More here: http://www.guardian.co.uk/science/blog/2011/oct/04/nobel-pri...
The real problem would arise a few years down the road, though, next time the Nobel gets awarded for some high-energy experimental work (discovery of the Higgs boson?). Those papers can typically have hundreds of authors, listing everyone involved with the project (the most extreme example I've seen has 2512 authors). Now all of a sudden pretty much everyone in high-energy physics is a Nobel laureate, and the proper Nobel laureates are a tiny minority.
I think the constraint that no more than three people can share it is fair. It just means the committee needs to look into it and find out who the real brains behind the operation were.
But the problem is that this is impossible. Who are the real brains? The person who had the idea, but didn't do anything about it? The person who organized the project and got funding for it but didn't do any real science? The person who build the hardware without which nothing would have been possible? The person who wrote the software, without which nothing would have been possible? The person who wrote the paper that explained the results? The grad student who found a crucial error in the analysis without which the results would have been wrong?
In many (most?) fields of science these days, progress is made by huge projects. Clearly there are wide differences between how crucial different people are to the success of the project, but to cut it down to 1-3 "winners" to the detriment of everyone else hardly seems fair. It seems more reasonable to say that if there are 200 people on the paper that's awarded the prize, everyone is a 0.5% Nobel Laureate....