Txurce
Deity
I've strongly considered this. The only thing I'm concerned about is it's unrealistic to have some of the autocracy/order policies in the Renaissance. Would this bother people much, or does gameplay outweigh realism here?
If a policy is clearly "modern," then it would bother me in terms of realism. This could arguably be addressed by putting those policies at the end of a relatively late-blooming tree. I'm more concerned about benefit inflation, which has steadily increased across pretty much all areas. Making more trees available earlier once again makes the game easier. (We disagree on the value of more choices vs tough choices.) If more trees were going to be available earlier, then I agree that specializing them for more specific (rather than general) purpose is the way to go. If you also weakened policies across the board, then the added choice wouldn't necessarily make the game easier. But that is an awful lot of reconceiving just to make more policies easier to get.
As has been mentioned elsewhere, the overall problem largely comes with our endless buffing spirals. If things are too easy, you have to reduce efficiency. There's nothing wrong with some stuff being less useful than others, as long as it's not worthless. I prefer the base game being harder, so that choosing an improvement, SP or GW delivers a satisfying result. That's why I liked the idea of changing the Village yields, for example. But you could just as easily go about trimming the fat from the buffing of weaker improvements, and many of the secondary +1's of this or that now dotting the GW landscape.