How to win a 16-bot vibrating Battle Royale
What we learned building good/bad robots at Experiment Studio.
On Thursday last week we held our 9th Experiment Studio at Fluxx HQ. It was fantastic. As well as the usual benefits of these meetups (free pizza, free beer, new people, etc.), we also had a fantastic and enlightening talk by Ross Atkin (who was recently featured on the BBC’s Big Life Fix, where he designed a machine that helped a man who had suffered a massive stroke tell Ross to fuck off in two distinct ways) and we got to build some cute little robots.
I’ve never built a robot before. The one above is the winner of the ‘Best Looking Robot’ competition, created by our very own Stefano Bellucci. We have an array of the robots that were entered into the ‘Best Looking’ contest on a shelf in the office, and visually they range from adorable:
By the end of the night I’d had a lot of laughs, and learned a few things.
1. Failure takes work
It’s fun to watch things fail, but they have to fail in pleasing ways. The robot kits Ross brought with him were Crafty Robots, and they are fantastically stripped-back. There’s not much more to them than a vibrating motor with a USB connector to charge it. You plug it in for 30 seconds, and it vibrates for roughly the same length of time.
That means they’re reasonably predictable in their flailing, and they fail in pretty interesting ways. If you enter the sumo contest your robot might just run out of the ring entirely. On the ‘escape the circle’ challenge (where the robot has to, well, escape a circle on a sheet of paper), it might just spin hopelessly in circles.
I can’t imagine how much work went into having this thing fail in such pleasing ways. They could catch on fire, which would be obviously wrong. They could completely fail to charge, which would be deeply annoying.
They don’t do either of these.
They could vibrate totally erratically, like a washing machine with a brick inside on a trampoline; this would be funny in itself, but removes the interaction from it and turns it into an unpredictable hunk of digital flotsam.
They don’t, though.
They all fail perfectly, doing just about what you want but with a pleasant enough sense of randomness that you can anthropomorphise them as being silly instead of annoying, and with just enough feedback that you can feel like you’re getting closer to good on each iteration.
2. The ‘Ikea effect’ is real
The winner of the Sumo competition, where these little vibrating robots had to push each other out of a circle on a sheet of paper, was this little guy. He’s called Mad Max.
I don’t know how clearly you can see Mad Max in this picture, but he’s made of rubbish. His torso is made of a blue paper cup we were using to serve beer in, his legs are straws, and his head is a ping-pong ball.
By all accounts he’s not a very good robot. He’s never going to replace anybody’s labour. He’s not particularly mobile, and in fact he’s not even going to move in the direction you tell him to. But he’s still loveable.
To Flow, his creator, I’m sure he is even more lovely.
The reason for this is the Ikea effect.
The Ikea Effect is part of the reason my house looks like it does, and why I bring bits of cheap self-assembly Swedish furniture I hacked together four years ago that cost next-to-nothing across the country when I move.
We value the things we create way more than things other people have made — having a little bit of interaction in the process of making a thing invests us into it, and changes how we assign value to it. We’d prefer to keep a coffee table that we built ourselves than get a new one that’s more expensive and higher quality.
Same with robots. We’d rather keep hold of and play with this:
even though the latter is clearly higher quality and more complex. This might sound a little frivolous but it can have massive effects on how you should design and market products.
You can win a sumo contest really easily if you buy the second robot. It’s mega-heavy and very durable.
You’ll enjoy it a lot more if you make your own Mad Max though.
3. Googly eyes are brilliant and you should get some for your office
Experiment Studio is a meetup where we welcome designers, developers, researchers, engineers, marketers, product managers, behavioural scientists, and many more, to play with the future. What unites us is an interest in getting things done in a way that is fast, iterative, and evidence-based.
We’ve had speakers who’ve worked with The Guardian, the Parliamentary Digital Service, The Times, and the Royal Society of the Arts, among others.
We’re always looking for speakers and activities; if you’ve got an interesting activity or have worked on an amazing project, get in touch!