On nation number:
- One nation: More or less standard Aurora game, with not much more interaction
- Two nations: Will probably end like Steve's NATO/Warsaw pact game: Outward expansion until someone gets lucky or enough people are off earth, then sieging of other systems.
- Three nations: Depending on the eventual strengths, this has the interesting result of allowing a weaker nation to form the lynchpin. As long as no nation is stronger than the other two together, there is always incentive against allying with an equal partner - that of betrayal once the enemy is defeated.
- Four nations: Similar to above, except that a complete alliance of two players might be able to steamroll one of the others. Only then, it becomes the same problem as for three nations; that is two strong and a weaker nation.
- Five or more nations: Very interestig due to more political dimension, but probably too much work for the SM.
I myself would prefer three or four nations, as I believe that would balance the risk of war with the risk of not-war. Implementing an espionage system would help making backstabbing less attractive.
On tech trading:
I believe that wholesome tech trading (p.ex. Ultraviolet lasers for Ceramic composite armour) should be allowed, but have some in-game drawbacks entailed. For example, you still need to research a certain percentage of the tech (representing integrating it into your tech base), or the tech transfer being easier to intercept via espionage.
Also, I like your idea of component trading and licensing.
On tech osmosis:
Another idea would be an automatic transfer of technological knowledge if in use or traded.
Sublight has had the rule that prerequisites of a traded technology would be revealed to everyone. Another possibility would be that, once you research something, everyone gets the tech two steps behind - for example 10cm lasers if you're using 15cms.
I myself would prefer the trade variant at most, though I dislike tech osmosis for the fact that it has the risk of making empires too similar.
On orders:
My idea would be for everyone to give standing orders that should be followed, and plans which are one-time changes to that. For example, you could have all exploratory ships follow a certain plan, all warships another, and so on. Production plans would be made a few months before the construction actually starts, and for a year or so. This would tie in with the below espionage system, would award planning and give you a clear set of things we want ships to do.
On espionage:
Now, it gets interesting. Assuming that you want a system completely outside of aurora, I propose using several areas of espionage: Military deployments, military plans, ship capabilities, research projects and industrial projects plus counter-espionage. Each of these areas can have several espionage targets. For example, ship capabilities could be a short description, or complete blue-prints.
Each of the above areas would fill up with a random number of points per turn/year/five-day increment. Whenever there are enough for one of the espionage targets, you slowly get corresponding information. These points represent the penetration in that area. The more points you have in one, the higher the chance of being detected, but the more valuable the information.
For example (and those are all numbers I've just made up on the spot): Assume that we are talking about ship capabilities, with 100 points representing a description, 250 representing solid numbers and 1000 blueprints. For 2500, you could build the ship in your own facilities.
Each five-day-increment, you gain between -2 and 7 points. So, on average, you require 100 days to slowly gain descriptions, 250 for solid numbers and 1000 for blueprints. If you have, for example 100 points, you have a chance of 5% per increment to gain a description of a random ship - so, if the enemy has twenty ship types, you'll need about five and a half year to get information on all.
On second thought, the above might be much, much too complicated.
