I think a first person RPG similar to Lands of Lore (LOL) or Double Dungeons (DD) might be interesting and achievable in a few months time frame. While both games are immersive, LOL has a more advanced storyline, level/character artwork, and GUI. DD, with very little of the aforementioned, still draws the player in and is more casual; the controls are very simple.

I've developed a quick method in creating levels for these types of games, which are essentially "mazes" filled with enemies, items, and loot. A maze can easily be translated from a sheet of graph paper into 3D Gamestudio. I'm thinking that anyone within the Gamestudio community who wants to contribute can make a maze or dungeon on a piece of graph paper, scan it , save it as a .jpg and upload it to a central web page that can perhaps be created for this project. From there, someone(s) assigned to translate the jpegs can enter the info into Wed and add it to the overall project. I would be willing to do something like this.

George, Thank you for proposing this initiative; I'm sure our community will grow stronger as we work together. It should be interesting to see what comes of it in months ahead. Good luck to everyone involved.

I think a single player puzzle solving 1st/3rd person game is a good mini project to see if this team is able to create anything.

The models created for this project can be used in RPG/FPS like games if the style fits.

To start with an RPG you would need (technically) inventory, text engine, ... which is complicated enought
But more important: a great story and a story teller who is able to transfer the mood, to keep the player in the game...

On topic:
I'm also against voting in a poll now: we need more talking first, there hasn't been enough precise propositions, just brainstorming mainly.

Ok, here's my (final?) idea:
Small arenas in which the Player battles AI with a few characters he can choose and equip (with a drag-and-drop style equipment system). After a quick initial deployment phase, you start controling your units (around 5, depending on how you equipped them, they have different abilities) and using cover to avoid enemy's attacks.
+ Small arenas would mean little art work, so we can focus on code, a nice GUI and a nice style.
+ As slin said, anything big is unrealistic. I think this type of game would give us the opportunity to try mutliplayer; Two people could try coding multiplayer-code for this game, and if that fails then there's still the actual game that works.
+ Arenas could get created randomly, meaning more code and less art work and more fun.
+ no need for a huge long story line, no need for big explanations, the game would be small. no need for many different models.
- we would need to settle on a style. I really don't care what it would be but I'd like "cute". Sort of like the warsheep from ICQ: they think themselves to be real cool...
- balancing and the item-equipment is a little hard in this one.

Reference:
http://www.4warriors.de/shots/shot7.jpg (four little warriors)
http://www.nokia4me.ru/uploads/posts/2009-10/1255857805_bezimerprni-1.jpg (icq-warsheep)

LarryLaffer: here you go.
The demo can now be downloaded from Peter Soxberger's site:
http://www.anet-plugin.com/downloads_eng.htm (shooter example)
Please also read the forum thread (link on the site, just under the download). There's some mistake I made when writing the readme, I mention it somewhere in the thread, I don't remember what it was though.
Note that this is no good for a shooter. It's a simple demo thrown together in a day or two. I am against making a shooter because a) there's too many and b) if done in multiplayer, it would only be frustrating, as people have said. Precise prediction would be needed which is totally unrealistic for a few indies who have their own projects to work on.

Some years ago I was working with Rhuarc on a project with lots of people from all around the world.
We used dotProject as management tool.

It is a great way to list all assets we need for the game (models, textures, sounds, functions) and assign them to team members. If someone leaves the team - and this will happen - we can see which assets are still open and needs to be assigned to someone else.

Together with subversion and other tools it would be possible to build the required infrastructure for an international online project.

I am happy to see that so many talented people want to get involved in this project. Let's keep the ideas flowing for a few more days.

"How you will sort out who'll get what". It's the amount of work that makes it into the final version of the game that will help us decide that. The project will belong to its creators; if people leave they'll get a smaller share, provided that their contributions make it into the final version of the game. Right now I'm more interested in seeing us producing a game from start to finish; I think it's going to be a very useful experience for any Gamestudio user and especially for all the team members.

I wouldn't go for a big project; I'd rather have us finish something of small proportions instead of dumping an ambitious project after 6 months of work.

i still think a single player, short missions, collect items and weapoms for protection and off to the next level with a goal at the end. it would be easy to put together. even if the action was a bit arcade like with lots of enemies to avoid or attack, and then go collect items and solve a puzzle to complete the level.

if i can help with non-organic models, 2d artwork(humanoid model concept art), sound effect, voice acting, or story writing, I am in! i can probably help with animations as well.

I could help with textures, non-organic models and some level design.

Germanunkol: Thanks. I'll try setting this up when I have some time. Any chance you already have an executable out of this? It would save me some time from gathering all required files and publishing it for my friends that don't have GS installed.

I had this idea about a co-op adventure since I played Legend of Kyrandia on the Amiga. How cool would it be to have two or three people in the world trying to solve the story? Maybe in a fashion like The Lost Vikings where each person has his unique abilities. I'm not aware of any game that has done this and maybe there's an obvious reason for that that I fail to see, but this idea always intrigued me. It might be possible with GS since adventures don't require precision accuracy multiplayer, but again... multiplayer just complicates things. The whole code needs to be written to conform with the network code. Pathfinding, Movement, Ai, it will all need to go through the server if we want to reduce traffic. I really see this as a project doomed to fail.

@larry,
in obscure series, that coop characters with special abilities are used a little.

in this game, you generally use 2 characters thought the game, if you play coop each player controls one, if you play alone, one character is AI controlled and one caracter is under control of the player, player can switch active character by pressing a button.

in some parts of the game you need ceratin characters to be together.

i.e. in a puzzle there is a ventilation cover(if that's what it called) on a wall, but it is too high to reach, there is no ladder or something, you have a screwdriver and it's certain you should be going thru the ventilation duct.

To reach the cover, you need to be playing with the most muscular guy with lightest girl. Girl steps on the guys hand, guy raises her up, she uses screwdriver on the cover, then goes in. from that point girl keeps going alone. I dont remember how it continued but probably, we re sending her some items thru the duct, and by controlling her, we were trying to find some way for others to get to the other side.(probably unlocking a locked door or finding a key.)