A good QC owner would use it as follows:
We don't think that QC development will happen step by step.
Our expectation is that someone will find a QC technology, that allows "far beyond expectations" numbers of qubits, that will allow this QC to get all private keys immediately.
We think that such a QC will surprise the Bitcoin community and only thereafter we will upgrade to a quantum resistant Bitcoin network. We hope that the user of such a QC to get the private keys, knows exactly how Bitcoin works and allows the owners to transfer their coins to the new QC resistant addresses. It would be a win-win game: the QC user would get the "lost" coins, the Bitcoin owners could transfer their coins to QC resistant addresses, the Bitcoin ecosystem wouldn't be affected, we would have a stronger Bitcoin network. How would a QC user act: starting with the oldest "lost" coins and moving them, so that the Bitcoin community can realize that someone is moving the "lost" coins (e.g. a special posting board here on bitcointalk) but gives the owners the possibility to transfer their coins to other addresses. In the meantime we will have a very quick "quantum resistance upgrade". And it will continue like DannyHamilton described it:
The coins that are still remaining in the weak transaction outputs once Quantum Technology becomes a realistic threat will be those coins that are effectively "lost". The QC owners will become the new owners of those coins, and Bitcoin will carry on as it always has.
but stronger
My bold.
Point 1 - There is a common misconception about quantum processing power. With a classical computer, this scales linearly. With a QC, it scales exponentially with the number of qubits, 2
n. So as you increase processing power:
Classical: 1, 2, 3, 4, 5, 6 etc
Quantum: 1, 2, 4, 8, 16, 32
I definitely think that once we have a reliable low qubit QC, then the steps to a powerful QC that can break public-key cryptography may be achieved more rapidly than commonly anticipated. It's a mistake to think in terms of how 'normal' power in computers scales up. Not saying you're doing that at all, it's just a point of which we should all be aware.
Point 2 - It's one option, but I thinking burning the coins that aren't moved to q-safe addresses is preferable. Ideologically it's questionable, sure, but 5m or 6m bitcoins suddenly available to possibly a single bad actor could quite reasonably be considered an existential threat. And it
would be outright theft, not a 'reward' for developing a QC. Unless the real owners consent, which of course they don't. Hard fork and a burn seems the sensible option. The question here is: what should happen when the purity of the original vision intersects the problem of basic survival? A safety tweak, or death?