One doesn't even have to read past the first three sentences to know that this guy is clearly talking his book. He owns DAO and also has a couple of other Ethereum-supported projects.
It's a well-written piece. He makes some strong arguments. He does a good job refuting some of the counter-arguments to a hard fork. But I think his fundamental argument that it would be MORE valuable is completely unsupported. Truth is, nobody knows the impact of the loss of trust caused by the sloppy work upfront and the efficacy of the attacks that inhibit the soft fork. What will the next attack be? Where will it come from? Will they roll everything back AGAIN to thwart it?
His argument that it somehow IMPROVES mainstream perception of Ethereum is, well, not actually an argument. It's an anecdote followed by an opinion. I happen to have the opposite opinion.
Finally, he doesn't address my main concern, which is the precedent of colluding with miners to do the bidding of the Foundation. This is NOT GOOD for a lot of reasons. One of those reasons is the precedent it sets of collusion. What happens if a government wants to intervene and threaten the miners to fork because they THINK there is a homeland security issue? What happens if they want to roll back just because they feel people aren't paying capital gains taxes? What happens if MegaCorp sees that they could easily afford the required bribes to do whatever it is that THEY want? ETH is roughly a billion in market cap. Apple has $160 billion on hand an Microsoft has $85 billion.
ETH isn't too big to fail. On the contrary, it's too small to save.
Colluding with the miners is not a problem. Some people think it is fine for the "hacker" to take away the money from the DAO by exploiting a bug in the system. It is also fine for miners to get the money back by not exploiting a bug. That is all the 51% attack does. It is in the design of any block chain.