Interesting challenge! I'm a fan of Innocentive. Most of their challenges have respectable prizes and best of all, they are guaranteed to be awarded. I think a lot of us are held back by a mentality in which our ideas and creations don't seem "good enough," but this perfectionism holds us back: le mieux est l'ennemi du bien. Engineers don't waste time trying to perfect their designs, they seek a balance between competing objectives, and one of these is always to minimize time and effort spent working on the design. The end result is often a crappy compromise that happens to be less crappy than all the others. Look at how stupid turn signals are! When the signal is active, current flows through the wire to light the bulb. The wire heats up, causing it to bend until it touches some metal and shorts the circuit. This makes the wire start to cool down, so it straightens back out and stops touching the metal and the light bulb circuit is restored. It's a perfect pattern for a maddening electrical malfunction that takes forever to locate, the bread and butter of auto mechanics. In this case it works great, is utterly reliable and costs $3. A tough aspect for the NASA challenge will be the size and shape constraints. The existing ballast masses are tungsten, which is almost twice as dense as lead. This is why tungsten has replaced the depleted uranium once used in balancing weights in aircraft, after crashing 747s led to some understandable health concerns. Mass is the important factor for liftoff, but NASA seems to discount volume when they ask "If you had up to 150 kg of ejectable mass prior to entry ... what could you do with it?" A Raspberry Pi is 45 grams in a 100cc package. If you used 150 kg of them to construct your ballast gizmo, it would be bigger than a 55-gallon drum. NASA might not be comfortable wrapping their rover around such a device. To maintain the desired mass, the winning entry will probably be a big chunk of tungsten with its own small "science payload" aboard. NASA is already studying impact craters from the ejected balance masses. What if we glue a GoPro to them?