4 ms·
You're 100% right. That's why I argued the GS case. Competitor or miner A's profitability could change in a heart beat if a new competitor with gravitas entered
by scrubs 5y ago
You're 100% right. That's why I argued the GS case. Competitor or miner A's profitability could change in a heart beat if a new competitor with gravitas entered the market. They could take a lot of market share and induce existing miners to switch block chains. Since all one has to do is have power, computers (agree: gross simplification there) one's business could change easily. Using BC as collateral, for example, isn't the slam dunk it seems. As I say the miner agreed BC isn't currency; the fallback is BC is an asset. But it's an asset that seems fundamentally more risky than a house.
This contrasts with AWS, Nvidia, GM, or AIG: it's not so easy for new comers to take market share. You gotta have a lot more going for you than a data center monthly rental bill.
- saurik 5y agoIf all the miners in the world are mining something, it doesn't really matter much if the users (who are paying for the security through inflation) don't want to agree to the new rules. In fact, the miners that defect to your system are all just hurting themselves: no one wants their bogus blocks, and so they are just throwing energy away. Sure, you are supposedly paying them (by your tweaked rules that gives them more reward per block), but you are paying them in a currency that is incompatible with the software used by every exchange, every bank, every supermarket, and every other user, who all see your "more profitable" blocks as "invalid"... they are buying real energy and you are giving them a currency no one has heard of or accepts in exchange for goods and services. What makes these networks hard to get everyone to switch between, even when they suck, is thereby all of the momentum of software deployment and brand recognition and "belief in the value of such" that they have. This network effect is extremely difficult to quickly replicate and is a massive barrier to entry for not just newcomers but even mere upgrades... bug fixes to these protocols require everyone to agree! Truly: if you could just create a new network that was slightly better or which paid miners slightly more per unit (which isn't real value) and somehow immediately sweep away all of the miners, Bitcoin--which is clearly an antiquated piece of shit to most of the people who work in this space--wouldn't exist, as not only do new networks come out every day that offer small improvements but new networks exist that offer large improvements, and it is extremely painful to get anyone to switch.