Start with the moment, not the technology
The first question we ask is not which technology you want, but what the visitor should feel. Surprise? A jolt? Pride at having made something happen?
That sounds vague and is not: the answer decides the technology. Surprise calls for something that responds bigger or faster than expected. Pride calls for clear feedback that you got it right. A jolt calls for timing on the frame.
Only once that is clear does the question become which technology makes it possible affordably.
The three layers of an interactive installation
The input: how does the installation notice what the visitor does? A button, a sensor, a touch, a movement, an object placed somewhere.
The logic: what happens with that input? Usually more than it seems. May it repeat immediately? What if two people trigger it at once? What if someone does it ten times in a row because it is fun?
The output: light, sound, picture, movement. That is the part the visitor sees and usually the smallest part of the work.
Build first, promise second
For anything we have not done before, we build a working mock-up before the quote goes out. Not a render, not a drawing: something on the table that does what it should do.
That costs a few days and saves three surprises during the build. You know whether the idea works, how it feels, and whether the pace is right. And you know what it costs, because we built it.
It is also the most honest way to say no. If the mock-up shows the idea does not work as imagined, you hear that before you have paid for an installation.
Building for the thousandth time
An installation that works perfectly once is a demo. The question is what happens at interaction ten thousand.
Anything that gets touched we build heavier than necessary: industrial components instead of hobby parts, strain relief on every cable, connectors you replace one-handed. Moving parts get position feedback, so the system knows whether something actually happened rather than merely being commanded.
And the installation starts itself. On with the building in the morning, off in the evening, and back on its own after a power cut.
Who will look after it?
This is the question most often skipped and the one that most decides whether the installation still runs in two years.
Is there staff walking past every day, or does it stand in a hall nobody visits? Should someone be able to intervene, or must it be fully self-sufficient? Who gets called when something breaks?
We build the controls around that answer: a screen for the crew, or deliberately no screen but a remote notification when something is off.
What you get handed over
Schematics, source code and credentials are yours the moment the invoice is paid. You are free to have the installation modified by whoever you like.
Beyond that we record what sits where, which components it is built from and where to order them. Even if we are not around in five years, an engineer we have never met can work with it.
Why you should get the source code →
The process in four steps
Conversation: what should the audience feel, and what is the space, the budget and the deadline.
Design: a cue list you can read out loud, a schematic your electrician understands, and a budget made of items you can recognise.
Build: cabinet, code and bench test. We build and test here, so the days on site stay short.
Handover: install, commission, brief the crew and hand over the documentation. After that we are on standby in the background.
How the process runs →
To the relevant pages
