Thanks for the feedback.

I was thinking about creating an algorithmic design tool which, when given input (number and kind of controls, profile rect, iteration level, mutation index, etc.) would automatically create a variety of geometry previews which could then generate MoI scripts which are then processed. The technique would be easier to build as a 2D depth map for a voxel modeler like 3D-Coat, but it would be much cooler to create reusable geometry directly in MoI.

Given time, I plan to continue to work on creating an app which creates a z-depth preview 2D (2.5D actually) first and see what I can come up with. Once I have something running and able to show, I may ask for some basic help porting the data, in a direct and simple way, to MoI as a POC. FWIW, a few years ago I programmed a set of CAD tools in Python for e-on's Vue software (VueTools), so I hope with a small bit of direction (and some time) I may be able to create a very basic form generation tool (fingers crossed).

FWIW, I believe the next big breakthrough in 3D is automated form generation. Too many people do not have the design nor technical skills to create functional, high quality 3D models, and with the 3D Printing revolution well under way, there is a serious void in the ability to allow 'everyman' to play in this field. Humans are much better at critiquing than creating OR put another way it's easier to select what you like from a series of choices rather than start with a blank canvas. There are attempts at this already at places like Thingiverse with OpenSCAD. Grasshopper is also cool, but at this time it focuses on more abstract than real solutions.

I've often thought (and spoken a couple times to anyone who would listen) about using genetic algorithms and neural nets to help generate lots of high quality design choices-- where the algorithms eventually become design personalities based upon a subject matter experts and some judicious knowledge engineering. I was in fact talking about this to some board members from the Foundry today and as an aside, did mention to them to check out MoI and how Michael has done such a marvelous job at hiding the hundreds of possible commands and icons from users and instead giving them an elegant and extremely usable interface with unsurpassed learning curve and surprising power. The Foundry's products are all but impossible to 'casually use'-- which isn't a huge problem now as their current market is mostly professional.

I believe the ability for users to be able to 'casually' use 3D will become more and more important as skills like 3D printing along with product and idea visualization allow for a broader skill set base for average folks. The CGI and VFX industry is only just so large-- and not really growing. MoI, Keyshot and SketchUp are a few of the products with a super fast learning ramp time along with retentionability (???) --> the ability for one to come back to a tool months later and remember how to get something done. Both criteria, IMHO, are most important for success in the next generation of 3D software.