RE: Who likes hard forking?
You're right to clarify that. Thank you.
However, I would add that the main branch that matters to most stakeholders here is the one that is tied to the STEEM symbol on the exchanges. There is also a difference between a hardfork in the historical sense of the term on the Steem blockchain, and a hardfork of the BTC/BCH, ABC/BSV, ETH/ETC types that leave two viable chains behind.
When I think of hardfork in the Steem context, my default mental model is one in which only one branch survives. In that context, I think terms like "property rights" are appropriate. It could be argued that they're also valid in the other sense, too, because the split is merely a dilution of the original chain. That's not as clear, though.
How do you suppose it would go if IBM tried to say, we're doing a 2 for 1 stock split for every shareholder except the California Pension Fund?
There is no "main branch" predetermined, nor is it predetermined which chain will inherit the name, until after the fact, when it is observed which branch most or all of the users/economy/community deem worthy of support (this includes exchanges as "users" who have the ability to decide which chain(s) to support and which name(s) to use, and they occasionally don't agree on this). During the previous forks which you mentioned (which of course after the fact all turn out to be chain splits), there were discussions about what would happen if "unexpected" chain ended up being dominant. Even Vitalik, who was on record in favor of the ETH (as opposed to ETC) fork, stated that if the community did not support the fork, he would accept that and the chain with unmodified rules (the path which is now called Ethereum Classic) would continue as "Ethereum".
There is no difference that can be determined until after the fact. At least in the case of ETH/ETC, it was not widely expected to leave two chains behind and came as somewhat of a surprise when exchanges began listing ETC. In fact it started out much the same as an upgrade hardfork, but it didn't end up that way. The difference is determined by the emergent properties of voluntary individual (and the collectively community) choices.
The only one your mental model matters to you is you. You can not impose your mental model on others. They are free to make different decisions than you, whether that results in one chain or two. If significant groups of people have a different idea of which rules are best, there will be two chains. If not there will be one. Nothing other than individual choices and free associations determines this.
Appealing to centralized entities and securities instruments which are not based on voluntary association and are in fact legal claims on real world assets (of which decentralized blockchains and cryptocurrencies are neither) is at a minimum a seriously flawed analogy. The only question is one of degree in how flawed. Stock splits are not forks, and you can not fork a building. The CPF has a legal claim on its proportionate share of the IBM assets. A blockchain has no assets and a cryptocurrency holder has no such claim to make against the blockchain itself (a nonsensical concept) nor against someone else who wants to run different code.
Agreed. Which is why my previous paragraph stated that it is less clear in the case of two viable branches. But DPOS is also different from other blockchains in the regard that stakeholders other than miners/witnesses have an on-chain voice in which fork gets selected. (unless the fork were to reset that mechanism too...) Flawed though it is, I think it's a useful analogy to consider when casting one's witness votes.
The complexity that you accurately highlight in the rest of your reply is the reason that I'm scrutinizing the witness statements so carefully. Risk and uncertainty abound in the cryptocurrency space, but I'd prefer not to vote for witnesses who (in my own estimation) are prone to increase those aspects of the ecosystem.
Which reminds me... Thank you to you and other witnesses for posting these statements. I should have said that in my first reply. And thank you for your additional dialog, too.