Monday 16 September 2024, late

This past week I've been digging through the newly released Dynamicland archive, and chatting to Vic about ubiquitous computing, "pervasive media", and what that could actually mean in 2024 (or more likely, 2025).

So at the weekend I decided to see how far I could get with building a physical computing system like RealTalk. I got pretty far! There are bits of card that represent programs, which talk to each other enough that you can draw a picture with a laser pointer.

a child's drawing of a house, being drawn with a laser pointer, projected by a contraption

Some thoughts:

The whole setup is very compelling to play with. How much of that is fundamental, and how much is just novelty? Unclear, maybe unimportant.
This thing wants to be big. I managed about two meters across on my living room floor, but bigger is definitely better. Leaves me wondering what version would work in a more space-constrained environment. If this is the future of "communal computing", what does home computing look like?
It's hard to come up with ideas that aren't silly little toys. That's not necessarily a criticism — ideas are hard and toys are great — but it does feel weird. What's the obvious kick-the-tires simple project of something like this?
Putting stuff from the computer into the world is easy, the hard bit is putting stuff from the world into the computer. This makes sense because the world is big and the computer is small.

More technically:

You need your foundational components to be as robust as possible in a system like this. It's incredibly frustrating when you're trying to think about high-level interactions, and instead you have to dig back into some OpenCV code to adjust the threshold for laser pointer detection again. Getting something that works out of the box with varying hardware, light levels, surfaces, etc, is a real obstacle to people messing around with these ideas.
It probably works a lot better if you have more of a straight-on angle with the camera/projector. Unfortunately, that's a bit of a pain from a mounting perspective.
I was worried that the only projector I had was pretty bad, but actually the stumbling block is the camera and associated computer vision code. I think in particular, the webcam I'm using can't really focus at the appropriate distance.

Also, I set up a new website, but it's boring to talk about websites so I won't.

Want to leave a comment? I'd prefer an email at [email protected]