1. "term limits" to each of the major rolesIf I were going to go with a multiple-rank system for Admirals, I think I'd probably have suggested exactly these two ideas myself. If we look at this as an economic issue of supply and demand, both term limits and having to be voted into roles help to maintain the supply of admiral roles. (Gotta love supply-side economics. :) ) My feeling is that this is usually a better way to go in MMOGs than demand-side restrictions on gameplay opportunities -- those tend to make players feel like they're being arbitrarily punished (which isn't far from the truth).
2. "voting" into each role
So these ideas -- term limits especially -- do address my question about how many players can fill a relatively smaller number of slots. I wouldn't say they solve all problems, however. For example, how long should someone have to wait for their turn on the swing? One of the standard policies in MMOG design is instant gratification, but that's clearly not how slot-based gameplay would work (even if term limits were used). Is there any other way this approach could come a little closer to allowing everybody to wants to be an Admiral to do so without having to wait for weeks or even months for their turn? (And what about that idea of "taking turns?" Is that valid within a MMORPG?)
One other note concerning voting players into roles... I have to admit, I like this idea. It really captures something that many people probably aren't aware of WRT high-level military commands, which is the intensely political nature of who gets which jobs. (I have to note here that I've never served myself; this is all the product of years of reading and observing and talking to people and generally being an interested civilian.) I don't mean "political" in the sense of political party affiliation, which would be a serious breach of military effectiveness -- what I mean is the internal wrangling that goes on, and the more general game of power-brokering within the Establishment that's responsible for nominating and approving high-level commanders.
The idea of voting for the top Admiral jobs is really attractive because it selects for those players who not only have proven track records of tactical and operational military success, but for the ones who are also good at seeing power structures and how to manipulate them... which I would say is a pretty good indicator of strategic capability. In fact, if I could be permitted one tweak to this idea of voting players into top Admiral roles, I would say that it shouldn't just be players voting -- let the (presumably NPC) "Federation Council" members vote as well!
What if the NPCs on the Council each had their own factional and policy interests? When considering applicants for a top Starfleet role, the game could compare the interests of each Council member to the service record of each applicant, generating a "yes" or "no" vote based on the closeness of fit. For example, an NPC from Betazed who favors a diplomatic solution to any conflict (whether that makes real-world sense or not) might choose to support the nomination by the Federation president of a Rear Admiral who has a history of solving problems without violence, while the member from Andoria might oppose that Admiral's nomination.
As long as player votes (i.e., votes from other Admirals) also matter, and as long as there are other factors that could be brought into play dynamically so that a character's military history alone doesn't determine for all time whether he or she gets a particular slot, I really like the idea of letting the Federation Council have some meaningful say in who gets which jobs.
On the question of actual gameplay from multiple levels of Admirals, I'm not entirely convinced (but not closed to the idea) that this is the way to go for this particular game.
First, and simplest, it's true that nobody does this stuff alone. There are support staff. The thing is, though, how would this work in a game where the problem is usually not that there's too much to do, but the lack of content?
Second is the question of overlap. If regional commanders, and sector commanders, and even some departmental commanders are all in charge of resource distribution, how would conflicts be resolved... and how could that be an activity that gamers at this level of play would consider to be fun? If as a sector commander I notice that the Bad Guys seem to be making inroads in Zone 11, and I order the USS Corbomite there to patrol that area, what happens to my success rating if you, my regional commander, can divert the Corbomite off to some other sector you think presents a more pressing challenge?
Is that realistic? Sure -- diversion of resources happens all the time in any enterprise, and managers just have to figure out how to get the job done anyway. But is it fun as gameplay?
At this point, I'm thinking we can see the ends of a spectrum with respect to Admiral-level play. On one end of the spectrum is seeing Admirals as basically Ensigns-on-steroids -- they play exactly the same game as Ensigns; they're just "higher-level" and so have more character abilities for doing the same stuff as the rawest Ensign. Most MMORPGs do a little better than this... but not much. I would very definitely hate to see Star Trek Online go this way in its design.
Your suggestions for multiple Admiral levels, and for various rules for how players can navigate between these levels, comes in at the other end of the spectrum. I think a lot of us would be happy to see ST:O's developers implement anything like this; the question is how likely they are to do so.
Something tells me the reality of Admiral-level play will be somewhere between these two possibilities. I don't know where, exactly; it might be the one-Admiral-rank, very strategic concept I've floated, or it might be closer to the other end, where admirals are basically just guild leaders. Or perhaps it splits the difference somehow between these approaches.
The question here is, should we be hoping that the developers try to split the difference? Or should we hope they go with a full multiple-Admiral-rank system? If the latter, what would be the best arguments we could marshal to help encourage them in that direction?