0
I'm reading about the concept of UTXO. Everywhere I read, it says that a UTXO in an account is indivisible. One analogy said: "It's as if you wanted to make a $5 purchase with a $10 bill. You wouldn't tear the bill in half."
Let's say I have 0.1 BTC as one coin in my wallet and I want to send 0.05 to someone. What I'd do (minus the fees) is send 0.05 to the person and 0.05 coin back to myself, right? That is, it will be marked in the coin's ledger that I and they respectively are assigned that fraction of it. At the moment of the transaction, wouldn't the coin be split?
Also, at the moment of original mining, aren't all new coins worth 1 BTC? If that is that case, they must be split in some way at some point. Otherwise, the amount of bitcoin that people own in the world would all be integers.
Re: deprecating the term UTXO in favour of payment. Im curious, where have you seen this proposed/discussed? I haven’t seen that before, and it seems to be a confusing change that is less descriptive and accurate. – chytrik – 2019-02-02T02:27:19.967
The practice was identified to me by Eric Voskuil, the lead developer on the libbitcoin Team. The rationale is that not all outputs can have an associated address. Payments can be made to scripts. Additionally, new address formats come to fruition, aka P2PKH, P2SH, P2WPKH. All transaction output scripts of a transaction be effectively hashed to be uniquely indexed. According to Eric, Electrum has been using such an approach for a while. It is easier to index one payment ID than a UTXO ID along with a specific output index. This appears to be a technique to future proof core infrastructure. – skaht – 2019-02-02T03:21:51.947