Andresen's point is that as long as Bitcoin Core defines the consensus rules, the few people that control Bitcoin Core control Bitcoin, but their status won't last forever.
The good news is that the oligarchy is already starting to crumble. There now are a few independent implementations of the consensus rules and their influence is growing.
Bitcoin XT wasn't the first alternative, so I don't feel that Andresen and Hearn can claim that they started the movement away from Bitcoin Core's control of the consensus rules, but they certainly have publicized the issue.
Someone has to define the rules, and of course it won't be the same group of devs for ever (it's changed a few times since Satoshi already). But there can only be one set of rules to govern a concept called "consensus" (the hint's in the name...)
You seem to be arguing for competing sets of consensus rules. May I suggest that you re-acquaint yourself with the implications: very serious adverse consequences for the network. Disagreement between nodes on which chain is valid will cause unplanned blockchain forks, and competing consensus rule invites this explicitly. That will manifest in an experience for the user somewhere between inconsistency and serious confusion, and that will damage confidence (and the exchange rate).
I am not arguing for competing rules. I am arguing for competing implementations of the rules.
Multiple implementations is better than one because one implementation means centralized control. Nobody can unilaterally change the rules or break the network. Also, keep in mind that when there is a change to the "one" implementation, the result is
two implementations (which was the source of the fork in 2013).
I agree that a scenario with only two implementations is risky. Ideally, no implementation would have a majority so that if a change to one implementation is incompatible, only that implementation suffers. I think that four or more is sufficient.
