The reason I'm skeptical about Reprap is that it's not a business, but they're trying to make hardware. So I'm not sure that they'll ever achieve their goals before a startup does it first and better. Apple had a lot of money, and they used plenty of it to create the Mac, which was the first desktop computer that had the interface and capabilities that most people really wanted (easy desktop publishing, for example.)
1) Low cost
2) Open source
3) Self-replicating
These attributes make it a lot more economically accessible to people who wouldn't otherwise be in a position to learn about and use this technology.
It could allow a bootstrapping hardware startup to produce their first product prototype. The economic value that could result is immense.
Just because RepRap doesn't meet your needs doesn't mean it's not worth talking about.
I am not saying it needs to be able to use dirt, sunlight, and water to make another copy. Just that the components need to be a little simpler before the cost really drops.
Sure it can be improved on massively, but you must acknowledge that it is at least a step in the right direction.
Last time I looked projects like Firefox and Wikipedia weren't businesses and they are still doing a sterling job competing with large enterprises like Microsoft and Britannica.
I'm way more excited about lower-cost selective laser sintering (SLS) machines. This process yields high-quality (and high resolution) metal parts that most people would want to use in their daily lives. It's also something that we're not going to see out of the RepRap project in its current incarnation. SLS is startup territory. Current machines run something like $100,000, and anyone that figures out how to make this much cheaper without losing quality will do very well!
So I agree with the larger point, which is that something like RepRap (self-replicating, etc.) is interesting. But I'm not convinced that RepRap is that project.
But what are the alternatives to RepRap? Makerbot maybe? Any others?