I think the problem here is not the risk of bricking the rover. It has two on-board computers in a redundant configuration, so if the update fails, the watchdog is going to switch to the secondary computer. Also, the integrity of the update is most likely protected by a checksum.
The real risk is that a valid on-board software could send unexpected commands to the other subsystems (that is what happened on MGS and many other failed spacecraft). For example, it could instruct the rover to drive off a cliff;-)
Slashdot Mirror
I think the problem here is not the risk of bricking the rover. It has two on-board computers in a redundant configuration, so if the update fails, the watchdog is going to switch to the secondary computer. Also, the integrity of the update is most likely protected by a checksum.
The real risk is that a valid on-board software could send unexpected commands to the other subsystems (that is what happened on MGS and many other failed spacecraft). For example, it could instruct the rover to drive off a cliff ;-)