Seventy tickets in a weekend

On 20 February, version 0.7.0 of Scopa Fortuna was playable from start to finish. It was also full of small wrong things: a button overlapping another on mobile, a round score always at zero, a special card losing its artwork after being played, text cut off in the guide. Nothing blocking. All of it irritating.

I was playing on my phone, on the sofa, and had no desire to stop at every flaw to open an editor.

What I believed

That review had to happen at the desk, in front of the code, fixing as I went. That is how I had always seen it done: test, find, fix, retest. A short loop, but one that interrupts the game every two minutes and never lets you see the game the way a player sees it.

What it was

I ended up doing the opposite. Play without fixing anything, and write everything down.

Voice notes, dictated while playing: “the Mago’s plus-one upgrade forces a direct capture when it should leave the choice”, “La Torre does not replay”, “a King with plus one should capture for eleven”. And screenshots, systematically, screen by screen: thirty-one on mobile in portrait, twenty-nine on desktop in landscape.

Then I handed all of it to the tool with one instruction: turn it into self-contained tickets, understandable without context, each with the observed behaviour and the expected behaviour, ranked by severity. Out came seventy, in four files, numbered BUG, IMP, MOY, MIN.

### BUG-03 · The +1 upgrade does not raise the capture value above 10
Observed: the capture value stays at 10 despite the upgrade.
Expected: a King +1 captures for 11, so it can take 9 + 2.

On 23 February those tickets were worked through in three sprints within the day: nine bugs, ten improvements, then the rest. Seven commits. Version 0.7.1 shipped that evening, 0.7.3 the next day, and the game went to itch.io.

What I take from it

A well-written ticket has always been the currency between the person who sees the problem and the person who fixes it. What changed is that the person fixing no longer needs things explained twice, and does not get tired by the seventieth. The cost of fixing went down; the cost of describing well did not. It has even become the main item.

The part I had not anticipated is what the method does to the game itself. Playing for an hour without stopping is the only way to feel that an act is too easy or that a shop comes too early. Those remarks are in no ticket. Yet they changed more than the other seventy.

Play without fixing, then fix without playing. Doing both at once, you do neither well.