The Christmas excuse is no excuse whatsoever.
If you don't want to work on Christmas, close down your shit, and especially don't push some potentially catastrophic update out.
Oh, it'll cost you too much money to close down over the holidays? Then too fucking bad, I guess your multi-billion-dollar international company will just have to find some way to have staff to cover the holidays.
You have zero perspective on how online companies operate, then.
There are always on-call engineers/people on other teams, hence why the problem got fixed. However, do you propose that the ENTIRE COMPANY not take holidays and stay at the office/working? Or that some people be forced to work while others are away? Or that the entire storefront be closed because the entire company isn't sitting at their desks?
On-call rotations exist for this very reason: always online services such as Facebook, Amazon, Google, Steam, etc, need someone to constantly be at the beck and call if something DOES happen, and they were, and it was fixed within hours. We're still humans, though, and that person needs to get a call/text/email, get to a computer with the appropriate environment, and diagnose/fix the problem. The issue of releasing a public statement for a huge data breach is even trickier, because there are likely only a handful or people who are authorized to publish anything to the public, and who know where or who those people are. You think any PM or engineer at Valve can make a statement? No, everyone is probably told to go through a PR person before doing ANY public messaging, and that's just how it is.
More likely than not, they DID reach an on-call person from the PR department, and got a statement they COULD give. Any further statement is probably going to require collaboration between responsible engineers and their team/leadership, management team members (perhaps up to Gabe himself, or his direct reports), the legal department, and full understanding of what happened, looking at logs, and discussing at length before PR can summarize all the data and give the public a response. There are most likely people losing their time off over this right now, but I am sure that the full collaboration to give the public a more detailed response won't happen until everyone, or most people, are back at the office on Monday.
Most likely ramping up infra ahead of the sale/holiday rush.
Makes sense. We also have a rule of "nobody talks to the server team after Dec. 1st", since any and all changes like this could be catastrophic. Seems like unfortunately, in this case, someone messed up, and it was.