# swapcoding: a competitive way to build software

this is probably a terrible idea but here it goes. suppose there are two expert coders and they want to compare their skillz and learn from each other. what can they do?

i propose the following game. i named it as swapcoding. the duo gets together and agrees to build some piece of software. something they both have the necessary skills to build without much research. e.g. maybe it's a terminal based tic-tac-toe game or a sudoku solver written in html + javascript. something that can be implemented with 1-2 hours independently by both.

both people implement the challenge themselves. but they will work on it in a turn based manner in multiple rounds. the "swap" part of name comes from this aspect: that they swap places during development.

each round starts with a small discussion where they agree which feature they will both implement in the next round. then the first person starts working on implementing the agreed feature while the other person just watches. during this time the second person cannot touch their laptop. there's a chess timer counting the time used. once the first person finishes their feature, they press the timer and now the time is counting for the second person. the second person starts working and now the first person watches. then another discussion happens and then the next round starts. each player builds on top of their solution in the subsequent next round. these rounds continue until some total timeout is reached (e.g. 4 hours).

example rounds for the terminal based tic-tac-toe's case:

• round 1: set up the coding environment, implement hello world.
• round 2: implement tic-tac-toe board drawing.
• round 3: implement two player game.
• round 4: implement ai.
• round 5: add mouse support.
• round 6: implement timers and print statistics for the 2 player games.
• round 7: add game controller support.
• round 8: ...

once the game is over the duo compares the end results. if both solutions are comparable then they can look at their time usage from the chess clock to determine the winner. but the game is not really about winning. for me it would be about watching how people approach developing code. e.g. how much they rely on autocompletion? is it slowing them down or does it give them an advantage? the fact that i'm also implementing the same problem puts me into the right mindset to really appreciate the specific problem's nuances to fully understand what the other person is doing. simply watching others without this extra competitive pressure on me would be boring to me.

i never really tried this game so there's a lot of details to iron out. but maybe some day i come across someone who really wants to play this game and then i can determine whether this is fun or not.

edit: see @/codingtime for a suggestion on how to measure time.

published on 2022-06-20, last modified on 2022-08-21

to the frontpage