So there, I finally got it out of my system. Some people (especially my architect friends) tend to suggest that what I do is too virtual, and I'm really starting to understand what they mean by that. I've actually built a physical tensegrity, and it's great fun to dribble it on the table. When you tap on the top it starts to bounce back and if you do that rhythmically you can move the tower around, bouncing it like a basketball. Good fun.

It sure ain't easy for a coder like me to get something physical built! I have a number of friends who are really good at building physical things, but I'm definitely not one of them. Typically when I build things the result makes me wish I had hired somebody to do it instead. But this time I've apparently outdone myself. Even the architects are impressed.

It's this virtual world that I find so much easier to build in, and one of the main reasons is that the virtual model avoids "tangles", since things just pass through each other. Not so in the real world, working as I did with pieces of wood, and little nylon rings and cable. I had to develop the step-by-step techniques to build this thing without getting things all tangled up and I can tell you that it took a few iterations!

The way I approached building this tensegrity was pretty unique, I think. My virtual model (seen here at the right) was exactly what I used as a basis. I set it up so that I could read the exact lengths of all the cables, recorded them, studied how it all fit together by looking at the structure in stereo (set it up in OpenGL, with two eyes, two points of view, then you see the depth). Then I mass-produced the parts of the tensegrity, cutting things precisely to the right measurements. After that I built individual three-stick modules using the pre-cut nylon rope assemblies, and when I had a bunch of modules (alternating right-left-handed) I was able to quickly assemble the tower by making six connections where each module joins the next. The whole thing took only a few hours, once I had figured out the system, and failed a couple of times.

Anyway, I did it, and I'm really proud of it. I can pick this thing up and put it into peoples' hands so they really know what tensegrity feels like.

Click on the picture for a big version.

My friend Hans Moor would describe the whole process I went through building this as a kind of "evolution" involving many steps of "selection" to get to the end result.

Now that I've built it I'm itching to get back to building evolution software. About time to put out another podcast as well.

Don't forget that there's a new Greythumb gathering set up, and it will be happening in the Royal Academy for the Visual Arts in The Hague on the 23rd of September. We have at least three speakers, so don't miss it!

When: Sept 23rd, 18:00-21:30

Where: The Academy