It sounds somewhat like a game my uncle used to have, around that time, called Sundog. You had a ship with lots of components, some of which started out broken and you had to find/buy/steal parts to fix it so that you could travel around shipping goods. There was some combat (both in space and on land), and stuff was always breaking.
GoodOmens
I bow down to Neil Stephenson; "Snow Crash" is great fun and intriguing social commentary, "Cryptonomicon" is a vast work about information theory, and his "Baroque Cycle" is incredibly ambitious (not to my taste, though). I've just started reading "Ilium" by Dan Simmons and love it, the reimagining of "god powers" as nanotech is wonderful. I plan to read his other works also.
It's true that the galactic resources are powerful, but remember that to use them effectively you need to devote alot of time and money to starbases, which are vunerable and expensive.
The GalCiv 2 buid system does lack a certain flavor. It's flexible, intuitive, and stylish, but it doesn't feel "sci-fi" enough. A nice addition would be customizable weapon names; it wouldn't require any change in AI tactics or rebalancing, but it would add some extra personality.
I generally don't even try for the trade goods, except occasionally the hull plating (just to make sure the Drengin or Yor don't get it). They're nice, but I prefer spending my resources elsewhere.
Using multiple starbases on a single planet can be very powerful, but remember that those extra starbases get EXPENSIVE after a while. You have to decide whether it makes more sense to have a few really great planets or to spread your bonuses out more. That said, a high-PQ planet with a bunch of economic starbases can be insanely productive.
No, but you can purchase the manufacturing center to speed it up. That's the problem with most methods of ground attacks; this game's system seems to encourage destroying a village in order to save it.
It does seem that the frequency of habitable worlds is less than in the first GalCiv, but that's based on very limited evidence at this point.