I think the case could be made at that point, but I think the issue will be moot by then anyway. It’ll be like IPv6, everyone will say it’s needed, but we’ll never need it, and there will still be new and better ways to accomplish the same thing, and then we’ll probably have another paradigm shift in cryptography or who knows what with layers on top (like Nostr and pear are likely to remake the web), and then we will be so excited and busy with that, that a hard fork for higher tps just won’t be worth all the trouble. 

Thats at least how I picture it in my mind. We usually suck at predicting the future and people come up with bonkers stuff that seemed impossible in a shockingly short amount of time.