Some years ago I designed this exercise for a Pair Programming simulation using the media of the game of chess. The idea was to have a pair on each side of the board, they would work together to finish a famous game against their opponents (a pair also).
I never used the exercise because the developers I was working with didn't know the game of chess, and didn't seem interested in the simulation. If you use this please let me know how it works out.
Garry Kasparov (White) vs X3D Fritz Computer (Black)
Man-Machine World Chess Championship 2003
Game 1 (after 31 ... Bxa2)
White to move
Instructor material for Chess Game Pairing Exercise
Chess board and pieces (one per 4 students)
Diagram of an in-progress chess game(s) (attached)
Introduce the Exercise
Pair programming is generally thought of as an XP (extreme programming) practice. In true Agile form we are borrowing a practice that has value to use in our process. Pair programming is when two (or sometimes more) developers work on a task simultaneously at the same computer.
This exercise is designed to give you the feel of pair programming. You will pair up with another person and become a chess playing savant team! Each team will be pitted against another team to solve a world class chess problem. Setup the chess board in the exact position described on the problem sheet. Then playing as a pair (perhaps with limitations – see variations below).
The Exercise – Part One: Setup the board
Instruct the students to retrieve a chess set and the sheet describing the chess problem.
Their task is to setup the chess board as diagrammed.
Exercise – Part Two: Let the games begin!
The instructor can give limitations or not, constraints, rules, etc.
Then they are to begin – White moves first!
- Only allow the pair that is in play to talk, the pair not moving must stay silent.
- No talking and strict turn taking for first 5 moves, then instructor announces that communication and cooperation may take place for the rest of the moves.
- One minute time limit per move.
- Pair beginners with experts – have the group self rate from 1 – 10.
Debrief the Exercise
The only reason to do an exercise is to get to the debrief stage. This is where students will apply the simulation to the “real world” and learning will happen. Please give at least 20 minutes for the debriefing phase.
Facilitator will start the debriefing – ask open-ended questions such as:
- What happened during the exercise?
- What did you notice about the act of pairing in a traditionally solo activity?
- Which activity was harder, more fun, required more concentration, stimulated creativity, produced more interesting results?
- What do you like/dislike about pairing?
- How did you decide to what moves to make – did you talk? Did you test out move?
- Did you find that one partner lead and the other followed, did the lead change?
- How does this simulation compare to the task of programming?
- What lessons can you learn?
The debrief should take 10 – 30 minutes (if the students don’t have anything to discuss, then it was not very instructive).
Major points of the exercise:
Pairing can be more fun, more creative. Problem solving is creative and one person’s idea can spark better ideas in another person. To effectively pair one must take turns, and communicate their ideas and intentions, they must actively involve the partner.
Bobby Fischer (white) vs Greenblatt Computer (black)
1977 Computer Match
White to move