There is a specific pleasure in a hard problem. It has edges. You can see the whole of it, hold it, turn it over, and know that somewhere inside it is an answer that will either work or not. The feedback is clean. The test passes or it fails. The system holds or it falls over. Nobody argues about whether you were right, because the problem itself decides, and it decides in your favor or against you without malice. You emerge tired and certain you did something. The day has a shape.
I love that feeling. I chase it. And I have started to notice what it lets me avoid.
The hard problem is bounded, and the bounded thing is the easy one. Not easy to solve — easy to inhabit. It asks something difficult of my skill and nothing at all of my nerve. It will never require me to sit with a question that has no test suite. It will never ask me what the last six months actually did to me, or whether the thing I keep calling a strategy is really just a flinch I have organized into a plan. Those questions have no edges. They do not pass or fail. They just wait.
So I pour the hours into the problem that answers back. And here is the part I have been slow to admit: the problem did not ask for all those hours. I gave them. The real work fit in a smaller box than the one I built around it. What filled the rest was not the problem's demand. It was my preference for its company over the company of everything unresolved.
Busyness reads as necessity, which is why it works so well as a hiding place. If someone had asked, on any of those days, why I was still at it, I had an honest-sounding answer ready. The problem was hard. Hard problems take time. Both true, and both a way of not saying the other true thing, which is that I would rather solve something clean than face something open.
The tell is the overshoot. When a problem could have closed at four and I kept it open until eight, the extra four hours were not engineering. They were avoidance wearing engineering's clothes. I know the difference now because I have felt both, and they do not feel the same from the inside. One is pursuit. The other is refuge.
I am not arguing against the hard problems. Solve them. They are the work, and they are also genuinely good — the concentration, the closure, the small clean proof that you can still do the thing. Keep all of that.
But it does not need all of you. That is the whole correction. The hard problem will happily expand to fill whatever you feed it, and it will look, the entire time, like it needed feeding. It didn't. Solve it in the box it actually occupies, and then — this is the hard part, harder than the problem — walk out of the box and toward the question that has no answer waiting, the one you have been outrunning by staying busy with the one that does.
That is what the writing is for. And it is the first thing I drop. Every slammed stretch, the writing is the cadence that goes quiet — not because I decided it was expendable, but because it is the only thing on the list that does not shout. The problem shouts. The problem passes or fails by end of day. The writing just waits, and waiting things lose to shouting things every time. So I cut it, and for a while nothing seems worse. Then the thinking starts to thin. I notice I am reacting more and understanding less, moving fast through days I could not describe if you asked. The practice was not a luxury I trimmed. It was the thing keeping the rest legible, and I cut it at the exact moment I most needed it — because it is the only place I go where nothing passes or fails, where the problem does not decide for me, where I have to sit in the open and figure out what I actually lived.
The bounded problem is a mercy. The unbounded one is the work. Stop using the first to hide from the second.