I think the way we should limit the number of these things that get built is by making them really expensive, not imposing a slot system
Maybe by a high maintenance cost?
When talking about building, I think we should define what kind of building we are talking about. So far, there is:
1. infracture building: things like nanobots enhanced soil, holothingy, facories etc.
2. ships: military (or supports) units who can move from system to system.
3. wonders: ala civ wonder, a building that has a defined effect on a planet/system/empire/universe, usually boosting capabilities or allowing new capabilities
4. wonder ship: ala DeathStar
5. defence type building
Here are my ideas about those:
1. nobody really wants to micromanaged this (as in a Sid Meier game, you always build the latest when you can and it is profitable). I like very much the way Stars! abstracted this with factories and mine: those two thing were the only buildable that affected production, and they were just a number (as in 588 mines and 1100 factories on a planet). To me, SM developments were only an increase in a stat (ah a bank, 50% more money) and it was really annoying to build each and every one of them (or trusting some kind of domestic advisor).
2. they are the reason people don't want 3, as en empire could pump all its production at any controlled planet, effectively making a productive empire everywhere at once, military wise.
3. a fun thing about wonders in SM games is that there is some kind of race to completion. If a large empire can divert all its ressources for a 2-3 turns to complete anything, there isn't much race, only the will to compromise empire development for a wonder.
4. there should be suspens and danger associated with such undertaking. Since they hold such great strategic importance, they could be dealt separately from any wonder.
5. like drek already said, they should be dealt with when we are ready to do the combat thing.
A lot of people seem to be uncomfortable with post-placement because things "pop up" out of nowhere. To me, this amounts to no more than a thinly-disguised realism argument. I don't think it matters whether things pop up out of nowhere or not. It's a video game. Things always pop up out of nowhere.
and not just for the megaship argument. If I find a colony to near of my frontier (or even worst, in the middle of my empire) that is well defended (so I won't just stomp it), I want to starve it from ressources from the rest of its empire, not it being able to pump out ships from its whole empire.
It's not about the reality argument, it is about how production is modeled. I always thought 4X games modeled country/world production: you move ressources where it's needed, things are produced where it can/is the most efficient, etc. An empirewide production, as in proposition 3, is like a perfect electrical network: anything produced anywhere on the network can be used anywhere on the network, without any loss. Why are we having system distant if it isn't to create and maintain an efficient production network? Not having that would be like dodging an interesting feature.
With the country/world model, if bandits/pirates/enemies/whatever are not strong enugh to take you head on, they can flay your support network, isolate and conquer some parts of you empire and eventually do something else than being weaker, if you are careless and don't protect the innards of your own empire.