Planning a GDK

Sorry long stretch of not doing much with Axiom or checking the forums.
I am still planning, but the GDK team kind of fizzled out, so not much progress.
Spending time now helping building the pieces that are missing, such as DotScene loader.

Basically is getting a core game engine together, dotscene loader, physics, etc
alot of which Bostich is thank fully porting like a man possessed by the one ring

And building an IDE to put all the pieces together, since Express does not support Add-ins
I plan to use some of the Framework from MonoDevelop
i.e plugin frame work so the plugins will work with MonoDevelop (hoping that the differences between MonoDevelop and VS are not too big)
some of the plugins that I am hoping to build are:
Scene Editor
Material editor
Particle editor
Media Manager
etc



but that's all a little way off without a GDK team.
So still planing to do it just scaling back and focusing on the parts I can do alone.
theMouse

Remember to take it step-by-step. Good luck with the project, and let me know if you need help. I wouldn't be devoted, but I could definitely take some time to check it out.

BTW, one of the things I liked about OGRE was that it didn't force a certain addin, such as physics on you. The user should be able to create a plugin for his own physics engine so that anyone can add new systems.

3) (personal) I don't like the in-game editor approach. I'm a coder and I like my ide's

I also don't like in-game editor approach. GDK's are always in between the frustrating line of 3d-rendering-engines and the-game-itself.
If developers are semi-professioanal, they mostly prefer to collaborate a rendering engine, a physics engine, a sound library, etc. by their own. Because usually it is not the most challenging side of the coin. But semi-beginner game developers even can't manage a GDK's usage and they mostly prefer -modding- the games they have played before.

For a 1-3 developer based indie game, there is too much code to do even using a GDK.

So GDK's always having the risk of popularity, and many of them wasted for that reason.

My advice would be weird but:

Make it very simple and single-genre-targeted at the first phase. (rts, side-scroller, fps, hack&slash, shoot'em up, adventure, boardgame, puzzle, etc.)
then try to encapsulate it for other genres step by step. Develop separate game/level editors for each genre types. Allow a good scripting support at the background (something like Lua). At last develop a server that could offer hosting multiplayer action for all games developed by your GDK (like Steam). I didn't seen any GDK that offer a good front-end for all genre types and such a project would achieve big success.

If anyone would interest this kind of approach, i would really want to join in.

I was thinking one step removed from that actually.
The IDE are just tools to edit the components of the GDK.

The actual games would be VS & Mono Develop starter-kits.

Basic process,
Create game solution via starter kit.
Open Solution,
Open IDE to edit items. (or just use VS/Mono Develop Plugins if you are not using express.