NASA has started the switchover process to Side B - it turns out I was a bit fuzzy in my understanding before. The two "sides" are not the power source, they're computers that interpret commands from Mission Control, and send data back down to the ground. Either way, Hubble can't take any data while this equipment is down. Because it's actually a computer, and the process of switching from one to the other is complex, they haven't tested Side B since it was launched - why mess with a good thing as long as Side A was working. So during the process of switching over, they hit a snag, though they haven't detailed what the problem is. Time will tell how easily they'll fix it. At the worst, the next Shuttle Servicing (I think it's scheduled for February?) could bring a whole new "side" up and fix the problems.
In the meantime, many astronomers are losing their time on the Hubble. It's my understanding that if you lose your time, you have to reapply for more time in the future, but you tend to get a high priority if that's the case. (This is a similar process to how it works if you're on a land-based telescope and you get clouded out.) There are a few projects that have guaranteed time, and they essentially get a percentage of the functional HST time, so for any time that the HST is down they just lose their time. The people/institutions who "own" that time can choose to reschedule as they wish.