Flashcard/Spaced Repetition System/Free

• Free Spaced Repetition System

  • SRS is the category.

    FSRS is a specific scheduling algorithm inside that category.
    
    

    ---

  • SRS (Spaced Repetition System)

    
    A broad concept:
    

  • Any system that spaces reviews over time
  • Can be implemented many ways (Leitner, SM-2, custom heuristics)
  • Defines what you’re doing:
    • adaptive review timing
    • based on memory strength

      
      It’s not one algorithm — it’s a family of approaches.
      
      

      ---

  • FSRS (Free Spaced Repetition Scheduler)

    
    A modern algorithm for SRS scheduling.
    

  • Developed as an alternative to older models like SM-2
  • Used in tools like Anki (optionally) and projects like repeater
  • Defines how scheduling is computed

    ---

  • Core Difference

    	SRS	FSRS
    Type	Concept / paradigm	Specific algorithm
    Scope	Many possible implementations	One concrete implementation
    Goal	Efficient memory retention	Optimize retention with predictive modeling
    Flexibility	High (many strategies)	Fixed mathematical model (tunable)

    ---

  • What Makes FSRS Different

  • 1. Explicit Memory Model

    
    FSRS models memory with variables like:
    

  • Stability (how long memory lasts)
  • Difficulty (intrinsic hardness of item)
  • Retrievability (probability of recall at a given time)

    
    Older systems (like SM-2) use simpler heuristics (ease factor + intervals).
    
    

    ---

  • 2. Target Recall Rate

    
    FSRS lets you aim for a specific retention probability:
    

  • e.g. 90% recall (as mentioned in repeater)
  • schedules reviews just before that probability drops below target

    
    This is much more explicit than classic SRS systems.
    
    

    ---

  • 3. Data-Driven Optimization

    
    FSRS can be trained on your review history:
    

  • Fits parameters to your memory behavior
  • Adapts to:
    • how quickly you forget
    • how you rate difficulty

      
      Traditional SRS algorithms are mostly static.
      
      

      ---

  • 4. Continuous vs Discrete Thinking

  • SM-2 / Leitner → step-based (“increase interval by factor”)
  • FSRS → probabilistic / continuous model of forgetting

    
    It’s closer to a prediction problem than a rule-based system.
    
    

    ---

  • Intuition

  • SRS = “review things less often as you learn them”
  • FSRS = “predict exactly when you’ll forget this, and schedule just before that”

    ---

  • Practical Implication

    
    With FSRS:
    

  • Fewer unnecessary reviews
  • Better long-term efficiency
  • More consistent retention at a chosen level (e.g. 90%)

    
    With generic SRS (older algorithms):
    

  • Simpler
  • More predictable
  • Slightly less optimal scheduling

    ---

  • Mental Compression

    
    SRS is the philosophy.
    FSRS is a modern, data-driven implementation of that philosophy.