Can a website feel like a fidget toy?

I have a box full of fidget toys.

Some of them are beautiful. Some looked incredibly satisfying when I bought them. I was convinced I’d love the way they felt.

And then I actually held them.

Sometimes the texture isn’t quite right. Sometimes the resistance feels strange, or the sound isn’t what I expected. There’s nothing objectively wrong with them. They just don’t do anything for me.

And then there are the others. The ones I keep picking up without thinking. Maybe it’s a particular texture, a tiny clicking sound, or the way something compresses and springs back. I can’t always explain why I like them so much.

I just do.

I’ve been thinking about that feeling a lot while building Synfield.

The difference between looking and feeling

Synfield is a collection of twenty interactive worlds made of light, particles, movement and sound. You can explore them with a mouse, a touchscreen or your hands in front of a camera.

When I started building it, I knew I wanted something visually beautiful. But I didn’t want people to simply look at it and think, oh, that’s pretty.

I wanted something closer to the feeling of picking up a fidget toy.

That slightly hypnotic experience of doing something, watching it respond, and wanting to do it again. Not because you’re trying to complete anything. Not because you’ve been given a task. Just because the interaction itself feels good.

The tricky part is that I don’t think you can fully predict that feeling.

You can imagine an interaction, design it, program it, and make it look beautiful. And then you try it and realise it doesn’t have that something.

It works. It looks nice.

But you don’t particularly want to keep playing with it.

When technically correct isn’t enough

This happened quite a few times while I was developing Synfield.

I’d have an idea for a world, spend time building it, and eventually get it working more or less as I’d imagined.

And then I’d play with it.

Sometimes it was interesting for a few seconds. Sometimes it looked lovely. But something was missing.

In another kind of project, I might have kept pushing. Tried to improve it, optimise it, find a way to make the original idea work because I’d already invested time in it.

With Synfield, I gradually got better at letting those ideas go.

If something didn’t feel right, I could try something else. Sometimes a much simpler idea ended up being far more satisfying than the elaborate one I’d been chasing.

There wasn’t a specification to fulfil or a particular result I needed to deliver.

I could simply follow what interested me.

And occasionally, that meant discovering something I hadn’t been looking for.

A stress ball made of particles

Nebula is probably my favourite example.

It’s a world of particles that react to your movements. You can draw them together, compress them, release them and watch them respond.

At some point, while experimenting with the interaction, I realised it reminded me a little of squeezing a stress ball.

Not literally, obviously. There’s no rubber, no physical resistance, no texture against your skin.

But there’s something about the rhythm of it. The compression, the release, the way the particles move when you let go.

I hadn’t started out trying to make a digital stress ball. I’m not even sure I’d realised a website could evoke that kind of sensation.

Once I noticed it, though, I wanted to make that part of the interaction more satisfying.

And that’s something I found myself doing repeatedly across Synfield: playing with a world until I stopped thinking about whether it worked and started enjoying what happened when I interacted with it.

That became one of my most useful, if rather unscientific, design criteria.

Would I keep playing with this if I hadn’t made it?

Of course, that’s entirely subjective. The interaction I find satisfying might do absolutely nothing for someone else. Just like the fidget toys in my box.

But Synfield was never meant to provide the same experience to everyone.

The awkward business of explaining play

There’s a complication when you make something interactive, though.

People need to know that they can interact with it.

With a mouse or touchscreen, that’s relatively straightforward. Move, touch, drag, and something responds. There’s a good chance you’ll start experimenting without needing much explanation.

The camera is different.

Synfield lets you interact using hand gestures in the air. You can move your hands, pinch and release, and manipulate things without physically touching a screen.

For me, that’s one of the most interesting parts of the experience. There’s something peculiar about moving your hand through empty space and seeing a world respond as if you were touching it.

But how is someone supposed to know which gestures are possible?

Without any guidance, some of those interactions could remain undiscovered. Almost like Easter eggs, except you don’t necessarily know there’s anything to find.

So I needed instructions.

Which presented another problem: instructions aren’t particularly immersive.

Synfield is dark and deliberately minimal. I wanted entering a world to feel like stepping into its own little space, not opening a piece of software and reading a manual.

Initially, help was just a menu. Eventually, I started working on visual hints that could appear during a first visit, using gesture icons alongside short explanations. If you dismiss them, they stay out of your way, although you can bring them back when you need them.

I wanted guidance to feel like part of the experience rather than something pasted on top of it.

I’m still not completely happy with the result.

On mobile, especially, every additional interface element takes away a little of the space available to the world. The interactions can feel wonderfully toy-like on a small touchscreen, but the help elements compete with that feeling.

I’ve wondered about making the gesture icons smaller and animating them so they could communicate more with less text.

Maybe that would work better. I haven’t solved it yet.

And I haven’t conducted formal usability testing, so I can’t claim that the current solution is intuitive for everyone. Much of the process has involved trying to imagine what it feels like to arrive without knowing anything about the project.

Which is surprisingly difficult when you’ve built the whole thing yourself.

You don’t have to discover everything

There’s another question hiding behind all these design decisions.

What if someone never discovers the camera controls?

What if they explore two worlds with their mouse, move some particles around for a while and leave?

Would they have missed the experience?

I don’t think so.

Obviously, I’d love everyone to discover all twenty worlds and every possible interaction. If we’re making wishes, why not?

But that would contradict the whole reason I made Synfield.

There are no levels, scores, objectives or correct ways to explore it. It would be strange to remove those expectations and then quietly replace them with an expectation that everyone should discover everything.

Some people will want instructions. Others will want to figure things out themselves. Some will spend ages exploring a single world. Others might try a few interactions and decide they’ve had enough.

All of those are valid ways to experience it.

I think the challenge is to make the possibilities discoverable without turning discovery into an obligation.

And perhaps that’s why I keep coming back to the comparison with fidget toys.

You don’t pick one up because you have something to accomplish. You pick it up because something about it makes you curious. You squeeze it, spin it, click it, or turn it over in your hand.

If it feels right, you do it again.

I don’t know if a website can ever quite reproduce that feeling. A screen can’t give you the texture of an object, and waving your hand through empty air isn’t the same as holding something.

But after spending so much time playing with particles, light and movement, I think it can get surprisingly close.

And sometimes, when I open Nebula to check something and realise I’ve spent several minutes just squeezing and releasing those particles for no particular reason, I think that’s probably what I was trying to make all along.