That's awfully cynical. They took a wait and see approach. Would you run your engineering team any differently? How do you tell a viable fork from the (many) others that have failed? If BCH falls to $10 in two weeks, would you still direct your engineers to support it?
If you knew me in person you'd know the cynicism was intentional. I've got quite a bit to start with and double so when it comes to anything involving digital currencies.
> They took a wait and see approach. Would you run your engineering team any differently?
> How do you tell a viable fork from the (many) others that have failed? If BCH falls to $10 in two weeks, would you still direct your engineers to support it?
I don't think it's possible to tell after less than a day if it's a viable fork. It's way too early. That's why I'm thinking this is driven more from the perceived cost of lawsuits vs. the development cost of supporting the feature.
I see. I made this point somewhere else in this thread, but there's no lawsuit threat because of the terms of service (just arbitration), and if someone were a big fish but asleep at the wheel for the last few months, Coinbase could easily make them whole quickly by just manually sending them the BCH they wanted. Engineering full support as they're doing almost surely costs more than any conceivable loss customers could suffer in the interim (assuming anyone with a nontrivial balance takes responsibility for their actions or inaction).
My impression from reading their announcements before the fork is that it had always been their intention, if the new coin had any value after the fork, to later allow all users to withdraw their "cloned" balance. The transaction freeze around the fork plays into this: not only it protects them against any instability, but it also is a convenient time to snapshot everyone's balance, and to keep a copy of the private keys for future use. They probably planned the balance and keys snapshot since before the first announcement, but just didn't want to promise anything (remember that for instance the fork originally didn't have strong replay protection, which was added just a few days before the fork).
Since the fork didn't die immediately, they are going ahead with the "allow everyone to transfer their BCH out" plan. Had it died, they probably wouldn't bother.