You can always create and altcoin with the same addresses as Bitcoin and with the same distribution of existing Bitcoins over those addresses. Than everyone can choose which Bitcoin to use (or both). I would personally feel safer using both concurrently than to change the existing social contract of the original Bitcoin. When the first change is made, the discussion for the 21M limit will surely start as well ....
I believe that is the spin-out proposed by Peter_R. While there is any chance of retaining the Bitcoin brand and its connected customers, I will work entirely in the framework proposed by Satoshi who allowed for changes via a consensus of full node miners. There are about 8300 of those today, and a million wallets at Blockchain.info.
In a possible world a couple of years from now, there is a peer-reviewed and invulnerable Proof-of-Stake version of Bitcoin Core - not yet deployed but running great on a large scale Bitcoin testnet. The sales pitch to Blockchain.info is simple. Blockchain.info arranges to lease m3.large instances from Amazon Elastic Compute Cloud that support Bitcoin Proof-of-Stake Bitcoin Core full node instances for an online wallet. Blockchain.info makes an offer to its million customers to pay a minimum of annual 10% bitcoin dividends to each online wallet that chooses Proof-of-Stake on the condition that at least 50,000 enroll. Blockchain.info keeps a fee of 2% of the earned rewards.
50,000 low-latency, well-connected full nodes on the new Proof-of-Stake version easily outvote 8,300 full nodes on the old Proof-of-Work version and on the pre-announced day, the blockchain forks. Ordinary bitcoin users are unaffected. Suppose at that time bitcoin is valued at $10,000. The daily mining reward is 3,600 bitcoin divided among the 50,000 Blockchain.info enrolled wallets. Blockchain.info's daily commission is $720,000 from which they pay for the Amazon instances at $72,000 daily, a gross profit margin of 90%. The deal is great for the 50,000 enrollees who get daily dividends each of bitcoin worth on average $705. Of course these numbers are for the early adopters. Once the remainder of the million online wallets enroll, the average daily dividends drop by a factor of 20 yielding a daily average dividend of $35 dollars.
From a simplified, possible world business case such as this example, it is clear to me that a very large opportunity exists to disrupt the disrupter. The challenge is to create an invulnerable and broadly acceptable Bitcoin Core Proof-of-Stake.