This game was originally invented in 1956 and improved in the 70s by Robert Abbott, but I came across it this week through the game review site Shut Up and Sit Down. When I watched that video, I immediately recognized it as a cousin to both The Dice Game and Zendo, because of its use of inductive logic and modelling the scientific method.
The video by SU&SD explains the rules thoroughly enough, so I won’t repeat that here, but the general idea is this: The dealer invents a secret rule for what makes a valid sequence of playing cards. A simple example might be that you must alternate odd and even cards. Players then take turns playing a card from their hands onto a shared sequence in the table, and the dealer will tell them whether it is valid or not. If it is, play continues with the next player. If not, the card is moved aside (still next to the position it would have been, so people can refer back to the mistake) and the player suffers a penalty of picking up more cards. The goal is to figure out the rule and be the first to be able to empty your hand.
This quote from the video review really highlights how the game captures one of my favourite elements of bug-hunting:
It genuinely captures that theme of scientific experiments. Every card you play is called an “experiment” so you say “I’m going to make an experiment that does this” but when you really start to get your head around it, you deliberately play cards you don’t think it is to eliminate possibilities.
Making experiments, running tests, varying one thing at a time to explore the parameter space. All that fun stuff.
Where it differs most from Zendo is that you don’t end the game by guessing the rule correctly. Rather, it adds elements of outsmarting your opponents, a countdown timer, gambling, and pushing your luck. All of this seems to make it much more “game-like”, and probably quite entertaining to play. It will be interesting to see how much that distracts from the core “testing” mechanic, though. I quite liked Zendo as a workshop on testing because it was so easily themed as a testing activity. I don’t know if Eleusis will stand up quite the same way, though I certainly intend to try.
Also on the point of theme: In Eleusis the players are God, a Prophet, and Scientists conducting experiments. To re-theme for testers, God could be the software itself and the scientists become testers running tests, but I don’t know what role the Prophet could be. Maybe the test suite? It’s a bit more of a stretch. Though this tidbit from the review can definitely be read as the situation where you suspect a particular kind of bug but you don’t have the resources or tools needed to test your theory:
That hand of cards you’ve got is really quite limiting. You might find yourself wanting to do certain kinds of scientific experiments to deduce God’s rule, but actually you just don’t have the numbers for it.
There’s also a simplified version of the game called Eleusis Express that was designed for classrooms. This might be the version I take to my local community of practice. The express version does away with the prophet, gambling, and countdown, and instead allows ending the round when someone successfully guesses the rule. This makes it much closer to a direct analogue of Zendo, though the mechanics of when people can guess are slightly different.
One thing both versions of Eleusis miss is requiring that the dealer provide counter examples to disprove guesses. That makes me think it might still have a bit of the antagonistic air that I disliked about The Dice Game, since the “God” role is never faced with a challenge themselves.
Though I haven’t had a chance to play it yet, it’s clear to me that this should be in the repertoire of anybody who enjoys games about testing. I’m looking forward to trying it out with my local testers. Once again for reference, the full version of the game is described on YouTube here, and the Express version rules are here.