Okay. First, thanks for keeping the discussion going. I was worried this wasn't going to get off the ground. I wanted to examine some of the ideas, from my point of view, maybe get some further discussion going. Please note I'm trying to look at a lot of this from the point of view of a coordinator first, and then a writer second. Thus, I understand what I think is a bad idea, might not be to some of you.
As a note, the ideas I covered were ones people said "I would do this" to. If you like the other ideas, I encourage you to speak up.
Horror Romantic Comedy
+Best opportunity for group building within the writing forum. Not only would we be sharing characters, we would be tying them into relationships of all kinds with one another. The writers would have to know all the characters, and thus be on working terms with their creators.
-Weakest chance of crossover when put into context of the forum. Paradoxically, lends itself the best to a verbatim translation to webcomic, but beyond that doesn't have the hooks the other projects do.
+Best option for character-on-character writing, which I know is a big draw for a lot of the writers here. Both genres involve a lot of interpersonal reaction to a variety of emotions. If you want to have a character building exercise in the literary sense of things, then this would be the best option.
-Speaking of character on character, I'd like to point out that this project would also mean the writers knowing each other's personal boundaries as our characters intentionally misconstrue and push theirs. Some people don't post on the smut forum, some people get freaked out by gore or themes of horror. We'd have to learn to play to eachother's strengths or the project would flounder.
+However, assuming we could get that covered, this would probably be the easiest project to get done. A small timeline and a cast who has active reasons to be with each other means that if everyone plugs a couple chapters in, then we can easily get the whole thing done in no time.
-Hard to write sequels for. Both genres we would be borrowing from tend to be very negative toward direct sequels. Even slasher flicks tend to not have anything beyond the premise and big bad in common, and if a character returns they're typically there to be killed off. RomComs almost universally end with a big, wrapped up "and then they lived happily ever after". We would have to get most everything we could out of these characters in this context.
Country-Building Exercise
+Strongest chance of crossover. Artwork of characters, locations, and historic events for the country could easily be brought up in the art forum, the various artifacts, manses, or armies could be designed in the crunch forum, the setting itself could easily be used for the RP or the webcomic. This is truly the project where the sky is the limit.
-Strongest need for canon police. Let's be honest, this one would be a pain the ass because we'd have to make sure all accounts matched up on a certain level. This isn't to say people couldn't do AUs or wildly subjective in-character vanity pieces, but on a certain level we would have to underline what was and wasn't true and keep everyone on the same page. This only gets harder the more we open up the setting.
+Most fluid in terms of what you can do with it. Writing could be day-in-the-life accounts, historical records, or standard fiction. It could be set nearly anywhere in the timeline for the project, from the first age to past the current point in canon. This isn't a crossover project in as much as it'd be a shared universe.
-Technically near impossible to ever actually finish. We'd have to likely decide a specific event in the country's history and write around that.
Strategy Game
+Good opportunity for crossover with the other forums, especially the roleplaying and crunch forums.
-Requires an insane amount of front-end work. Generals and lieutenants must be designed, and then their armies, including fluff details, specialized equipment, tactics, history, and so forth. Mechanical means of handling things must also be determined: are we using Mass Combat or Mandate of Heaven rules? One of the boardgames? Another board game (we could just play Risk or Axis and Allies and change details)? How will the game be run? If we're determining which side wins first, then we need to decide details of battles and exactly who wins and what makes a good story.
+Beyond the massive front end, this project needs the least coordination. If we approach this from the notion of using a mechanical basis, then we have randomness to determine the outcome of the battles. If we instead decide which factions will win over which ahead of time, then this is just pure writing, plain and simple.
+Very easy to revisit! Armies have this tendency of being regrouped and good generals know when to make their escape and start anew.
TFS Visual Novel
+Minimal need for canon policing. Given that most visual novels tend to have multiple paths (typically based on romantic choices, though in some of the better ones it's just by plot decisions), every author could do their own thing and go in wildly different directions.
-Might be tricky to coordinate. Obviously not every story path could branch directly from the initial jump-off points, people would have to work together. Sure, this is fine for the first two or three people who get the initial branches, but when you're branching off a branch of a branch, you might have had some constraints put on your creativity.
+Strong group-building. Considering we're having to build characters from the point of view of everyone writing them, it means that we're essentially doing design by consensus.
-Your character becomes common property of the whole project. If we go with the idea that everyone would be writing a different story path, then this means accepting your character can't be a badass (or at least the badass) in all routes.
+Potential crossover with the art forum and the rest of the forum at large. If this is popular enough, we can put out an open casting call for people on TFS to voice characters and draw scenes, and code the thing together with one of the free, open-source visual novel programs.
-Does not lend itself to revisiting. Most sequels for visual novels are spiritual, because at best your other options are just more story paths, or declaring one ending as the actual ending (which invariably means some bad blood with the other writers). In short, if we do this option, there's really no chance of ever revisiting these characters.