Next we can make the goal-nets out of mosquito netting, so we can trick them out of dying of milaria.
This is colloquially referred to as a "bad sign".
In all seriousness, though, this looks like a pretty crappy product. "Shake-light" generators like these are already pretty miserably inefficient, and you'll only be getting maximum power when a kick is perfectly aligned with the axis of the coil, with the magnet at the far end. If you kick it at a right angle to the coil, or with the magnet at the near end, then it won't generate any power at all, and, even better, will react differently to the kick.
You could use three generators, for all three axises, but that would triple the BOM, and the magnets will interfere with each other. And you still have the "magnet on the wrong end" problem, and the damn thing will still wobble.
Additionally, electronics designed for use in adverse environments can either be cheap, or they can be good. Soccket could make them cheap enough that the target demographic could actually buy them, and get tons of bad press when they break more or less instantly, or they can make them good, and have one NGO buy a hundred of them... and have them break slightly less frequently. Soccer balls are not terribly durable, (because they have to be light enough to kick) and when your reputation rides on miraculously making them much more so, you're in trouble.
There's a veritable glut of seriously awesome designs designed to bring benefit in the developing world -- two of my favourites in the last decade are ColaLife (http://www.colalife.org/ ) and Josh Silver's liquid adjustable glasses (http://www.guardian.co.uk/society/2008/dec/22/diy-adjustable... ). While both are incredibly neat, both seem to have have had poor traction (the glasses are, admittedly, doing well, but there's 20 years of work and still only a relative handful in the wild.
In the CNN interview they mention it's provisionally "Not that much more expensive than a regular high-end soccer ball". Sounds like they plan to use the XO get-one-give-one plan.
Unless they can lower the cost, and come up with a funding scheme that doesn't rely on a few nice folk in rich companies buying what will essentially be a useless novelty to them, I don't hold out too much hope for this changing any lives very soon. Sadly.
What happens when it's kicked into a pole, a car, a wall, etc? What happens when someone directly kicks the output port? I can't imagine this not being quickly damaged in real use.
Would it weigh more than a regular ball? Seems like it would have to. Worse, the internal movement would probably cause it to "wobble" through the air, at least a bit. Wouldn't kids avoid a ball that played different than regulation?
Almost all of the ideas for harvesting human energy or other incidental energy has this problem where people forget about conservation of energy. If you gussy up your road to collect energy from passing automobiles, the fuel economy goes down. If you hack your floor to do the same to passing people, that makes your floor harder to walk on. People don't really output that many watts under any circumstances, the only thing I've seen that is useful is extracting milliwatts for very small electronics like pacemakers or something. These ideas keep winning design competitions but I think we never see them because they don't actually work in the real world with real people. It seems like the same amount of money put into solar panels is very likely to be a far more effective investment.
Couple that frustration with the fact that regulation balls are common and cheaper, and I don't think you'll find these popular at all in practice.
I'd like to be wrong about that; it would be great if it worked out. I just don't think there's much chance of it going anywhere in reality. Any money behind this would be better spent on distributing small solar panels and batteries to serve the same purpose.