A full datacenter power failure just caused about 55 minutes of downtime.

Of course we weren’t able to alert you through the blog — it was down — so we’re doing to use a 3rd-party system for that starting immediately.

We’re also looking into what it would take to do some sort of (probably read-only) off-site backup that would allow us to at least continue basic service in the event of a complete physical disaster, but that will probably take a few months to come to fruition.

I won’t say “thanks for your patience and understanding.” Instead I’ll say: We’re going to start building out new ways of weathering even this sort of storm.