AbdulAlhazred
Legend
Right, so my take is that it would have been virtually impossible to alter D&D MORE than 4e did and still credibly package it as an edition of D&D! I mean, starting from the most basic skeletal outline of 3.x was not based on a desire to 'keep what is good and iterate', it was based on a desire to remain within the overall structure and genre of D&D itself. They basically had 2 choices here, they could take d20 and strip it to the bone and build from there, or they could go back to 'classic pre-WotC' (IE 2e or something approximating it) and start there. They really didn't have any other options from a business/product point of view.Hang on. I'm saying it is part of a chain (or really chains) of iteration, with roots in 3.5e. As much in reaction to 3.5e.
I did not and do not say that the game remains the same. Iterative design processes very often result in leaps - substantial innovations - while at the same time having their roots in their predecessors. It's product design 101. Identify problems for your audience (I think we can agree that 3e did a fantastic job of providing a rich supply of those!) Divergently ideate solutions. Where possible, prototype your favoured solutions. Analyse other work in your context for problems you might have missed, concepts you might learn from, and preexisting "prototypes" of ideas you are interested in. Define your criteria for what = good. Using your criteria, converge to a set of solutions - the solution space - that you will then develop. Try to get your core gameplay rapidly to the table, and iterate from there (test, learn, adapt.)
Identify problems > Ideate (diverge) > Define good > Solutions (converge) > Develop.
Typically, the solution space makes its own demands and offers its own opportunities. Certainly some concepts came - as they rightly should! - in from the wider context. Your list is not dichotomous with my claims, it's orthogonal: your list and my claims are both justified. Nothing there "pushes back", which is not to say that the process didn't also bring in elements that did not identify their problem spaces in 3e. That should be anticipated: it's the process working as intended. New things ought to be learned as the solutions are developed.
So, yes, I agree with your assessment in that we can agree that 3e was at least an attempt to evolve D&D beyond its classic design and create a more solid structure around combat (mostly) as well as an attempt at codifying skills and moving them into the core of the game (IE using them to implement things like thief abilities). Note that 4e basically ditched that last one and started over with skills, which is a pretty major thing, actually! So, 4e does, in some sense "build on that" but: IMHO 4e represents what happens when you say "well, our goal is to make a game which serves an agenda derived from modern ideas about RPGs, but we have to build on a chassis that is recognizably derived from some edition of D&D." Had Rob, Mike, etc. been simply given carte blanc to make any old RPG they wanted which was in the heroic fantasy genre, I very strongly doubt they would have built it on anything that resembles 3e at all. In fact I would have thought that a game more closely resembling Hero Quest/Hero Wars would have been more likely.