I'm getting too specific, but my premise for designing a system is that each class ought to be as powerful as every other class is, inherently.
I think this is one of the main problems a project like this will face; different people have very different ideas about what constitutes an optimized set of rules.
Bauglir thinks that each class ought to be as powerful as every other class, inherently. I think pursuing that is both quixotic and ill-advised, and that a far better target is "Every player has as much opportunity for fun as every other player."
Actually, I agree with this. In my opinion, though, having equally powerful classes is key to making sure that each player has the same amount of fun. I don't want Fighters disappointed because they're obsolete, and I don't want Wizards annoyed that enemies either steamroll them or are cakewalks (pretty much the only two options with an optimized caster).
Take Ars Magica, for instance. They dealt with the problem of wizards being more powerful by embracing it. "Wizards are more powerful. They're the stars of the campaign, and the game revolves around them."
Here's the thing, though: EVERY player plays the wizard...sometimes. Every player has a magus character. Every player also has a companion--someone highly skilled in his own right, but not a wizard. Then there are a pool of "Grogs"--spear-carriers, basically--who are held in common.
Who is playing which role rotates. Sometimes you'll be playing your wizard, and the spotlight will be on you; sometimes, you'll be playing a grog, and very definitely in a supporting role.
Now: would dynamic character balance of this sort work for D&D? Probably not.
That's a solution, but I find it a poor one because it boils down to "everyone gets to have fun, but not all at the same time", if I understand you correctly. But yeah, I agree that's better than "The wizard gets to have fun, and everyone else might as well not bother to show up."
Instead, I think it would be wise to recognize that a lot of what was thrown out from first and second edition SHOULDN'T have been thrown out. Fighters were NOT irrelevant in first edition. Why not? Because no other character class could do their job. Clerics didn't have Divine Power. Wizards couldn't simply Polymorph into vastly-more-powerful frontline fighters.
3.0 and 3.5 embraced the "Characters should be able to do anything" philosophy. Add optimization, and the result is, "One character can do EVERYTHING."
That's where the game falls apart, plain and simple; there's no inter-party dependency.
I don't know 1st and 2nd, but I agree COMPLETELY. Each
character should have limits. Players should have as few as possible when building a character. I think what I said might be somewhat misinterpreted, because I definitely don't think Fighters and Wizards should be the same. They should be able to contribute equally to any given adventure, on average, but since adventures so often involve fighting things with class levels, it's important to keep class vs class balance in mind. But the Wizard's spells should have effects comparable to what a Fighter can do with a sword in terms of usefulness; not necessarily in terms of what is actually DONE.
One alternative to straight class balance (that I could get behind, anyway), for those who want classes better than other classes, would be a sort of circular hierarchy. Wizard > Fighter > Rogue > Wizard, for instance. That way each character has an advantage in certain situations, and by and large these advantages should balance out over the course of an adventure.
I'd also mention that I dislike the "certain classes are better at certain levels" because of the unlikeliness of 1-20 campaigns. In real campaigns, some of those balancing areas won't come up, so it's not a satisfactory balancer in my mind.
My List, in no particular order:
1. Polymorph and virtually all effects based on changing shape.
2. Different rates of power increase among classes.
3. CR
4. LA/Racial HD
5. Magic Items/Economy
6. Sunder, also other combat maneuvers
7. Grapple (gets its own entry)
8. Epic levels/Spellcasting
9. Metamagic
10. Certain tropes are poorly executed (1-handed fighter, sword and board, evoker (relative to other casters), etc)