As always, the version of the protocol utilized by the majority of the network stakeholders will control the chain.
This is largely incorrect, and if you are operating under this misunderstanding you will eventually break the coin and destroy its value entirely.
You can not have any significant
minority running an incompatible client or you will have transactions that confirm on either or both chains and it will never be possible to agree who owns which coins. The coin will be entirely destroyed.
A
majority of network stakeholders is useless for determining the network rules. You must have either near-unanimity on the rules, or any disagreements must be resolved by creating a new coin on a separate network.
I can't argue against the idea of conducting polls, as that is just free speech, but if your idea is to deploy a contentious change to the system (and taking away coins away from people who own them will certainly be contentious whether voted for or not) on the basis of a majority vote, then you are misguided.
You are under-estimating the ability for humans to reach consensus.
If there was a situation with contention, users have a vested interest in verifying and agreeing with the majority chain.
This creates a very powerful incentive to update.
Of course, the minority could choose to continue running incompatible software.
They would, essentially, be on their own network.
Ideally, the other major parties (such as services and exchanges) agree with the change; but, in the end an exchange client has no value if it refuses to verify the transactions of the economic majority.
Communication would prevent the network from getting to that point, but for illustration:
When faced with one support request for fork A, and 50 support requests for fork B.... The choice is quite clear.
How do you think new network protocol rules are implemented during a fork?
By definition, majority stakeholders control the chain.
That said, there is no covert plan to do anything at the moment except allow the community to make their own decision.
Then, if a decision for change is made, there is development, evangelism at the various services/exchanges for adoption, update and eventually a change to the network protocol.
You, nor I, have to agree.
In the case of a fork, that which represents the majority, and thus is the most useful, will undoubtedly be adopted.
Regardless, you are jumping ahead of the starting line.
First, the community gets a provable voice.