The things I watch on repeat (and what they did to my work)

The way you answer when you haven't actually sat with the question yet. I said something about reading and watching movies, which is true and also says nothing.

It kept bothering me for weeks, because the honest answer is stranger than that and I knew it while I was giving the short one.

So I took my time and wrote the version I wish I had given.

Electronics

This is the only one where I really use my hands. I read, I try things, I get a circuit to work, mostly by trial and error until something finally lights up.

The first time I burned an LED I didn't understand what I had done wrong. I had power, I had a component rated for less than that power, and I had skipped the small boring part in between. The resistor is not a suggestion. Nothing about the circuit argued with me or gave me a second opinion. It just released a bit of smoke and stopped being an LED.

I like it because it doesn't care about my intention. It works or it doesn't, and it tells me right away.

That's a very different kind of feedback than design, where something can be quietly wrong for months before anybody says a word about it. A button in the wrong place doesn't smoke. It just costs people small amounts of attention, thousands of times, and none of those people write to tell you.

So electronics is where I go to be corrected quickly. It has recalibrated how long I'm willing to wait before I find out whether something works, and now I build the shortest version of a thing first, just to see it fail early.

Animation, and watching the same thing many times

I watch and read a lot about how animation gets made, and about the principles underneath it. Disney and Pixar breaking down a single scene. Concept artists showing how a character went from a rough sketch to a final render.

Here is the part that sounds strange when I say it out loud: I watch the same things many times.

The first time I only see how it turned out. The second time I start noticing that there was a real decision in there, that two options existed and somebody chose one of them. The third time I notice the thing they left out, which is usually more interesting than anything they kept.

Nobody stops the video to explain why one option was better. You have to catch it inside the small choices, and I never catch it on the first pass.

The one that changed the most for me was spacing. Animators don't really draw movement, they draw where a thing is at each moment, and the distance between those positions is the whole feeling. Even spacing reads as a machine. Spacing that opens up as it falls reads as weight.

Once you have seen it, you can't unsee it in interfaces. A menu that slides at a constant rate feels like it was dragged by something. The same menu with the same duration, easing out, feels like it arrived on its own.

Physics and math, and the part where things feel off

I go back to this without any particular goal. Just returning to concepts I had already seen once and reading them slowly, until they settle differently than they did the first time.

There was no plan behind it. It was closer to curiosity than to studying. But it turned out to be the most useful of all of them, and here is the specific reason.

Take the simplest thing there is: something thrown, and falling. Horizontally it moves at a steady rate. Vertically it does not, because gravity is an acceleration, so the distance it falls grows with the square of the time. Half of g, times t squared. That single t squared is the difference between an arc that reads as a real object and an arc that reads as a sticker being moved across a screen.

If you draw that same arc but you space the positions evenly along the path, every number in the drawing is still correct. The shape is right. The start and the end are right. It still feels wrong.

Nobody looking at the left one can tell you what is wrong with it. They will say it feels floaty, or cheap, or off. Off is the word people reach for when their eye has caught an error their vocabulary hasn't.

And that is what I find interesting. You already know this. You have never once computed a trajectory, and you would still notice if a ball hung at the top of its arc a fraction too long. Every person who looks at your work is carrying a physics engine they didn't install and can't turn off.

The same thing happens with light. Two objects in the same scene with shadows falling in different directions will bother people who could not tell you what a light source is. It happens with perspective, where lines that should meet at one point meet at two, and the room reads as slightly haunted. It happens with scale, with reflection angles, with how a heavy thing starts moving compared to a light one.

None of these break a layout. They break belief, which is worse and much harder to debug, because the bug report is a feeling.

What happens when all of that leaks into the work

Now when I need something to move or bend or distort on a screen, I go look up how it actually behaves in the real world instead of guessing until it feels acceptable.

Right now I'm building a site for a hackathon and I wanted it to have a portal in it. The science fiction kind.

So I've spent an unreasonable amount of time making sure the distortion behaves like distortion, that the curvature makes sense, and that anything sitting inside it actually has depth.

A lens does not smear what is behind it at random. It bends light by an amount that depends on how far from the centre the light passes, so straight lines stay continuous while they curve, and the density of everything changes as it goes through. Getting that one rule right is most of the effect.

And it is not one drawing. It is a stack: the scene behind, the depth that makes the scene behind read as far away, the distortion the opening applies, and only then the thing you actually see. Every one of those layers can be individually correct and still fight the one under it.

Which is a little funny, because a portal doesn't exist. There's no reference anywhere. I can't compare my version against anything.

But I think that's the point. You already know how light bends and how things fall, even though nobody sat you down to teach you. So an invented object can still feel wrong the moment it breaks something you already understood.

That is the trade. You can invent anything you like, as long as it is consistent with the rules the viewer never agreed to learn.

Cases of graphic design that ended up mattering

I collect these. There's one I keep returning to because of how small the mistake was compared to what it caused.

In the year 2000, the supervisor of elections in Palm Beach County designed what became known as the butterfly ballot. The candidates were split across two facing pages, with a single column of holes running down the middle.

Her intention was good. She made everything larger so older voters could actually read it.

What happened instead is that thousands of people who wanted to vote for Al Gore punched the hole belonging to Pat Buchanan, a candidate who had spent years being accused of antisemitism. Another 5,304 ballots ended up marked for both men at once, which meant they counted for nobody.

Gore lost Florida by 537 votes. A study published by The New York Times decades later called it a slam dunk that the layout of that page swung the presidency.

It is probably the most expensive spacing decision in modern history, and it didn't come from carelessness. It came from someone trying to help.

I keep it close because it is the cleanest counterargument to the idea that this work is decoration. Nobody misread that ballot because it was ugly. They misread it because the thing that told them which hole belonged to which name was position, and position was ambiguous.

The other reason I keep it is the part that gets skipped. The designer tested it. She showed it to people. It went out with approval from more than one office. Whatever review process existed did not have a way of asking the only question that mattered, which was whether a person in a hurry could tell which hole was theirs.

Games, where the interface has to disappear

The other place I look at constantly is videogames, because a game is the clearest example of an interface carrying an enormous amount of information while trying very hard to stay invisible.

In something like StarCraft the player is tracking the minimap, watching where the enemy is moving, keeping count of how many units they have left, and noticing whether they're running out of minerals or gas.

All of that has to arrive without the player ever consciously stopping to read the screen. The moment they have to think about the interface, they've already lost the fight.

What I take from games is that most of that information is not delivered as text. It arrives as position, as colour that means one specific thing every time, as a sound you learn in the first ten minutes and never consciously listen for again. The screen is not explaining. It is training you, and then relying on the training.

Games are also honest about the cost of getting it wrong in a way that software rarely is. A confusing menu in a product produces a slightly worse quarter. A confusing health bar produces a player who dies and knows exactly who to blame.

So, the actual answer

None of these are really hobbies about making things. Almost all of them are about watching other people decide.

And then understanding the rules well enough to know which ones I'm allowed to break, and why breaking them was worth it. The animation people figured this out a long time ago with squash and stretch, anticipation and exaggeration. None of it is accurate. All of it looks more real than the accurate version.

That is not a contradiction with the physics. A ball that squashes on impact is wrong about the material and right about the impact, and the eye is grading the impact. You need the rule to know which part of it you are trading away, and what you expect to get back.

I don't think taste is something you're born holding, and I've stopped telling that story about myself. Whatever I have came from quiet hours paying attention to choices nobody bothered to explain out loud.

Now I want the other side of it. Tell me what you do, and which hobby of yours ended up changing the way you work without you noticing at the time.