Scoring

[Limpan]

It would be possible to score the users performance.

[Greenheart]

Decreasing reward

This builds on the idea that puzzles have a difficulty stored as a number between 1 and 100

• When a user starts a challenge, a variable with the reward is set to the puzzle difficulty's value (Ex. 73).
• For each failed try, the reward is decrease by some amount.
  • Maybe decrease the reward by larger amounts when the reward is high
• The reward can't get smaller than let's say 1 or 5 out of 100.
• When the user submits a valid solution, they get the reward.
  • But only if they haven't already completed the challenge
  • Maybe store an user's completed challenges as an dictionary with the challengeIds as keys and received_reward as values
    • User.completed_challenges = {"challenge_id": received_reward}
    • Ex: {"5": 18} - this user has completed challenge 5 and got 18 points.
    • Also keep a User.total_score to not have to compute this over and over if users at some point get hundreds of completed challenges.

Why this approach?

• Gamification
• Users get to gather points - could be showed in a profile and/or a leaderboard of some kind?
• By decreasing the reward on each submitted try, users learn that they need to be thoughtful about their guesses. This makes them think deeper about each attempt - which hopefully helps them learn more from each puzzle. Extreme example: Users don't learn much by mindlessly bruteforcing combinations and still getting points.

[Limpan]

I agree on the gamification of things. It seems to be a rather effective approach. It is a future feature but it should of course be somewhat considered for the database design.