Every peer has to download every transaction, in order to judge it for validity. If relying on someone else's claim about validity, not truly a peer. It is having many peers that makes the network resistant to control and attack. This inherently limits the number of transactions that can be made fully on the blockchain and fully peered to something that is not compatible with taking over the world. Thus scalability, up to world conquering scales, requires an architecture where most people do not have peer wallets, but client wallets, and most transactions are consolidated into fewer and larger transactions by important peer wallets functioning somewhat like banks before the transactions are entered into the blockchain. The Lightning Network is intended to have transactions consolidated by important peer wallets functioning somewhat like banks, but it is not altogether clear how this will work in practice, or if it will work in practice.