So back when I was working for This Guy, I met a really excellent venture captialist that was looking for ideas.  I came up with an idea, detailed below, and he was ready to fund it (I actually had cash in hand at one point) but the deal went sour thanks to Dr. CrazyPants.  

The project was this system of inter-locking entertainment and home automation tools.  We proposed some main components – an animatronic plush ‘character,’ like a stuffed teddy bear that could move and talk on it’s own; a set of computer controlled home automation units ( imagine a little box that sits between the wall-plug and your appliances, allowing wireless computer control) ; a cheap tablet PC that works as a wireless interface to the system; and a server that sits in a closet and runs the whole show.  We wanted to use a lot of off-the-shelf parts, and show that with a basic kit and some creative software, we could do a great deal.  

Looking back, the project was overly ambitious, and we surely would have failed.  We wanted the Teddy Bear to have a rich enough sensor suite that we could write contextually aware software, and we wanted the whole system to learn the usage patterns of the people that it lived with.  

One neat idea was a cross-over between real and virtual worlds.  For kids, we designed a software interface that acted like a stage, were virtual representations of real-world toys were the players, and interaction could work both ways – interaction developed in the software world could migrate to real-world interaction, and vice-versa.  As children grew up with the tools, the software would ‘mature’ with the child as well, starting out as a puppy or baby bear with simple behaviors, growing into adolescence and autonomously developing more complex behaviors.  We had some knowledge about neural networks, and stars in our eyes.  

We were undergrads an we didn’t really know how to do this, but we thought it would be incredibly fun to try.

From July 10, 2004:

Last night I had one of my recurring dreams. I dream that I’m building something. It’s this… structure… it’s not a physical thing… it’s a tree with leaves that are concepts and ideas, held up by branches made of dreams and memories. When I have this dream, I’m working on it in a kind of head-space. Just this side of lucid dreaming, I can conjure up whatever ‘tools’ I need. I craft a dream to a purpose, and hang it on the tree. The leaves grow by themselves. 

I remember feeling that familiar troubleshooting tension in the dream. Like I’m trying to get the tree to achieve something but I’m having trouble. I remember the feeling of satisfaction when it works. 

I don’t know what the tree is supposed to do, but it felt powerful in the dream.

I have a vision. 

Nighttime on a bluff. Overlooking a valley of trees. 
Still, cool summer air, bright stars, right after moonset. Only a few people. 

Floating lights. Hundreds, thousands of floating lights. Dancing. Music. 

I envision hundreds of lit white balloons, arranged in a grid so big it looks like a landscape from the a viewing location on a bluff.  Each balloon is a giant 3D pixel, able to move up and down, and change color by the driven light.

I imagine building audio-visual experiences, sound and color and movement, a giant musical visualizer.  I imagine walking among the construct, having it be the ceiling to a party, watching it undulate as the music passes through it. 

I first thought of this almost 5 years ago, I did some design, helium baloons on fine wire-strings with motor reeling mechanisms at each base.  Each balloon lofts a lighting package, a small microcontroller, small battery, some high-power LED’s.  The wire could be used as a one-wire communications interface to talk to the lighting package.  Each base is a larger battery,or power distribution node, another microcontroller, and the reel motor.  The whole thing is mastered by a PC running a coordination program, and a music interpreter, all written as a WinAmp visualization or something.

Individual balloon lighting schemes could include flashes, dims, slow color gradients, maybe even some kind of UV lighting.

At one point I calculated some costs, trying to figure out what it would take to put this all together, besides a small army of technicians.  It was too much.  If someone out there wants to build this, send money, or invite me to the party.

I’m starting this thing due to the fact that in my course of work, I have too many ideas, and not enough time to complete them all.  The Cult of Done dictates that I should post these ideas, thus spreading them into the world in the hopes that somebody will stumble across them and turn them into “done.”

Flickr Photos

PennSunset

TotalGlobe

Shiny

ProtoDuck

LearningToShare

More Photos