I ran throught a lot of the posts and saw some stuff similar to what I was thinking but not quite, excuse me if I did miss someone with the same idea though:
I think one of the most interesting aspects and mechanics that Orbiter has to offer a Massive Multiplayer Online community is it's true scale. I think this mechanic, along with the physics, should be promoted as the most important aspects of what drives desicion making in the world.
I envision much more than just a space travel simulation, if you fold that simulation into a sandbox open world game with autopilots for all the noobs so they can button press to get places (and die when they point somewhere they don't have enough gas for MUAAHAHA...ahem...sorry), and a switch to turn it off and save yourself a bit of gas by being a good stick jocky, manual flight, if done properly, should still pay off.
That being said, no one wants to fly for months and months and months to get somewhere, but how about for a few weeks? what if it took you several hours to get to the moon, a few weeks in real time to get to mars? Jupiter a month perhaps, at most, with good windows observed. Any more than that, and I think you start to lose people.
The trick to all this is DeltaV, of course, and keeping it interesting in the cruising stage. You could always blow a wadd of cash on boster that has DeltaV for days and get to where you need to go in a hurry (this can be tuned to find the right ballance of scale). If you were light on credits in the virtual economy, you could always take the slow route and get to the lucerative spots in a bit more time, but a lot less expense.
The important part is the economy, you let people get places fast, but you make it cost a lot. This fuels the economy itself, resource harvesting, transportation, logistics. It also keeps people near earth while they are noobs, around other people learning, and where there is the most metabolism for a virtual economy. Earth orbit, Moon, Mars, the Belt, Juputer's moons, all of these areas will eventuall be populated by in game players as they advance in the world and mass resources.
And on the back end (IT consultant o/) it also plays into a very elegent network architecture, with a central position tracking system that tracks all objests in sol SOI in real time, and a farm with each planet (and areas such as belts) SOI having it's own server, and even going so far as splitting orbit and atmos between yet more instances. As the central system tracks you passing through each layer you would be routed to a new cloud, and because of the natural scale of the system, it will take players a lot of time to initially spread out and populate the system, allowing for a dynamic response to traffic managment.
I've got endless idea's worked up already for a game based on this very concept, orbier could very well serve as the physics engine at the core of the server side postion tracking, and I've been tinkering with the idea of getting some people together to work on the MASSIVE amount of game development the rest of it would take. I don't want to give away too much just yet, I'm acctually thinking about writing a novel in this "world" to drum up interest, both in myself and with others.
All in due time though, right?
DevCo
*Edit: BTW, I realized it sounded pretty lame to expect people to even stay flying for several hours let alone a whole month at most, but I forgot to mention how another key aspect that needs to be integrated is "stasis" where basicly you can put you and your crew into stasis (log out) and the system will propagate your position in real time so when you log back in you have progressed in your flight. Anyway, I think this is KEY to playability. Again, no time acceleration, just lots of DeltaV for a big price, and state propigation when logged off (if you hit a planet, get intercepted by pirates, or miss your entry burn and fly into space, don't cry to us, space can be a ***** like that.)