That if your home computer had to do 2.25TB/day down to be a full node you might not be able to do it, which is a centralising factor.
You're saying that retaining the possibility for home computers to be full nodes is both necessary and desirable.
I understand that many people have made this
claim over the years, but I am not aware of any attempts to actually prove either of those.
He's saying that increasing the full-node bandwidth requirements is a centralizing factor. It is. If it requires more resources to run a full-node, less full-nodes will be run (all other variables held constant). What is "necessary and desirable" is a different question: we want the Blockchain to scale but we require Bitcoin to remain decentralized. Like you've pointed out several times, the proponents of keeping the blocksize limit small can't prove their claims; however, those who wish to increase or remove the cap can't prove their claims either.
Bitcoin has value because it is decentralized. It has value because the Blockchain is an unforgeable global ledger accessible to anyone according to the protocol rules, and secured by the largest single-purpose computing network ever constructed. It has value because another individual, company or (possibly) government cannot prevent you from doing what you wish with your coins. It has value because the majority cannot vote to redistribute your funds or attempt to paper-over economic problems by printing additional bitcoins. In summary, it has value because it behaves less like a human invention/institution and more like a phenomenon of nature itself. Just like there's nothing in the world we can do to stop gravity, we must also be unable to stop the ceaseless chaining of new blocks upon the Blockchain.
Full-node count is
currently estimated at 6495 nodes, and is steadily dwindling despite the fact that the blocksize-limit has remained static at 1 MB. How would full-node count change if the average blocksize were 10x larger? 100x larger? How many nodes do we even need? I think these are difficult questions to answer.
The current price reflects the belief that Bitcoin has the potential to one day become a significant world currency/SoV. Like Smooth pointed out earlier, that pie-in-the-sky $10,000+ price multiplied by the market's estimate of its probability of success sets today's price. If Bitcoin can't scale to the extent necessary to fulfil this potential, then the price of bitcoin today will become adjusted to this new reality (lower). How much scaling is necessary? I don't know, but I'm pretty sure the 1 MB limit put in place by Satoshi many years ago as a stop-gap measure is not it.
BTW, I don't think the "free-market-solves-all-problems-so-remove-the-cap-completely" approach can fly. The Satoshi Social Contract requires that certain system-level constraints exist (for example, the inflation schedule
.we can't have the free-market solve that problem). Another requirement is that Bitcoin remain decentralized. Without a constraint on blockchain growth, it seems less likely that this would always be the case.