Hi there people! I wanted to first address Mr. Simon's observation that he doesn't see GNS as something important for game design. Two games that I feel have a very clear creative agenda written into their core are Burning Wheel and Pendragon. Now, I admit I haven't played much of either yet (unfortunately), but still, here is how I see them.

Burning Wheel seems to have Story Now written at the core of its mechanics, thanks to Artha. Artha is a resource a character can earn during the game. In order to earn Artha, a BW character must fight for what he believes. Only by driving toward his beliefs (statements chosen by the player about what his character believes) can a player earn Artha. In doing so, however, the player causes the situation to get more and more complicated. Eventually, the players will probably manage to solve their situation, thanks to the Artha (which can be used to earn very useful bonuses in tests). Solving the situation, will provide more Artha, change the player's beliefs in someway and, thanks to the resolution systems, leave plenty of loose threads. Thus Artha is always driving the game on. And since the way to earn Artha is basically to have a good Story Now game, the creative agenda is written in the very core of the game.

Pendragon likewise has, at its core, a Right to Dream system that helps using Arthurian Legend tropes to drive the story. In Pendragon there are these attributes called Traits and Passions. These attributes are numeric values which can be used in a variety of ways, including being tested to determine how a character will act, being tested against each other to simulate inner conflict, being tested against the traits of another character to solve disputes of will, etc. These attributes can be tested during play to gain heavy bonuses, but they can provide heavy disadvantages too. But the reason Traits and Passions drive the game toward a Right to Dream agenda is because they are only somewhat under the player's control. The way these attributes rise, fall and are tested may be influenced by the player, but are more or less determined by how Mr. Stafford felt an Arthurian tale should go. Thus the game will, ideally, take the players ideas and input without deviating from a predetermined core, the passions of knights that drive Arthurian tales.

By the way, an interesting fact is that, while I am not sure about the first example, at least Pendragon was made way before the Mr. Edwards wrote his essay on the GNS modes of play. I think the creative agenda has always influenced RPG design, even before it was formalized by that essay. So I think that the GNS is very important in determining a game's design, but it can do so without the designer actually knowing about it. It is the ideas behind them that are important, not the name they are given.