I don't see what's weird about that. Plenty of social/government systems just wouldn't be practicable without things like roads, printing press, banking, telegraph, flight, computers, etc.
I don't care about realism of social policies. Social policies were a gamey and unrealistic concept to begin with and if Firaxis wanted realism they should have kept civics. What I care about is gameplay. And making a secondary technology tree with it's seperate resource dependant on your primary tech tree is bad gameplay.
If I want lots of good social policies I should have to focus only on culture, not on a combination of culture and science.
It might would be useful to make it so that unlocking later trees would not require entirely filling out early ones.
Perhaps make it so that one only needs a certain amount of policies from earlier trees to unlock later ones.
That would be another option, but I'd prefer my version. It would give you a reason to fill out the entire policy tree instead of just cherry picking the best policies from each one.
Also, it might (not sure) be useful to keep era based requirements on the unlocking aspect of later trees (although this is a problem for Order and Autocracy).
No, that's exactly what my propositions are supposed to prevent. Era requirement is bad. I shouldn't have to worry about stupid stuff like getting a policy 3 turns before advancing to the next era.
Another change to policies I'd make is to remove the number of cities penalty for policy cost. But instead of getting total culture towards policies you'd only get average culture per city towards policies.
For example you have a capital producing 10 culture per turn and a second city producing 4 culture per turn. You accumulate (10+4)/2=7 culture per turn towards getting new policies, but the cost of the policies themselves would be the same if you were running OCC or ICS.
This has two beneficial effects. First it means that if you have a small number of cities producing lots of culture and then you expand shortly before getting a new policy, you still get the new policy faster than under the current system. That is you still gain the benefit for running a small number of high culture cities for most of the time.
The second benefit is that it would put an end to the exploit where you save culture, then sell cities to get low policy prices. This benefit would of course only apply if we could save policies.