I liked the "drag population to different boxes" method of MOO2 and the "drag population to different map squares" method of Civ for specializing production. However, the primary drawback for each is the amount of micromanagement that it requires in large empires. Reducing late-game bloat has always been a high priority for Sven. Micromanagement can be mitigated with automation, but if the AI is good enough (which is a difficult programming task) at auto-specializing, then there's rarely any need to override it, and it becomes a largely empty feature. That's the other drawback; that it really doesn't offer all that much of a choice... the vast majority of the time you'll want to specialize 100% for one type of output, so all you'd end up doing most of the time is pressing one of the AI target buttons.
The current system in SIS is a bit simpler: you build different types of infrastructure on each planet to improve its efficiency at a certain type of output, and then you can change its production target (from shipbuilding to research or trade or growth) to specialize on the fly.
We intend to add some more complexity to the production model, so it's possible that the final system may incorporate more of the visual "drag population to specialize" model, but we do want to keep colony management as streamlined as possible.
enpi wrote:-choices, choices, choices - technology is divided in groups of three techs per group. you can choose only 1 technology per group, and the other 2 are lost. (you can get, maybe anytime later in the game, those lost techs per espionage and tech trades) This very elegant system allows races with a really unique profile which produces an instant ingame narrative by the inclusion of techs AND the exclusion from techs. It provides also great stategical choices in which direction you want to develope your empire.
As for the MOO research model in which the techs in the same group are mutually exclusive, I think you haven't seen it in other games for two main reasons. First, it doesn't make logical sense, and seems like a very artificial choice: why should researching Neutron Blasters make it impossible for you to later research Neutron Scanners? It seems like the reverse should be true: knowing one tech in a field should make the others easier to research, not harder. Second, because the techs are mutually exclusive, that means you're going to miss out on most of the techs, and therefore none can be critical for you to have, and that means there have to be far more technologies in each category so that you can be sure of getting the kinds of techs that you need. The technologies lose their uniqueness, and the number of them explodes. MOO2 has 9 types of beam weapons and 10 types of shields, all of which are virtually identical in function with exception of each being more powerful than the last.
Rather than having 10 technology tiers, SIS has 4, but we try to add interesting choices within each tier between weapons that are functionally different. In tier 2, for example, you have a choice between turbolasers, ion cannons and disruptors; each does comparable damage, but has different special characteristics: one is better against shields, another better against armor, and a third causes additional crew casualties. There are also non-beam alternatives in the same tier: kinetic weapons, missiles, and shorter-ranged plasma weapons. And they aren't mutually exclusive; it's probably not an effective use of your science resources to research all of the options, but there's no need to prevent the player from doing so if that's what he wants.