I don't support everything that comes out of the Forge, but I do support many of the key points, or at least twist them to meet my own ends.
1. Simulation As Tool, Not Goal.
This is the big one for me, and the example that The Shaman pulled from the RPG.net thread is what I'm talking about. There's a vast world of difference between trying to simulate a reality in an RPG because that's what you think you're supposed to do, and doing that because you know it's what you're supposed to do.
To take the starship fuel example, I might want a tightly focused game where the players spend 90% of their time talking to aliens, negotiating treaties, and so on. I don't really care about running scenes where the PCs fly their ship around.
OTOH, I might want the game to incorporate that because it makes for more interesting gaming. Do you press on to a primitive alien world knowing that you might not have enough fuel to make it there, but your rivals from the Klingon Empire have already sent a delegation? Do you risk being stranded at the cost of stopping the klingons from exploiting the planet?
The key is that, as a designer, I'm making a conscious choice. I add, or ignore, such rules with the intention of shaping how the game plays. I recognize that simulation is useful for extending the game and for making things easier for the players to cope with, but I also recognize that a game can't do everything. Ideally, if I choose to ignore rules for space travel, I make it very clear in the game that it doesn't support that style of play. OTOH, if I try to do a more thorough job of simulation, I don't write about how my game is designed purely to support storytelling. It isn't - there's a level of sim there that, unless you're willing to ignore rules, pushes drama behind it. The GM can't just say "You're low on fuel and have to land on this planet."
2. System Matters
This is another big one. The rules of the game shape how the game works. If a group has to make lots of house rules, then maybe the game doesn't fit what they want. They may have been better off with something else. This isn't always the case. Sometimes, only a homebrew does what you want. But, all in all, a designer should strive to build his game so that the closer the players stick to the rules, the more fun they have.
I truly and utterly hate the idea that if the game goes wrong, it's always the players fault. Could you imagine a car dealer telling you that it's always your fault if a car breaks down? Would you buy an XBox if it crashed every half-hour and Microsoft's tech support said, "It's your fault, you weren't playing the game the right way"?
I hate the idea that rules get in the way of fun even more. If that's true, and if a GM can make any game system fun, why even bother buying an RPG? Why not just find a good GM and play in all the games he runs? What's the point of even designing games? The designer's efforts mean nothing if we accept that rules don't make any difference. I've played lots of RPGs, and I can categorically say that some games are more fun than others.
3. Many Game Play Problems are Relationship Problems
If Bob's the one who always plays the character who ruins plots, attacks other PCs, willfully tries to derail interesting scenes, and can't shut up when others are trying to talk, the problem is with Bob, not his character. If Bob says, "But that's what my character would do," he's just hiding behind the game. Kick him out of your game. Don't try to use game rules to "reform" him into playing the way you want him to play. The problem isn't with Bob's character. The problem is Bob. Game rules won't make Bob into a different person.
RPGs are collaborative exercises. Even in a pure hack n' slash game, everyone is there to have fun. If someone is doing things to prevent others from having fun, kick him out of the group. If you have a friend who hates bowling, who when you go bowling does everything he can to get you kicked out the alley, would you keep inviting him to go bowling? Of course not. Same applies to RPGs.
4. Put Up or Shut Up
This ties into system matters. If your game has the same basic design paradigm as D&D, and if it features heavy sim, don't slap some prattle in the intro about how it's the "true inheritor of the shamanic story telling tradition," or some other bunk. It's a game designed to simulate something, or it's built to provide interesting challenges to the players. If it's all about storytelling, then that's what the rules should talk about. Don't just tack on some grad school reject essay about theme and expect that your game is now about storytelling, and people who play it are suddenly Real Roleplayers.
5. The Forge's Shortcomings
The main problem with the Forge is that it ties its theory together with the "indie" publishing label. I understand why Ron chose to do that, but I think that in the end it makes it harder for RPG theory to spread. Ron really doesn't like the "industry" (for good reason, IMO), so the Indie thing is pretty much there as a firewall.