Or at least, we've started to. When you've got millions of people installing your software each month (I'm not exaggerating), you take it slow. We've got good QA, and we've had an extended beta on this thing, but you still run into problems in the wild that you just can't encounter anywhere else. So we start off slow. We'll start by releasing it to a small subset of our Zango installs, let it run for a couple days, analyze the numbers, listen to customer feedback, and make sure that everything looks good. If there aren't any red flags, we'll release it to the rest of our new Zango installs; wait a while; analyze; and then start upgrading our existing audience, again starting small. And once we're finished with that: then we start over with our next brand (Hotbar), and rinse and repeat until they're all done. The process will probably take two months.
And of course, by that time, we hope to be well into coding 10.1. Now if I can just keep the feature set down to a 90-day dev cycle . . .