See, it's this "one true way" philosophy that really irks me about you. I was providing a single counterexample that proved your premise false. Now, if you want to go ahead and start using logic like a competent person, then by all means, prove me wrong. But do it with logic, and not fallacies.
I will address your points now. If they can be called points. Your "complainfest".
1: see above. A single counterexample was enough to prove your absolute premise false. It doesn't matter how extreme. I could have gone with a level 10 rogue fighting level 10 enemies, but then I'd have to go through and make an entire freaking campaign to generalize. And yes, it's possible for this sort of thing to happen. I've seen it. You won't believe me that I've played in a campaign where a stealth character mattered, but it happened. It was actually one of my gestalt ideas that I played, a Pixie monk//rogue. Hit like a truck, fun as anything else I've played (except Doctor Ooze, that was my favorite character ever). Mattered to the campaign, never seen by an enemy when I didn't want to be. Even the BBEG, who was a high-level demon/caster. I think Balor, but I can't remember, it was a long time ago.
2: you keep saying the same things over and over again, never proving anything you say. You say a Rogue can't beat a caster. We ask "why?" and you say "angel summoner vs BMX bandit!" or whatever.
3: I have no idea what you're even talking about. I'm going to assume right here that you're claiming that a trap is universal (not good in any campaign). In which case you're right. Now will you understand that just about everything you're calling a trap is not in fact a trap? It's only a trap if it's viable in NO campaigns. This means even campaigns designed to handle them. I've shown that a stealth character is viable in at least two campaigns (there's actually a whole range of them), so I've proved that it is not a trap.
4: If a campaign is supposedly designed to support a specific character concept (i.e. stealth), and the character concept cannot succeed because of encounter design, then it is indeed the DM's fault that the concept is failing. If the DM is not doing such things (for the stealth example, if the DM is restricting access to sensory-defeating options while throwing only opponents with extra-sensory methods of perfect detection, including but not limited to insane perception checks, blindsight, and touchsight) and the concept is STILL failing, then you look at the player, and if that's not the problem, then, and only then, can you begin to claim that something is a trap.
So basically, I'm saying that this thread continues to be worthless. Here's a guide that might actually be worthwhile: a list and discussion of options that fail in campaigns designed to support them. The list would be incredibly small, actually. I'm thinking Truenamer, and.....actually, that's about all I can think of. This supports an incredibly wide range of campaign power levels and styles. You absolutely HAVE to take this into account, because the styles of campaigns vary so incredibly wildly.