Anyways, that picture does not mean what you think it means, it is incorrect in nearly everything that's written on or below it and even though a lot of people suddenly seem to act as if they got a PhD in distributed systems design just by looking at this, they did not. Sorry. It's very nice to look at and apparently easy to understand though, so it might have some value as example.

I might not have a PhD in distributed systems design, but I'm pretty familiar with them, and at least at this point I think those pictures are fairly accurate representations. Although one could very well argue that bitcoin is more similar to the picture on the right as well with mining pools being the centralized authorities.
This is one of the more interesting ones - the thing is that with payment networks like Ripple and also Bitcoin it depends a lot on network effects and adoption rates. There is little reason to use MasterCard if you can use Visa everywhere it is accepted. Starting your own competing credit card however is going to be really hard, if not impossible unless you can beat MC/V on nearly every battlefield that exists - and even then, there is a good chance that they will adapt and absorb your innovations. Just like people predict Bitcoin will act in the Altcoin scene.
Well most altcoins are crap that don't solve the distribution problems of bitcoin, they specifically want to emulate it for a pump n dump. Ripple, on the other hand, made no secret that they would have control. And it's been what, 1.5+ years now and bitstamp is still really the only major gateway? I may be way off on this, I haven't kept up with Ripple. I would like to think that is because of the XRP distribution, but it's probably more because the world isn't really ready yet for it. There is still ample opportunity for a new contender that will obviate the need for RippleLabs. For the future of all forms of cryptocurrency, people really need to tell centralizing schemes to fuck off.