[ODE] Geist3D, ODE and Open-source

jstier@cs.uvic.ca jstier at cs.uvic.ca
Wed Apr 12 00:10:10 MST 2006


> 
> My 2c of advice as someone that has a fulltime company that works almost 
> exclusively in open source development and has half a dozen different 
> projects out there under varying licenses, and have been doing so for 
> the past 5 years (and closer to 10 for personal projects).
> 
> An open source license will not do this for you. In fact, quite the 
> opposite. Basically it puts everything out there and people can do with 
> it what they want. The only way to make money on a codebase is either to 
> sell services around it as an open source product, or to stay 
> proprietary and license the codebase to others. One of the definitions 
> of open source is that you cannot maintain control in any way. A user 
> must be free to do whatever they like with the code, without having to 
> report to a single central entity. For example, one of the points of the 
> GPL is that changes are made, but only need to be sent along with the 
> code that gets redistributed downstream - they're not required to be 
> sent back to the original author.
> 
> Even going with GPL, you won't make any money from the codebase 
> directly. Where you may make money (and it's far from guaranteed) is 
> that end users will pay you to make bug fixes or to add improvements to 
> it. Most of the time all you will receive are bug fixes from people. 
> Unless you've got a huge userbase already (measured in the 10s of 
> thousands), donations will not form any meaningful level of income 
> stream. Occasionally, but far less so for a game engine, someone will 
> pay you to make an application (in your case, a game) over the top of 
> the base codebase.
> 

Hmm,

certail food for thought!

See, my problem is that I don't see how to make money of the code base anyways,
because the tool is not done yet. And, there so still much to 
do that I cannot finish myself. It's not yet a game engine, for example. It
needs new types of scene tree nodes that contain 2D GUI widgets like buttons
and sliders. It also needs a character node. I have already developed the
cylindrical ODE avatar.... anyways you get the point. There is still much
missing. 

One option is that I drum up some VC funding for a grandiose plan and then hire
a fixed set of developers who may, or may not, be motivated. If the endeavor
succeeds then financial success is probably guaranteed as well. In this case, I
get to eat...

The other option is to open-source, and hope that other people will pick up some
of the development work. I already had 1200 downloads in the first week and I
think that more interest could be generated. The advantage is that Geist3D is a
complete IDE (once everything works), so hobbyists and especially University
groups may be inclinded to use it and inevidably contribute. In the mean time,
I can go out, and this the hard part, and find anyone who needs simulations
done. Hopefully with my knowledge of computer graphics and know-how about how
Geist3D works, I can be more competitve. And in this case, I also get to eat
... but I think this case is more achievable :-)

If anyone else finds a way to make money then they will have to make their code
available because distributing the code includes installing it even on a single
client site. By releasing their code others then improve the tool and generate
more interest. Check out "open scene graph" ... worked for that guy :-). 

does anyone share this my view ?

Cheers
Jochen 

----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.


More information about the ODE mailing list