Usage: Move the mouse around in the flash window to rotate around the model as it assembles. 

Directions: Click and drag on the edge of a part to alter its length. Click on a part to set the camera to view it, select ’3d view’ to return to the orbiting view. Change the ‘tooth size’ slider to alter the tooth size/number of teeth per face. Export is coming!

(For images documenting the process that has come to this point, check out my Flickr page)

Housifier involves alternating between engineering problems and User Experience problems*. The engineering refers to the rules that govern the actual shape that has to be constructed – considering material properties. measurements and thicknesses. It’s new and pretty interesting territory for me. After years of considering pixels and clock cycles it’s refreshing to contribute to the process of making stuff. Thankfully, the right answers are relatively clear -a press-fit join either holds or it doesn’t.

The third part,  the User Experience, is a lot less clear-cut. There are principles to apply and yardsticks to measure actions by, but trying to anticipate what a system that people are going to find clear, powerful and entertaining is a difficult process. Similar to the concept of the Minimum Viable Product being the cheapest thing to test, I’m trying to focus on the Minimum Viable Cognitive Load – the smallest required engagement from a user that will result in satisfying usage of the application. The one thing I am constantly surprised by is how little attention people want to give to something in order to understand it!

My current goal is to remove the need for sliders or other excessively numerical forms input from the design to produce a box. From there I’ll be integrating with the Ponoko API so that orders can actually be made through the system. After that I’m going to look at options for surface detailing- engraving patterns and cutting holes, though there are some technical hurdles there that might require a jump into Unity3D. From there we should be ready to get into the extrusion of any old shape, and you’ll finally have your dollhouse!

* And programming problems, which are by no means trivial but place a distant third after the engineering and UX considerations.

I wasn’t sure what to start with though – something too easy and you won’t learn anything about the nature of the medium  - cutting width, material tolerances, that sort of thing.  On the other hand, though, if you reach too far  you won’t end up learning anything either – if you try to design an ornate interlocking castle without some appreciation of how things actually work you’ll end up with an elaborate mess.  So I tried to strike a balance somewhere in the middle.  I tried to design a remote control car with rack-and-pinion steering (though I didn’t know what that was until I looked it up either)

The first thing to consider with laser cutting things is that you only get flat sheets and holes carved into a single dimension – so when you have a mechanism where two parts have to be hinged at right angles, you’ll need to make that hinge  happen at right angles to both of those surfaces.  I’m not an engineer so it’s a novel experience trying to figure these things out. Once I had an overall idea of how the car was going to fit toegether, I could start doing more precise drawing.

Vector graphics applications are a wonderful thing.  Using Illustrator, I was able to draw the shapes I wanted, measure them up against one another and fit them on to the specified sizes.  I was hoping to keep the thing as small as possible, so it’s on an 18x18cm board with a lot of room left over!  I used some of that room to try and figure out how a bearing assembly might work made out of cut board.

Next I had to wait. Ponoko say they’ll clear their queue of designs inside 28 days, and mine was done within 11 so I was pretty happy with that.

Then, it arrived! Actually I picked it up but it was fantastic to see.  Unfortunately at the time of first unwrapping the sheet with all 114(!) parts laid out together I didn’t have a camera with me, so that moment is lost to history.  Once I did manage to get everything assembled, though, it looked like this:

Unfortunately once it was assembled, it turned out that the motors I had planned to use were too big, and the drive axle for the rear wheels was too short.  The bearings worked out better than expected, though, and at about a third of a millimetre, the beam width will be easy to account for in future designs.  I’m looking forward to prototype #2! Between this and the Arduino micro-controller, I think it’ll be possible to do some pretty fun stuff.