[ODE] is ODE losing his 'O' ???

[Frederic Marmond]

Adam D. Moss wrote in the mail "Re: [ODE] We'll probably use ODE for an 
Augmented Reality exhibit..."

> [...]
> IMHO, probably of even more benefit and interest to the ODE
> community than your input code would be any bug-fixes or
> improvements to ODE that you might make, or any interesting
> insights into its practical application for your product.

I do agree.

By the way, is there any cleanings planned to ODE?
In my opinion, ODE, which goal was to be OPEN is not growing in this way.
I think of connecting the pure Physic engine with external libraries.

When you get ODE, you get the Physic, but also the collider and the 
graphic stuff that goes with. And it seems that the Physic part is 
developped regarding this colliding and this graphical stuff.
In my project, I use OpenSceneGraph and OPCODE with ODE, and it's bad 
that I have to make so much efforts to connect ODE with the other 
libs... OSG and OPCODE can connect very well, but, I have to say that 
ODE is less easy to integrate.

Russ created ODE for a Open Dynamic Engine, and I remember he wanted it 
to be equivalent to openGL for dynamics... Am I wrong, or are we going 
in an other way?

What do I suggest?

=> make the physic, the collider and the graphics more independants
collider and graphics are good things to see what physic can do but they 
may have their own libs

=> make more tools to integrate with other libs:
OPCODE doesn't provide the contact point for a collision. We have to 
implement our own, having the 2 colliding triangles (I'll release mine 
when it will be good enough, by the end of the next week I think)
Matrixes that can be directly used with graphs libraries (OSG, 
Performer), changing the order of columns/rows

=> concentrating attention and efforts on the physics instead of the 
cosmetic
Few months ago, someone asked for a different implementation of the 
solver, but nothing since... I know it's a hard part of maths (I'm not 
skilled enough to work on this... :-( but I think it's the kernel of 
ODE, and, well, it's not as good as it could be! (see the number of 
mails speaking of problems with world-step, number of bodies, ...) 

=> make separete documentations
When you try to find something in the doc, it always speak of the geoms, 
but it may not be handled by ODE! (OSG in my case). It's hard to see 
clearly things, as they are all gathered in only one place.
By splitting libs (physic, graph and colliding), we would be able to 
concentrate each on our competences and make things betters (independant 
code, specific docs, ...)
Did we forget the main computing law: 'divide and conquer' ?

Hope theses thoughts may make it grow up in a good way!

fred