1. I don't see what's so interesting about "a low productivity city that has lower happiness requirements but still gives you territory," but regardless, that is how I would describe puppets in their present state. What the equation was before isn't particularly relevant to me.
In their present state, they aren't really worth the happiness, because they still cause almost as much unhappiness as a regular city.
If they caused less unhappiness, then they might still have a role.
What the equation was before isn't particularly relevant to me.
If you didn't think they were too weak before, then changes that nerfed them should presumably want to make you want to compensate for that in some way, no?
I am worried that if puppeting is too weak, then we go back to a situation where razing the city is better, or we end up with a situation where the rewards for warmongering are too low.
recently you said city spamming is now back (a nerf to tall empires), and then today say Liberty's happiness has been nerfed too hard (a nerf to wide empires)
I don't think I said city spamming was back (if I did, then I probably mis-spoke, and should correct my impression), I think I said that the balance in TBC has shifted away from tall empires towards wide ones. I think this is correct, and a bit unfortunate. I dislike the general shift from +X% modifiers towards +X modifiers from buildings.
and then today say Liberty's happiness has been nerfed too hard (a nerf to wide empires)
Yes, I think my initial impression on Meritocracy may have been incorrect, given all the other happiness changes. I don't like forcing happiness onto Wonders and social policies, but if this the design we are going with, then they need to give significant happiness.
Changes should not be made in a piecemeal fashion
when you posed a general set of changes ..., I opposed it on general principle
So, you oppose both general and specific changes I suggest on general principles. Nice.
But if we were to, for example, approach the SP's one tree at a time, and pause to discuss the overall effect proposed changes may have on the game, then I would be more comfortable.
Why? What is so special about the individual tree structure?
Why is the relative value of policies within a tree more important than the relative values of policies across the trees?
You are not forced to pick policies only from 1 tree at a time, you can play across trees. I don't see any reason why the tree is the appropriate scale for changes.
I prioritize the order of doing changes by:
How important it is.
How easy it is to implement.
This sounds like a good approach to me. I think there is room for a lot of gradual tweaking and changing things; you test something out, if it works, you keep it, if it doesn't, you change it back.
and 
to near-zero for a while until we can readjust. Conquest has the advantage of a larger, more experienced, but lower-tech army (military over tech).
hit at very-unhappy status. A warmonger has all the advantages of both quantity and quality (better military and tech).
Each vanilla patch adopts things from the mod, so the two become steadily closer and less different over time. I haven't made any fundamental gameplay changes in several months. It's vanilla that's changing, and here's a list of the ways vanilla has become more similar to the mod. You can see the devs are actually including things from the mod at an accelerating pace, with more in recent patches:
