It depends on whether you believe in the wisdom of the market or not. Earlier, @jonny100 objected to a comment I made when he saidI just realized that to me, to hardfork is basically to undergo a small inconvenience in order to obtain the ability to harvest multiple realities in the future.
Is Ethereum better off undoing TheDAO or leaving it be? Is Bitcoin better off with a big blocksize cap or a small one? For a small price in inconvenience, each community of investors can try out both possible futures simultaneously and benefit if *either* succeeds. While not being harmed at all if the fork turned out to be pointless (except the minor inconvenience). In contrast, choosing NOT to fork during controversy and NOT to have a persistent split is subjecting the ledger to great risk.
Framed this way, doesn't it seem like hard forks (with persistent splits if there is traction) should be the obvious default thing to do in any controversy? Hodlers win because their bets are hedged with no downside, and partisans stand to win big if they bet correctly and be rid of their opponents. It's looking more and more to me that we should be taking advantage of every hardfork opportunity as it can only add value and resilience to the system. I believe in the future this will be obvious. The ledger is all. Copy it a hundred times and purchasing power is never diluted. (This also spells doom for the altcoin concept, since new ledgers are completely unnecessary, even as backups.)
"Nobody wants top down diktats from Core. This is just a political misrepresentation of your opponents. This language is not helpful to healthy debate."
But I don't think this is true. There is a subset of the community that believes that leaving bitcoin governance to the market will result in failure. I don't want to dig up the quotes, but Greg Maxwell, Adam Back and many others have made their view clear that bitcoin governance needs to be curated by "wise developers." If this is your view, then hard forks are bad because they give the market a chance to make a decision and--according to them--the market could easily make the "wrong" decision.
If they didn't have this view, then why wouldn't they make it easy for people to change their node's block size limit, as we did with BUIP001?