Status Update

I thought I’d take a moment to provide some status and make a call for comments.

Apparently Wiiusecpp is the most popular/easiest to find software on my site, based on the number of visits and downloads.  I have had no significant feedback as to how well the library works,  if anybody was able to get it to work on windows, or if there should be any API tweaks/changes.  I have not really been using it since I’ve been mainly working on the spreadsheet project.  My intention was to use the wiimote to control my 6-dof simulation (see the visualpyode page) but I have just not gotten around to it.

On the pysheet front I have a much much better version of it (multiple sheets, plotting, etc.) which is nearly a complete rewrite from the pysheet I released, but since I can’t really gage the interest in it I’m not feeling compelled to rush out a release.  I’m also a bit worried that a company might co-opt it into their for-fee product.  If there is revenue to be had I could sure use it if for no other reason than to allow me to continue working on new projects (I’ve been sidetracked of late, seeking income).  So it seems I have three choices:

1) GPL it, and don’t sweat the outcome

2) Use a more restrictive license to prevent commercial use, leaving me the option

3) Try to make a business of it right from the start (free to students, fee to companies type of thing)

A vibrant and interested community (and a few other developers) would certainly persuade me towards #1 or #2, but right now I’m just not seeing it.  That leaves me tempted to either form a relationship with one of the companies I have identified or compete against them.

On the visualpyode side of things I’ve kind of reached the conclusion that visual python is the weak link there and have been contemplating how large the task would be to convert it to being OpenGL based or something like that.  Don’t get me wrong, visual python is a neat project, but their goals are really oriented towards the eliminating complexity and using it in a classroom setting. My goal of visualpyode was to allow visualized simulation without the complexity of designing a video game — do your simulation in PyODE and visualization ‘comes for free’. The visual python UI shortcomings (and no realistic hope of working around them) really hamper its use in ‘real’ applications so I think my visualpyode library falls short.  I need something that will easily render in a ‘canvas’ provided by wxPython or other serious UI library.  However, the sticking point is the same thing that brought me to visual python in the first place — mouse/object interaction.  It may not be obvious from the demos but with visualpyode you can interact with your physical objects using the mouse.  I’ve used this to have the mouse apply forces and even do construction of objects.  This requires that the graphics library be able to identify the ‘contact point’ between the mouse and the 3D scene.  This is turn-key in visual python, I’m not sure how easy it is to accomplish in something like OpenGL.

If you have any ideas, comments, or opinions you want to share please post them here, in the forum, or send me an email (jt@missioncognition.net).  If even one person would take the time to share how they are using the software or perhaps something have learned from reading the source code, it would really be meaningful to me.