Reference and Selective Reading

You do not need to read SICP front-to-back for this module. Use the concept and practice pages first. Open these local chunks only when you need alternate exposition, worked examples, or depth on a specific step.

Source Roles

Source	Role	Why it is here
SICP (Abelson, Sussman, Sussman)	Primary teaching source	The spine of every cluster. All primary concepts point to specific SICP sections
The C Programming Language (K&R)	Contrast	Shows what "no closures, no first-class procedures" looks like and why abstractions in C are awkward
Computer Organization and Design (Patterson & Hennessy)	Contrast	Where interpretation bottoms out in real machines (relevant to Concept 12)
Norvig -- lispy (external)	Primary external	The minimal-viable interpreter in Python, directly supports Concepts 10 and 11
Nystrom -- Crafting Interpreters (external)	Primary external	A full treatment of tree-walking and bytecode VMs
MIT Press SICP online (external)	Authoritative external	Canonical source for any SICP passage

Read Only If Stuck

Cluster 1: Abstraction as the Engineering Move

• SICP 1.1 The elements of programming
• SICP 1.1.2 Naming and the environment
• SICP 1.1.4 Compound procedures
• SICP 1.1.7 Square roots by Newton's method
• SICP 1.1.8 Procedures as black-box abstractions
• SICP 2.1 Introduction to data abstraction
• SICP 2.1.2 Abstraction barriers
• SICP 2.1.2 Abstraction barriers (part 2)
• SICP 2.2 Hierarchical data and the closure property
• SICP 2.4 Multiple representations for abstract data
• SICP 2.4.1 Representations for complex numbers
• SICP 5.1.2 Abstraction in machine design
• K&R: Basics of functions
• K&R: Scope rules
• K&R: Basics of structures
• K&R: typedef

Cluster 2: Higher-Order Procedures and Closures

• SICP 1.3 Formulating abstractions with higher-order procedures
• SICP 1.3.1 Procedures as arguments
• SICP 1.3.2 Constructing procedures using lambda
• SICP 1.3.3 Procedures as general methods
• SICP 1.3.4 Procedures as returned values
• SICP 1.3.4 Procedures as returned values (part 2)
• SICP 1.3.4 Procedures as returned values (part 3)
• SICP 2.2.1 Representing sequences
• SICP 2.2.3 Sequences as conventional interfaces
• SICP 2.2.3 Sequences as conventional interfaces (part 2)
• SICP 3.1.1 Local state variables
• SICP 3.2.3 Frames as the repository of local state
• K&R: Pointers to functions

Cluster 3: Evaluation Models

• SICP 1.1.5 The substitution model
• SICP 1.2 Procedures and the processes they generate
• SICP 1.2.2 Tree recursion
• SICP 3.2 The environment model of evaluation
• SICP 3.2.1 The rules for evaluation
• SICP 3.2.2 Applying simple procedures
• SICP 3.2.4 Internal definitions
• SICP 4.2 Variations on a Scheme -- lazy evaluation
• SICP 4.2.2 An interpreter with lazy evaluation
• SICP 5.1.4 Using a stack to implement recursion
• SICP 5.4.2 Sequence evaluation and tail recursion

Cluster 4: Metacircular and Interpreter Design

• SICP 4.1 The metacircular evaluator
• SICP 4.1.1 The core of the evaluator
• SICP 4.1.2 Representing expressions
• SICP 4.1.2 Representing expressions (part 2)
• SICP 4.1.3 Evaluator data structures
• SICP 4.1.4 Running the evaluator as a program
• SICP 4.1.5 Data as programs
• SICP 4.1.6 Internal definitions
• SICP 4.1.7 Separating syntactic analysis from execution
• SICP 5.4 The explicit-control evaluator
• SICP 5.4.1 Core of the explicit-control evaluator
• SICP 5.5 Compilation
• SICP 5.5.1 Structure of the compiler
• SICP 5.5.2 Compiling expressions
• SICP 5.5.6 Lexical addressing

Cluster 5: State, Mutation, and Language Design

• SICP 3.1 Assignment and local state
• SICP 3.1.1 Local state variables
• SICP 3.1.2 The benefits of introducing assignment
• SICP 3.1.3 The costs of introducing assignment
• SICP 3.1.3 The costs of introducing assignment (part 2)
• SICP 3.3 Modeling with mutable data
• SICP 3.4 Concurrency: time is of the essence
• SICP 3.4.1 The nature of time in concurrent systems
• SICP 3.4.2 Mechanisms for controlling concurrency
• SICP 3.4.2 Mechanisms for controlling concurrency (part 2)
• SICP 3.5 Streams
• SICP 3.5.1 Streams are delayed lists
• SICP 3.5.2 Infinite streams
• SICP 3.5.3 Exploiting the stream paradigm
• SICP 3.5.4 Streams and delayed evaluation
• SICP 3.5.5 Modularity of functional programs

Optional Deep Dive

• SICP 2.3 Symbolic data -- the first appearance of "programs as data"
• SICP 2.3.2 Symbolic differentiation -- a tiny language evaluator in disguise
• SICP 2.4.3 Data-directed programming and additivity -- dispatch-table version of operator overloading
• SICP 2.5 Systems with generic operations -- polymorphism via dispatch, no classes needed
• SICP 3.3.1 Mutable list structure -- why mutation changes list semantics
• SICP 4.3 Variations on a Scheme -- nondeterministic -- the amb operator, logic programming's precursor
• SICP 4.4 Logic programming -- the full query-language evaluator
• SICP 5.1 Designing register machines -- the bridge from interpreters to hardware (Concept 12)

Concept-to-Source Map

Primary concept	Best source if stuck	Why this source
01 Procedural abstraction	SICP 1.1.8 Procedures as black-box abstractions	Explicit statement of the black-box idea
02 Data abstraction + ADTs	SICP 2.1 Introduction to data abstraction	Rational numbers, fully worked
03 Layering abstractions	SICP 2.4.1 Representations for complex numbers	Two representations behind one barrier
04 First-class procedures / HOFs	SICP 1.3.1 Procedures as arguments	sum, integral, Simpson
05 Closures + lexical scope	SICP 1.3.4 Procedures as returned values	average-damp, deriv, newton
07 Substitution vs environment model	SICP 3.2.1 The rules for evaluation	The precise rule of extension
09 Tail calls	SICP 1.2 Procedures and processes	Process vs procedure distinction
10 eval + apply	SICP 4.1.1 The core of the evaluator	The template for the interpreter
11 Writing a small Lisp interpreter	SICP 4.1.4 Running the evaluator as a program	The wiring into a usable REPL
13 Assignment: cost of mutation	SICP 3.1.3 Costs of introducing assignment	Sharpest statement of what is lost