Sometimes we run into a series of bad luck when it comes to downtime, and then we have long stretches without any. The last week has been especially difficult. Wednesday's downtime and rollback was due to a bug which _really_ needed to be fixed. Are we looking at other options for releases? Absolutely. We're trying to strike a balance between infrequent, huge releases that increase the risk of major problems, and too frequent, smaller releases. I think you'll see us move away from weekly releases so we can have more time for testing.
Last night we discovered an exploit that allowed non-owners to delete the content of other users. It was imperative that we close that hole as quickly as possible. A team of Lindens started working on the problem around 2AM and had it addressed and code re-deployed by 8:30AM. We're very sorry for the downtime, and we know it caused some _real_ problems for everyone, but sometimes it's to everyone's benefit.
Again - we're very sorry for the downtime this week. We will be making changes to the release process so the frequency of downtime should be reduced. We're also looking at ways we can deploy new features or fixes without taking down the entire grid. In the meantime, we sympathize with the difficulties the downtime caused and we apologize for any hardship you encountered.
__________________
You must Be the Change you wish to see in the world.